Talent Assessment for Growth Startups

TALENT ASSESSMENT

_ Talent Assessment is published by Notion, 91 Wimpole Street, London W1G 0EF. Registered address: Third Floor, 1 New Fetter Lane, London EC4A 1AN. For opportunities to contribute to future editions of Notion Insights and its associated publications please contact [email protected] Design, Layout & Production SunnySideUp, contact [email protected] Notion (OC364955) is Authorised and Regulated by the Financial Conduct Authority. _ Reproduction in whole or in part without written permission is strictly prohibited. © 2019 Notion. All rights reserved.

_ 1. Introduction 2. Before you start 3. Finding candidates 4. Interviewing candidates 5. From candidate to employee 6. Retaining your employees i. Contributors ii. Appendix

NO TION T 1. INTRO- ALENT A S SES DUCTION SMENT 4

When you’re building a SaaS business, your primary asset outside of your product idea is your people. This project aims to gather data and information from recruitment experts, founders NO TION T and investors across the Notion family. It is meant as a guide rather than a set path to follow. ALENT A S The only sure thing that has come out of SES SMENT aggregating information, collecting data and interviewing experts is that there is no one way to grow and scale your team and people. However, by sharing knowledge and ideas from our ecosystem we hope to reinforce proven solutions, uncover challenges and provide new ideas for solutions in how to build a successful enterprise software team. 5

One of the most common challenges when it comes to people and talent is articulating a return on investment (ROI). NO TION T ALENT A S SES SMENT _ The output of talent is not directly tied to business outcomes and has few measurable outputs. However, even small improvements in sourcing, hiring, onboarding, and managing talent can result in significant ROI for the business. We are increasingly focussing on a new concept being used to help assess the business impact of people operations: Employee Lifetime Value (ELTV). Here at Notion we are on a mission to help nurture a thriving ecosystem that builds, connects and recycles knowledge, ideas and skills for our portfolio and the wider community and we would therefore like to thank any contributors for sharing their experiences for others to learn from. As ever this is as much a learning exercise as it is a guide, so please feel free to contribute, comment or challenge any of our assertions and assumptions. 6

NO TION T 2. BEFORE ALENT A S SES YOU START SMENT 7

HOW DO YOU DECIDE ON YOUR CULTURE & ASSOCIATED VALUES? NO TION T ALENT A S SES SMENT 8

Culture and values represent the underlying fabric of any organisation. Despite this foundational role, culture and values are rarely prioritised by founders early on because they’re difficult to define, can be seen as transitory and the effect on the bottom line is indirect - especially at the seed stage. However, many founders emphasise the NO TION T importance of defining culture and values early on, before scaling the team and before culture is ALENT A S created on its own. SES SMENT One thing is certain: a culture will always form. The challenge for founders is being able to take on, drive and define that process and the eventual outcome. Research from Harvard Business School has shown that by actively managing culture, organisations are more likely to deliver on their strategic objectives in the long run. 9

WHAT IS CULTURE? _ Through decades of academic studies and research, company culture has been found to have three building blocks: ARTEFACTS NONO What you see (symbols, texts, ceremonies, rituals, facilities, processes, communication) TION TTION T VALUES What people say (norms for good/bad practices, slogans, mission statements) ALENT AALENT A ASSUMPTIONS The meaning behind it (why do we have the culture we have, how does it help us) SS SESSES In less theoretical language culture is the system of agreed and shared values and norms that define what is important and what attitudes and behaviours are appropriate. SMENTSMENT WHEN SHOULD FOUNDERS DO SOMETHING ABOUT DEFINING CULTURE AND VALUES? “Despite all the reasons (to not focus on culture) - not only should you think about culture, you should obsess over it as early in the process as possible.” – Dharmesh Shah, co-founder and CTO of HubSpot Why? Because your team is the biggest thing that will cause a startup to succeed or fail and culture is going to define your team early on. This is going to influence your overall probability of success more than most of the things you will do as a founder. However, this does not mean that you should not iterate. Culture is more like software than hardware. It requires running updates and continual work. As an example: HubSpot’s culture code deck has been updated more than 25 times. 1010

WHY IS IT IMPORTANT TO DEFINE CULTURE AND VALUES? NONO TION TTION T STRONG CULTURES ENHANCE ORGANISATIONAL PERFORMANCE IN TWO WAYS: ALENT AALENT A 1. They improve performance by energising employees—appealing to their higher SS ideals and values and rallying them around a set of meaningful, unified goals. SESSES 2. A clearly defined culture reduces the need for management hours to be spent; employees know what’s expected of them so they need to be managed less to be SMENTSMENT aligned. In Leading By Leveraging Culture (2003), Chatham and Cha from Harvard and Berkeley formulated a culture framework that shows the organisational results that come from organisational culture. Intensity No Agreement Agreement Low Weak Empty High Warring Fractions Strong Agreement Do people agree on the culture? Intensity Do people feel strongly about the culture? Founders want their company to be in the bottom right corner. 1111

AGREEMENT NONO TION TTION T ALENT AALENT A SS JUST DO IT BE ASPIRATIONAL SESSES Agreement does not have to be organic, Though you should be honest, make particularly when it is an ephemeral thing sure you are also aspirational and that SMENTSMENT like culture. Agreement should come from the culture and values can scale with the the top. Earlier is better than later, as company. HIroki Takeuchi, Co-founder and CEO of GoCardless, has said “At GoCardless we first DEFINE THE ARTEFACTS worked on the culture and values when This is critical in order to ensure that we were around 30 people. I’d encourage the culture and values mean the same to founders to tackle it, ideally, before this…. every employee and that the company Really try to get it done as soon as you can follows up on the espoused culture. because it dictates who you hire and how Artefacts can be physical: The Palo Alto you onboard them.” office of IDEO famously has an airplane wing jutting out from one wall, signalling BE HONEST IDEO’s culture of playful experimentation Though the founder(s) decides on the and free expression. They can also be culture and values at this early stage, make procedural: Zappos offered a $4,000 sure they actually reflect the reality of the quitting bonus during new hire training company. You cannot have a culture of for incoming employees to leave the “always shipping great products” and “move company fast and break things”, or “Output, not input” and then demand everyone works This sends a clear value statement. 9-6pm, for example. 1212

INTENSITY NONO TION TTION T ALENT AALENT A DON’T SEEK CONSENSUS / don’t like about their job and the team. The f ounders are the focal point, the Classify the feedback into groups. SS nexus and the future of the company. SESSES Sebastian Siemiatkowski, CEO and LEAD BY EXAMPLE co-founder of Klarna has noted how “I The culture flows from the founder. As SMENTSMENT wrote Klarna’s internal culture documents the founder, you have to reflect your own for three years – but the only thing that values and standards and be the strongest came out of company wide discussions on it representation of the culture. This is also was how words like ‘agile’ should be applied why it makes sense for the founder(s) to to the company.” define the culture. GET FEEDBACK ENSURE THEY AFFECT DAY-TO- This does not mean consensus. Ensure DAY people understand that you want input, Ensure that you have unambiguous but that how that input is turned into artifacts and values that formulate your output is up to you. Ensure that people culture. Furthermore, work your values are included in the process both so you get into the company OKRs and KPIs. Base to the actual culture of the company and your culture on an easily replicable, scalable, make people feel included. Interview your people-focused tradition so you don’t have to employees and ask them what they like change it every funding round. AS A FOUNDER YOU ARE THE DRIVING FORCE OF THE COMPANY, ITS VISION AND ITS CULTURE. DEFINING CULTURE AND VALUES EARLY WILL HELP YOU GET TO THAT VISION. 1313

HOW DO YOU COMMUNI- CATE CUL- TURE AND VALUES TO THE ORGANI- NO SATION? TION T ALENT A S SES SMENT 14

NO TION T ALENT A S SES SMENT Defining your values is a pivotal moment for company culture. In order to ensure behavioural change there has to be a concerted effort from founders and senior leadership to communicate and implement these values. 15

DEFINING VALUES ‘Customers will never love a company until the employees love it first.’ – Simon Sinek, author, Start with Why NONO TION TTION T ALENT AALENT A SS _ SESSES SMENTSMENT The co-founding team at Talent.io decided to present their values “during a long company weekend with all team members outside of regular work hours so we could take the time to go through them in depth and why we cared about it.” Not everyone will have time for a weekend away, but it is worth investing considerable time when implementing something which senior members and founders have spent time and resources developing. Remember that communicating culture and values is a program, not a project. It does not have an end date. LIVE AND LEAD BY EXAMPLE Leaders are always being watched. Setting core values, and then failing to abide by them, is worse than not establishing core values at all. PERSONALLY TEACH THE VALUES THROUGH ORIENTATION/TRAINING It’s unrealistic to simply send out an email or a document listing core values, or engrave them on coasters, and think that everyone will adopt them. Formal training communicates that the values matter and shows company commitment. REINFORCE THE VALUES IN ALL COMMUNICATION Revamp the all-hands meetings formats and launch internal newsletters. Every employee touchpoint from interview to leaving day should reinforce the values. 1616

RECOGNISE AND REWARD and ‘company value’ or speed and ‘not VALUES-CENTRIC BEHAVIOURS company value’. There are many ways to recognise and reward values-centric behaviours, INCORPORATE THE VALUES INTO including spot-bonuses, peer-voting NONO YOUR SALES PROCESS opportunities that give employees the Integrate the core values into the TION TTION T opportunity to nominate co-workers for proposal messaging. This goes beyond successfully living the values, and written simply listing the values in an opening recognition in newsletters or on the paragraph. The language highlights how website. Social recognition is powerful and the core values shape the customer ALENT AALENT A public storytelling about employees who experience. embody a corporate value has a profound SS effect on the rest of the organization’s INCORPORATE THE VALUES INTO SESSES understanding of those values. YOUR PERFORMANCE REVIEW PROCESS SMENTSMENT INCORPORATE THE VALUES INTO Once you’ve hired employees that YOUR HIRING PROCESS align with your core values, and you’ve Selecting candidates that culturally trained them on how they can live the align with - and add to - your organization values, you are ready to integrate the core is just as important as finding candidates values into the performance review that match your required experience level process. The performance review process is and skill set. where you inspect what you expect. ALIGN VALUES WITH YOUR OKRS DON’T KEEP PEOPLE WHO OR KPIs VIOLATE THE CORE VALUES Values indicate a way that employees Firing employees is always one of the behave and OKRs/KPIs are a benchmark worst aspects of being a founder. However, for the way those behaviours are measured when an employee consistently engages in in terms of work output. With this in mind, behaviour that contradicts the desired and they need to align. required behaviours of an organization, For example, if you have a value that is this impacts their personal performance as ‘We only ship great products’ but a KPI that well as the performance of the company. focuses on shipping products quickly, Every employee represents the brand, both there’s a likelihood that employees will be internally and externally. confused about which path to take: quality 1717

HOW DO YOU KNOW WHO YOU NEED, AND WHEN YOU NEED THEM? NO TION T ALENT A S SES SMENT 18

NO TION T We have crunched the numbers for 130 European B2B software startups ranging from Seed rounds of ALENT A S $60k to Series D rounds of $75m. The companies SES SMENT come from a wide range of geographies, with HQs in 32 different cities across 18 countries. Using Notion portfolio data, LinkedIn and Crunchbase we have layered this data with departmental headcounts broken down into three categories: Sales & Marketing, Tech & R&D and Other. 19

NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT _ There are no right answers when it comes to the makeup of your team; the nuances of how your business is set to scale, and the expertise on the existing team will influence who you need to hire and when, in such a way that many of the lessons of hiring don’t translate from one business to another as an exact science. Our purpose is to present founders with data that can help inform their hiring decisions and guide their thinking on what their company is going to look like (and all that this entails) as it scales. We present the relative sizes of the teams as we think the mix and the balance is the more interesting metric for founders rather than absolute numbers. 2020

WHAT DOES THE DATA TELL US? Sales & Tech Marketing / R&D Other Mean Seed 29% 55% 17% Series A 39% 44% 17% Series B 41% 41% 18% NONO TION TTION T Series C 36% 42% 22% Series D 35% 34% 31% ALENT AALENT A Median SS SESSES Seed 29% 54% 15% SMENTSMENT Series A 38% 46% 14% Series B 46% 38% 12% Series C 33% 40% 18% Series D 36% 34% 33% Standard Deviation Seed 14% 16% 10% Series A 18% 18% 10% Series B 17% 16% 14% Series C 10% 11% 11% Series D 10% 7% 9% 2121

NO TION T At Seed stage there is no ‘trend’ for how employees ALENT A S are split between functions. Some companies are SES SMENT 80% engineers, others are 5% engineers. At later stages, however, this isn’t true to the same extent: companies converge at around 1/3 Sales and Marketing, 1/3 Tech and R&D and 1/3 ‘Other’. This can be seen in the Standard Deviation table above, which shows a high variance at seed and a much lower variance at Series D. 22

_ SEED By virtue of the nature of software businesses, the initial stages of operation are typically tech heavy in terms of staffing. At Seed stage, approximately half of the headcount is technical, reflecting that most software founders are technical. The remaining employees are split between Sales & Marketing and other. Tech & R&D 55.1% NONO TION TTION T Other 15.7% Sales & Marketing 29.2% ALENT AALENT A SS SERIES A SESSES As you raise subsequent rounds the main focus shifts from building the product SMENTSMENT and achieving product market fit to scaling ARR. The organisation will go through substantial change during this time. Typically at Series A, software businesses will seek to increase the number of salespeople more rapidly than other functions. As the business approaches product-market-fit, the opportunity for generating revenue starts to take shape and the requirement for people to capitalise on this increases. Operations, finance and support staff numbers typically stay static from Seed to Series A, whereas technical hires increase proportionally. Tech & R&D 46.5% Other 14.2% Sales & Marketing 39.3% 2323

NONO TION TTION T ALENT AALENT A _ SS SESSES SERIES B SMENTSMENT This is the point at which the proportion of technical talent in the business starts to level out. We call this ‘The Reverse Wedge’ at Notion, by which we mean that space in the average ‘pie chart’ of functional split that was once occupied by tech, will start to be taken over by Sales and Marketing. With additional budget - and being firmly out of the bootstrapping phase - HR specialists are brought in to grow the team, a finance team is put in place to support the spending of the latest round raise, and operations skill sets ramp up to enable the organisation to shape itself up for scaling. Tech & R&D 39.5% Other 12.7% Sales & Marketing 47.9% 2424

_ SERIES C/D At this stage the average average software business has reached a steady state of functional split. Technical, Sales and Marketing, and other skill sets occupy roughly equal splits of the talent pie chart. The same trends as described above are also present when we look at round sizes. This is helpful because naming rounds is not always an accurate science and varies from NONO country to country. TION TTION T Tec & R&D 42.3% ALENT AALENT A Other 21.8% SS Sales & Marketing 35.8% SESSES SMENTSMENT SERIES D Tech & R&D 34.3% Other 30.8% Sales & Marketing 34.9% 2525

NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT Round Size Sales & Tech ($m) marketing / R&D Other 0 - 1 30% 55% 15% 1 - 5 30% 53% 17% 5 - 10 42% 41% 17% 10 - 20 48% 39% 13% 20 - 50 39% 42% 19% 50+ 36% 40% 24% The final way to test the viability of these departmental trends is to create a matrix that delineates departmental ratios by the round size ($m) and stage. 2626

Stage / Round Size ($m) 0 - 1 1 - 5 5 - 10 10 - 20 20 - 50 50+ Sales and marketing as % of organisation by stage and round size Seed 29% 29% 13% - - - Series A - 41% 42% 32% 12% - Series B - 17% 37% 48% 44% - NONO TION TTION T Series C - - 30% 39% 37% - Series D - - - 24% 38% 36% ALENT AALENT A Tech, R&D as % of organisation by stage and round size SS SESSES Seed 56% 55% 48% - - - SMENTSMENT Series A - 41% 40% 56% 75% - Series B - 55% 37% 40% 42% - Series C - - 43% 41% 43% - Series D - - - 37% 30% 40% “Other” as % of organisation by stage and round size Seed 15% 17% 39% - 12% - Series A - 18% 18% 12% 12% - Series B - 29% 25% 12% 20% - Series C - - 27% 20% 39% - Series D - - - 39% 32% 24% 2727

NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT The same trends are visible, though naturally with more variance - particularly around the larger and smaller amounts across the different stages. Smaller companies are still mostly technical and Series A tends to be the round at which Sales & Marketing numbers grow exponentially. The ‘reverse wedge’ is still visible in Series B, while departmental ratios slowly even out across Series C and D. IT’S ABOUT KNOWN UNKNOWNS The key take-away for founders is to be prepared for how their company is going to change as they scale. This change happens due to a combination of investor expectations, product and market demands and changed focus. Knowing that these changes are around the corner will allow founders to plan ahead and put in place plans and processes that will allow their company to make this journey. The rest of this book will outline how you find, assess and keep the people and talent that is so crucial to the scaling we see in the data. 2828

3. NO FINDING TION T ALENT A S CANDI- SES DATES SMENT 29

GENERATING INTERNAL & EXTERNAL JOB DESCRIP- TIONS NO TION T ALENT A S You know what company you want to build and SES SMENT you know what you need. How do you make it clear to everyone else what you want? The first step is the job spec. A job spec galvanises the fleeting thoughts and images founders have of their requirements into something specific, into a person. 30

BASIC BEST PRACTICES • Make the job title searchable. Job boards work like search engines so ensure you are clued into what language is being used for this type of role. • Sell your company and the candidate’s future in an engaging fashion. • Use words that evoke feelings while being specific. • Make them aspire and then act on that desire. • Use 'you' or 'we', don’t use the passive voice. NONO • Make it easy to apply! Submitting an application should take two clicks: one form TION TTION T and a CV upload. STRUCTURE 1. Company intro: Who are these people? Add a photo. Emphasise the growth and ALENT AALENT A exciting journey ahead. 2. Role intro: Why are they hiring for this position? This is where they should get SS excited. SESSES 3. Responsibilities. What would I do day-to-day? Who would I report to? Emphasise the importance of the role. SMENTSMENT 4. Profile: Are they looking for someone like me? Attributes and values. 5. Skills/Experience: Would I meet their requirements? Must-haves and nice-to-haves. 6. Extra role info: What are the benefits? Any perks? PUT TO WORK EMOTIONAL TRIGGERS THAT RELATE TO MASLOW’S PYRAMID OF NEEDS IN Y OUR JOB DESCRIPTION • The need to belong: Talk about the actual team they will be joining. Include a group photo or pointing out something unique about your team. Make sure to use emotional and inclusive language rather than bullet points. Leverage social proof using social media that showcases your team spirit. • The desire for esteem: Outline how critical the candidate’s success will be to the organisation. Do not only say that they will report directly to X, but expound on why. • The desire to grow: Everyone wants to be a better person. Include the opportunities for learning in the description and tailor it to the position (not just generic). Include examples of what people in similar roles have learned in the past. If you need more, specific inspiration you can find 100s of template job specs on Workable’s resource page. Keep in mind that these are generic and will need customisation and added emotion and character. 3131

SOME EXAMPLES OF JOB DESCRIPTIONS NONO TION TTION T ALENT AALENT A Accounts Supportocat SS SESSES Github is looking for a full-time Accounts Supportocats in the GitHub Austraia/Asia-Pacific timezones, as well as Europe / Middle- East / SMENTSMENT Africa timezones. Basically, we’re good on US business hours. Simple When GitHub users have questions, they email support. We help people as quickly and awesomely as possible. Accounds supporto- explanation cats handle passwords, email verification, billing, legal, spam and of duties abuse, account lockuts, fraud, and helping people choose plans. How to apply This requires a logical brain and a spidey sense for problems. Email resumes@github with the subject The most important characteristic of our support team is that we “Accounts Supportocat”. We want our interactions with users to be memorable, Spotlight on <3 helping developers. GitHub is a place for people to work better, so please make your email memorable core values together. Our job is to help. Support treats users the way we’d want as well. Tell us about how you match up someone to treat our friends. to the Supportocat characteristics, and what makes you a wonderful person Supportocats are excellent writers: stellar grammar, charming writ- to have around. Please include your Definition of ten personality, and the ability to explain complicated things simply. GitHub username, (it’s okay if you haven’t signed up before.) success & top We measure our success in swiftness, accuracy, clarity, and the performers number of exclamation points we receive in replies. The occasional use of an animated gif may be required. Original You’re good at: headings • logic and solving puzzles • advocating and empathizing • the English language]working remotely Extra awesome • customer and experience Few bullets • technical experience (QA, documentation, elaborate board- games) • you’re read “How to win friend’s and Influence People, despite its odd name White space Developers: this probably isn’t the job for you, but perhaps you Referral have a friend who would be a perfect fit. Send them the link. Request Thanks! 3232

NONO TION TTION T ALENT AALENT A Associate Community Specialists SS SESSES Meetup’s Associate Community Specialists support millions of Meetup members, tens of thou- Transparency - it takes a sands of Meetup Organizers, and more than 100,000 Meetup Groups on just about every topic SMENTSMENT particular type of person under the sun. To keep the Meetup ecosystem humming, Associate Community Specialists to want to respond to responds to mountains of email and take plenty of phone calls; we answer questions, offer mountains of email advice, hunt down bugs, moderate discussions, enforce Meetup’s terms of service, and serve as advocates for the Meetup community at-large. As an Associate Community Specialist, a typical day might include the following: • Congratulate the Organizer of a French language Meetup on reaching 1,000 members. Description of a • Call the Organizer of a Motorcycle Meetup Groupl, and walk her through her email settings. • Offer a new Meetup Organizer tips for recruiting members. day in the life • Send a member directions for uploading photos. • Iron out a billing issue and reset some lost passwords. • Run a quality check on recently-created Meetup Groups. Addressing the candidate This job might be for you if: directly, i.e., ‘You’ You enjoy solving problems. You love taking on difficult challenges and finding creative solu- tions. You don’t get flustered easily. If you don’t know the answer, you’ll dig until you find it. Jargon-free attributes You pay attention to the details. As far as you’re concerned, anything worth doing is worth of top performers doing right, every single time. You stay focused, and nothing falls through the cracks on your watch. Very short sentences You communicate clearly. You write well. You speak eloquently. You can explain just about anything to anyone, and you’re comfortable communicating in writing and on the phone. - - - - To land this gig, you need to have some experience helping other people solve problems. It Conversational and doesn’t need to be traditional customers service experience; we have a former park ranger personal tone and ex-Maitre D’ in our ranks. Internships, volunteer work, and side projects count. Just make sure we can see where you picked up your helping-people powers somewhere in your cover letter or resume. 3333

NONO everything me TION TTION T QA Engineer Location: Tel-Aviv, Israel ALENT AALENT A Compelling introduction Are you a stubborn person who’s not affraid of searching for a needle in a stack of hay? Do that grabs attention you over the boarder between development and product? Are you an ardent learner who’s always looking to improve? Can you play nice with both humans and machines? SS SESSES Eye- catching & tongue-in-cheek Waldo spotting score Outburst Rate Sen Level SMENTSMENT description of key qualities Original headings We Have: • A top notch application spanning multiple mobile platforms. • HTML5 web apps in abundance. • A team of aces to work with. • An open mind for new ideas and methodologies. Lots of white space You Have: • The knowledge to test for consistant UI, content and user experience. • A precise attention to detail. Few bullets • The ability to collect data, define problems, establish facts and draw valid conclusions. • The understanding of team strategies and objectives. • The motivation to learn and constantly improve processes and tools. • The people skills required to work with designers, developers and product managers. You might also have: Expereince in creating and implementing test automations. Working knowledge with UI testing frameworks Mobile device testing experience HTML and front end development knowledge ‘Apply with LinkedIn’ so Apply with LinkedIn or Email us at: [email protected] process is streamlined 3434

GENERATING VOLUME AT THE TOP OF THE FUNNEL NONO TION TTION T ALENT AALENT A Jobvite’s database of more than 55 million job seekers and 17 million applications SS informs their annual recruiting funnel benchmark report. This data comes from all SESSES industries and the numbers are likely to be more positive for more exciting and modern startup companies. However, the data hammers home the importance of improving quality SMENTSMENT and quantity in tandem on the top of the funnel. Overall data - conversion rates 2015 2016 2017 Visitors to applicants 11% 9% 12% Applications per 59 52 36 open requisition Applications to interviews 12% 15% 12% Interviews to offers 17% 20% 28% Offers to hires 89% 83% 91% Top-to-bottom 0.2% 0.28% 0.37% Average time-to-hire (days) 41 39 38 3535

NONO TION TTION T Revenue Applicants Applicants Interviews Offers to Start to ($m) per job Per Interview to Offers hires Finish ALENT AALENT A SS <1 54 15.7% 16.4% 81.5% 2.10% SESSES 1 - 5 62 16.1% 17.8% 82.1% 2.35% SMENTSMENT 5 - 10 53 18.6% 16.6% 52.3% 2.30% 10 - 25 54 14.8% 19.0% 83.7% 2.35% 25 - 50 50 13.9% 20.4% 83.4% 2.36% 50 - 100 48 15.8% 21.5% 83.2% 2.83% 100 - 250 41 14.0% 26.0% 57.3% 3.18% For every successful hire your company will need 50-100 candidate CVs. Startups often have little to no name recognition, so how do you ensure quality and volume at the top of the recruitment funnel? Matt Buckland, VP of Customer Advocacy at Workable has written extensively on talent acquisition and outlines four main arenas of talent acquisition on his blog: http:// thekingsshilling.io/ Data: Jobvite’s 2018 Recruiting Benchmark Report 3636

EVENTS NONO TION TTION T There are 1000s of events in the startup world. Staying tuned in to Meetup, Eventbrite, Tech City News, Google Campus London and Silicon Milkroundabout will give you a head start. There are plenty more directories and events focused on talent and hiring. Here are ALENT AALENT A a couple of things to keep in mind when considering an event: SS SESSES What? Why? SMENTSMENT Audience Is this a target group you want to recruit from? How many people will be there? When you look at a large number of attendees, bank on 50% of that number actually attending. ROI How much do sponsorship opportunities cost, and what’s included in the package? Get creative: sponsoring a lunch at an event may have more impact than just adding your logo to glossy flyers. Remember to factor in opportunity cost of spending time preparing for and being at the event. Speakers Are the speakers the kind of people you want to recruit and is the content relevant to what you’re doing? Better yet, can sponsorship guarantee a speaking slot? Past Is this an event that has been successful for recruiting in the past? Do Experience other companies rate it? Can you get an honest opinion from someone? Resources Do you have sufficient employee resources/time to have a meaningful presence at the event? 3737

REFERRALS PROS • Faster hires (29 days on average compared to 39 for job adverts) • Higher retention rate (46% vs 33% according to HR Technologist) • Reach candidates not on the market NONO TION TTION T CONS • Referral schemes might lead you to hire candidates similar to the employees you already have. Be aware of this and incentivise diverse referrals. ALENT AALENT A PRIZE Companies offer cash or prizes for a successful placement. However, you might decide SS on something that makes sense in your company – would your culture respond better to cash SESSES or a bottle of wine or an Amazon voucher? A trophy for the referrers desk or a lego figure representing the person referred? It’s important that the reward demonstrates gratitude and SMENTSMENT that it is meaningful. Only spending £50 on showing that gratitude is a sure fire way to dry up any referrals. Hiring is hard, and a referral can be a valuable short cut. Decide early on whether you want to reward referees based on passed probation, offer made or simply passing the first interview. This will affect how substantial the prize can be. Waiting for someone to pass their 3 or 6 month probation period might lead employees to lose interest. If founders trust their interview process it should be possible to reward on offer. SUBMISSIONS Make it as easy as possible for staff to submit referrals. Let your team know they should email referrals to a specific email address where your HR or recruitment manager can keep a record of high quality candidates if there isn’t an explicit job open. When you do receive a referral, take care to make sure the process is both quick and effective – and take the extra time to offer verbal feedback. The process could be abridged by for example removing the initial phone screening. OTHER THINGS TO KEEP IN MIND • Change the reward to renew employee’s interest • Identify and reward your ‘super recruiters’ • Publicly praise people who refer candidates 3838

JOB BOARDS AND ADVERTS NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT It’s not always true that the best people aren’t looking. This bias has led to firms sometimes missing out on great candidates who have actively tried to join their companies. Job adverts are a great way to advertise your company across multiple audiences and media. Try to A/B test different job titles, descriptions of the role and the call to action. Only change 1 thing against your control advert. Remember that the the metric you are assessing is not just number of clicks, but number of qualified applicants that come through. You should not decide the winner of the test on clicks (because it’s simply too high up the funnel) and not on # of candidates accepting offers (because it’s too far down and not enough data points). You can find a list of places to advertise in the appendices. Use a distilled, though still emotive version of the job description you have already created. Georgiana Barbanta, Talent Acquisition Manager at Paddle, who doubled their team between November 2017 and July 2018, uses a specific strategy here: “We bought one LinkedIn recruiter account (£6,000 a year) and we used that mostly for reaching out to candidates. We also bought two annual job ad spots for £1,000 each. We used these like parking lots, rotating job ads depending on our needs.” 3939

RECRUITMENT AGENCIES NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT Using an agency can be a great time saver. However, remember that they will most likely send good candidates to multiple companies and you can end up with a very top heavy funnel from recruiters. Take extra care to assess the key metrics and cut recruiters who don’t stack up. • Negotiate terms you’re comfortable with. For early stage startups, hiring fees are generally better than retainers since you’re only paying for results. Fees range from 15-20%. • Make sure you carry out reference checks or ask people you trust about which recruiters they recommend. Recruiters are good salespeople and it can be confusing to pick someone just based on a website or an ad. • You must give the agency time, direction and feedback. The recruiter will fail if you do not communicate with them about what you want and why, so help them calibrate and understand your team. • Give them your story. A recruitment agency will never sell the company as well as an employee can, so it’s up to you to give them your story. • Set a realistic deadline. Give them at least two months to find somebody. If things don’t work out, it’s okay to swiftly end the relationship. 4040

4. NO INTER- TION T VIEWING ALENT A S SES CANDI- SMENT DATES 44

Adam Grant, professor of Organisational Behaviour at Wharton, notes that "When seeking out specific qualities — research supports evaluating responses using a scoring key.” The challenges companies face are, at their core, similar across size and stage, though they NO TION T present themselves in different ways. As an early stage founder each hire is crucial and you ALENT A S want to take an active part in hiring. Later stage SES SMENT founders are likely to have an HR team in place that helps the department that is hiring with their interviews. However, setting a structure, whether for yourself and your small team or for managers of departments, remains a surefire way to improve hiring outcomes across company size and stage. 45

CREATING A STRUCTURED INTERVIEW PROCESS NONO TION TTION T Published research on employment interviewing dates back to W. D. Scott’s 1915 research into the “Scientific Selection of Salesmen”, published in Advertising and Selling Magazine. In that 100-year history few conclusions have been more widely supported than the idea that structuring the interview process enhances reliability and validity. ALENT AALENT A A 2014 meta analysis* of all the published scholarships on the topic outlined that the SS key ways to improve the reliability and validity of interviews are: SESSES 1. Base questions on a job analysis/spec SMENTSMENT 2. Ask exact same questions of each candidate (in preferably the same order) 3. Limit prompting, follow-up questioning and elaboration on questions (if you must it should be preplanned) 4. Use better types of questions. There are three types you should use: situational, behavioural and competency based 5. Longer interviews with larger number of questions are better than shorter interviews 6. Control (and perhaps limit) ancillary information that is not related to the role/job (less bias) 7. Only allow for candidate questions outside the interview 8. Rate each answer 9. Use detailed anchored rating scales (reduces ambiguity and semantic differences) 10. Take detailed notes 11. Use multiple interviewers 12. Use the same interviewer(s) across all candidates 13. Do not discuss candidates or answers between interviews 14. Provide interview training 15. Rely on scorecards rather than ‘expert’ opinion (i.e. statistical prediction vs. expert) *Levashina - “the structured employment interview” 4646

WHAT ARE THE EFFECTS OF THESE IMPROVEMENTS? R U U e s s Interv R duc er RC er R elia Validity eactionandidaeaction e iewer bility d bias Content te Base questions on a job analysis/spec + + + Ask exact same questions of each candidate + + + Limit prompting, follow-up questioning and + + - - elaboration on questions Use better types of questions + + NONO TION TTION T Use longer interviews or larger number + + + - - questions Control/limit ancillary information + + - - ALENT AALENT A Only allow for questions outside the + - - SS interview SESSES SMENTSMENT Evaluation Rate each answer + + Use detailed anchored rating scales + + + + Take detailed notes + + + - Use multiple interviewers + + + - Use the same interviewer(s) across all + + - condidates Do not sidscuss candidates or answers + + + - between interviews Provide interview training - + + + + Rely on scorecard rather than expert opinion + + + + means positive effect and - means negative effect 4747

WHAT ARE THE EFFECTS OF THESE IMPROVEMENTS? NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT Any interview process could be easily enhanced by using at least some of these components, according to the researchers. Startups will not have the time or resources to implement all 15. Don’t worry! Some components have a bigger impact than others, and we will provide more depth on these. 1. Use a job analysis to develop a scorecard and a rating system 2. Similar questions (interview script/flow) 3. Develop better questions 4. Train the team 4848

WHAT ARE THE SPECIFIC BENEFITS AND CHALLENGES WHEN USING A SCORECARD? NONO TION TTION T ALENT AALENT A SS SESSES THE BENEFITS SMENTSMENT Interview score sheets keep you focused and will help your hiring manager improve. Interview scorecards support hiring team collaboration and helps the team record separate judgements on candidates. Scorecards make your interviews fairer and more consistent and they can help in litigation. Candidate scorecards help you think through and define your requirements. THE CHALLENGES Active scoring limits eye contact. Make sure you write short notes. The scorecard should include what a good (and potentially bad) answer entails so you can tick off. Scorecards don’t allow you to stray from the process and explore tangents. When creating the interview questions you can include exploratory questions (and reflect this in the scorecard). Preparing scorecards requires time and effort, so standardise scorecards in departments and across the business if possible. 4949

CREATING A SCORECARD NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT - Include everyone that will be involved in the hiring process. - Create general categories with a percentage weight (e.g. cultural fit is weighted 20%). These weightings can vary between departments. - Create subcategories specific to the department. These subcategories are the specific attributes /skills / traits you’ll be looking for. Weight these with a percentage - e.g. 2 core values weighted at 10% each. The weight of the subcategories in a category should total the weight of the category. - Develop a question and an answer guide for each subcategory. Be as detailed as possible in the answer guide. - Calculate a total subcategory score which you measure against a pre-set graded scale. It should clarify who falls in your “No”, “Maybe” and “Yes” buckets. 5050

CREATING A SCORECARD NONO TION TTION T ALENT AALENT A Culture SS Sub Question Answer Category Sub Score Total SESSES category weight (%) category (1-5) weight (%) SMENTSMENT Value 1 Why...? Great answer (5) 20% 10% 4 0.4 Bad answer (1) Value 2 Great answer (5) 10% 3 0.3 Bad answer (1) Experience Sub Question Answer Category Sub Score Total category weight (%) category (1-5) weight (%) Specific How...? Great answer (5) 20% 10% 4 0.4 Skill 1 Bad answer (1) Specific Great answer (5) 10% 5 0.5 Skill 2 Bad answer (1) 5151

CREATING A SCORECARD Personality Traits Sub Question Answer Category Sub Score Total category weight (%) category (1-5) weight (%) Analytical What...? Great answer (5) 35% 15% 4 0.6 (make a list where you can easily tick off mentions) Bad answer (1) make a list Problem Solving Great answer (5) 20% 2 0.4 NONO (make a list where you can TION TTION T easily tick off mentions) Bad answer (1) make a list ALENT AALENT A Experience SS SESSES Sub Question Answer Category Sub Score Total category weight (%) category (1-5) SMENTSMENT weight (%) Leadership Who...? Great answer (5) 25% 5% 1 0.05 (make a list where you can easily tick off mentions) Bad answer (1) make a list Teamwork Great answer (5) 20% 3 0.6 (make a list where you can easily tick off mentions) Bad answer (1) make a list Sum 3.25 Once the interview is finished immediately fill out the entire scorecard, grade the candidate and provide feedback. Set aside 15-30 minutes and do not postpone it! 5252

INTERVIEW PROCESS DESIGN Now that you know what to look for, you should aim to make your approach consistent by training your interviewers how to look for the same things. The interview flow maps out the entire process (from first phone screen or test through to final on-site interview). The interview flow that’s shown below will help to minimise bias and improve the overall outcome of the hiring process. 1. Start your interview process with a phone interview. This will provide NONO an efficient way to assess basic traits, attributes and skills. TION TTION T 2. Ask the candidate to do a test. a. For sales roles: ask them to pitch. b. For marketing: have them produce content or ALENT AALENT A suggest specific channels for investment. c. For tech roles a test should be fairly straight forward. The point of the SS test is both to grade the assignment and ensure the candidate has the SESSES minimum level of competency. Don’t make the test too complicated, but rather make sure it relates to the job the candidate will be doing. SMENTSMENT 3. The third and fourth interview should be longer, face-to-face interviews. a. Focus on traits, behaviour and culture (i.e. the person you are hiring). b. Set aside 1.5-2 hours for each interview. c. Start preparing 15-30 minutes before the candidate comes in by going over your question sheet, reading their CV and prepping the room where they will be interviewing. HOW MANY PEOPLE TO INCLUDE? Most startups from seed through to series D+ involve 3-4 people in the hiring process: - One person owns the process. For a small team without a hiring manager this will be the candidate’s immediate superior. This person is in charge of managing communication and directing candidates through the process, starting with phone interviews and tests. - The founder/CEO will ideally meet each candidate at the final round. - A peer/team member who has been with the company for a while and their immediate superior (if the hiring manager is ‘driving’ the process). 5353

REMEMBER THAT CANDIDATES ARE ALSO ASSESSING YOU NONO TION TTION T ALENT AALENT A SS SESSES The interview is not only a chance for you to assess the candidate, but also an SMENTSMENT opportunity for you to sell the company. KEEP THESE SIX THINGS IN MIND 1. Make sure the process is clearly defined and communicated. Professionalism and preparation helps manage expectations, which we know is heavily linked to happiness/satisfaction. 2. Leave 5 minutes at the beginning of each interview for an informal introduction; try to make them feel comfortable and welcome. Leave 10 minutes at the end of each interview for questions. 3. Ask relevant questions. With a structured process any candidate is more likely to perceive the questions as more related to their role and not just ‘fluff’. 4. Let them meet other people outside the group (even if just for a chat that has no impact on the candidate’s score). 5. For internal stakeholders, walk through the reasoning for - and intended outcomes of - the structured interview process so that everyone is aligned with why it’s being done, and why it may feel a little ‘formal’ in places. 5454

TRAINING YOUR HIRING TEAM NONO TION TTION T ALENT AALENT A You should think about hiring a Talent Acquisition Manager once your company reaches the stage when the volume of hires is expected to increase sharply, which will SS usually be immediately after raising a round: read Gemma Lockhart’s take on this in her SESSES interview. Using KPIs such as staff retention measures in the long run and conversion rate SMENTSMENT from initial to offer will help both founders and Talent Acquisition Managers focus on this quality over quantity. Crucially, your Talent Acquisition Manager should help you with training hiring man- agers and peers that take part in the recruitment process. Below are some key elements to educate the hiring team on. TRAIN HIRING MANAGERS TO UNDERSTAND STRUCTURED INTERVIEWS Structured interviews are more objective and legally defensible than unstructured interviews, but it also means a less personable interview process. Interviewers who use this interview format should learn how to prepare questions, understand rating scales and score candidates consistently. Help the team understand the merits of structured interviews, clearly communicate the hiring objectives (and emphasise quality over quantity); help them understand the process and why each part matters. TEACH HIRING MANAGERS ON NOTE TAKING Research shows that the most structured form of note-taking results from: • Extensive notes instead of brief notes • Mandatory note-taking instead of optional note-taking • Recording answers or facts instead of evaluations or judgements • Note-taking during the interview instead of after the interview 5555

INTERVIEWERS SHOULD TAKE NOTES ABOUT: • Important parts of answers that will help them remember what candidates said • Extreme behaviours • Whether key criteria is not met, met or exceeded INTERVIEWERS SHOULDN’T TAKE NOTES ABOUT NONO • Subjective criteria, like “we went to the same school” or “funny” TION TTION T • Protected characteristics, like race and gender • Traits that aren’t job related, like physical attractiveness • Excusable behaviours, like nervousness ALENT AALENT A TEACH HIRING MANAGERS ABOUT BODY LANGUAGE Train interviewers to both be aware of candidates’ and their own body language. Being SS more aware of candidates’ non-verbal cues can help interviewers refine their interviewing SESSES skills. For example, if candidates’ body language suggests they’re anxious, interviewers can make a conscious effort to put candidates at ease. Even if hiring managers think candidates SMENTSMENT are unqualified, they shouldn’t let their body language negatively affect candidate experi- ence. TRAIN HIRING MANAGERS TO COMBAT BIASES The best way to combat biases during interviews is to be aware of them. This can’t be achieved overnight – it takes time and effort. A good start would be to help interviewers discover their hidden biases so encourage them to take H arvard’s Implicit Association Test. Make sure any person involved in hiring does this on a regular basis or certainly before a big hiring push. DEVELOP BETTER QUESTIONS The most basic component of interview standardisation is asking all the candidates the same questions in the same order. While this lacks some of the conversational freedom that interviewers and interviewees may prefer it ensures a similar experience and assessment across the board. In terms of generating questions there are three overarching question types: 1. Skill/competency based 2. Situational 3. Behavioural 5656

TYPE EXAMPLE Skills-based questions are directly related Explain how [a process at their previous to experience and competence with tools, company] worked? technologies, and industry standards. - What were the benefits/challenges They help you find the difference between with this? NONO theoretical and practical knowledge. TION TTION T Situational questions giv e you the chance How would you deal with [a situation that to see how a candidate would react to is inevitably going to be present]? specific scenarios in your office and helps ALENT AALENT A you gauge soft skills like problem solving, analytical thinking, decision making and SS creativity. SESSES Behavioural questions sho w how a How have you handled working with [type SMENTSMENT candidate’s aptitude and approach will of process or manager] in the past? be applied to a task, based on their past experience. They are useful for revealing soft skills like trustworthiness, attitude, values and work ethic. These questions ask candidates to narrate a situation or incident from their past experience which demonstrates a particular skill or attribute. You should have pre-planned a follow-up question for each question you ask, in case the first question is not adequately answered. Decide how to score the fact that the candidate could not adequately answer the first question. The key to interview questions is to ascertain if the candidate can , will and want to do the job they are applying for. These correlate to skill/competency, motivation/ambition and values/culture. In order to gauge these aspec ts of your candidate you should combine testing with standardised questions. We will go into further detail on this. 5757

CAN THEY DO THE JOB? NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT In this section we look at how companies at any stage can assess candidates’ skills, experiences and attributes. There are two tools that a hiring team have at their disposal: tests and interviews (phone and in-person). In this section we look at how successful companies and experienced interviewers ensure the best possible assessment of their candidates’ abilities. WHY USE SKILLS TESTS? 1. First, CVs should not stand alone. As the single mode of assessing candidates, they’re increasingly outdated. They give an overview of what a candidate did and where, but not how they performed or even what they really did. 2. Increase potential reach. When using job boards with CV criteria, you are only able to get candidates who are actively looking for a job. If you market tests you can actually find qualified and interesting candidates based in the first instance on skill rather than experience. 5858

SKILL TESTING FOR TECH 1. DEFINE THE ROLE In tech you have predominantly three types of roles: - Backend - Frontend NONO - Design TION TTION T The tests for these three roles are different. ALENT AALENT A Backend Test problem solving capacity, intellectual fire power and knowledge of tools SS and principles. SESSES Frontend Test problem solving capacity, organisational and structural capacity and SMENTSMENT knowledge of tools. Design Test their ‘eye’, creative ability and knowledge of tools. 2. DEFINE THE KNOWLEDGE Decide on what specific language/framework/tools you require for the role. These elements have to be present in the test result and be easy to tick as done/not done. Backend Required languages (Java, PHP, Python, Ruby, SQL), Frameworks (Zend, Symfony, CakePHP), Version control systems (SVN, CVS, Git), Computer science principles Frontend Required languages (HTML, Javascript, CSS, Actionscript, Objective C), Frameworks (Bootstrap, Foundation, Backbone), libraries (jQuery and LESS) Design Tools (Adobe suite, Sketch, wireframe software like Axure, prototype software like Mockplus), Basic HTML and CSS. 5959

“The questions don’t really matter. What matters first is a clear understanding of what you need.” – Soham Mehta, CEO of Interview, Kickstart 3. DEFINE THE ATTRIBUTES Decide on what intellectual / mental skills or attributes you require for the role. The central challenge of the test should enable candidates to showcase these. NONO TION TTION T Backend Problem solving, Critical thinking, Detail oriented, High level thinker ALENT AALENT A Frontend Organised, User orientated, Technical/Non-technically literate SS Design Visual communicator, Technical/Non-technically literate, Inspired SESSES SMENTSMENT 4. DEFINE “OTHER” SUCCESS FACTORS If there are skills or traits that you kno w increase the likelihood of an employee being successful at your company, make sure you include a way to check for that skill or trait. It could be something so simple as ‘comes up with quirky solutions’, ‘adds an explanatory README file’, or ‘familiar with Agile/Scrum. These skills and traits often transcend the role and relate to company culture and values. 5. CREATE THE TEST BACKEND / FRONTEND Create your own test or engage platforms like HackerRank , C odility and CyberDojo to help create and administer your tests. Make sure the test is representative of work the candidate will be doing. Based on the interview with Stephen Wilcock and testing software HackerRank, here are some examples of tests you can combine: a. Basic code challenge to see if they can properly code. b. Algorithm code challenges (particularly for backend to ascertain knowledge of first principles) c. Multiple choice d. Code completion The aforementioned platforms will help you find the best questions and correct answers. 6060

DESIGN TEST EXAMPLES WHAT? A challenging user sign-up flow (interaction design) HOW? WHY? We’re building a furniture Users won’t automatically know what recommendation web app where we need their favourite colour is. Does the designer to know the users favourite colour. try to create a small game or challenge to During the sign up flow, how can we help narrow down their choice? help the user find their favourite colour? Is the solution short enough to still be What other information might we part of the sign up flow? require for this web application, and how Does the designer suggest further NONO would we go about obtaining it from the interesting ideas for the furniture TION TTION T user? recommendation service? Do they exploit this by including furniture in the colour picking step? ALENT AALENT A SS WHAT? SESSES A mobile app for cruise ship crews (mobile design) SMENTSMENT HOW? WHY? Ask the designer to sketch out a few Testing specific skills related to the list screens for a mobile app. Cruise ships have crews of hundreds and can be away from port for days at a time. Ask the designer to flesh out some wireframes for an application that can: 1. Provide a list of all passengers for the crew member to look up 2. Allow the crew member to mark passengers as special (ie, VIP, or causing trouble) 3. Show some statistics about the ship, such as when it’ll dock next, the weight, speed it’s traveling, etc. 4. Provide a simple messaging service for crew members (ie, to ask for their shift to be covered) 6161

WHAT? A guitar tuner with only one button (product design with a constraint) HOW? WHY? Ask the designer to think about a Feel free to swap the guitar tuner for guitar tuner which only has one button. It something else (eg, an alarm clock or a needs to be able to: TV remote) if the candidate isn’t familiar 1. Turn off and on with the functions of a guitar tuner. The 2. Tune the guitar basic idea of this exercise is to see how the 3. Change between 5 alternate designer thinks about creative and unique tuning patterns approaches to user interaction with a NONO 4. Be easily viewable from a difficult constraint, and also to get them out TION TTION T distance (ie, on stage) of their digital comfort zone with a real life product design. ALENT AALENT A “Tell the candidate that you’ll start simple and make SS it more complex as you work through the problem.” SESSES SMENTSMENT – Greg Badros, Founder, Prepared Mind Innovations 6. ASSIGN INTERNAL ROLES Pick who will be involved from your hiring team. Get separate people to a) score, b) assist the test takers, c) provide feedback to the candidate. Remember, keep your notes and thoughts to yourself in order to avoid members of the hiring team being influenced by each other or prioritising conforming over stating their true feelings. 7. IMPRESS THE CANDIDATE Make sure you tell the candidate early in the process and in clear terms what is expected and how the test process works. Hiring is a two-way street and you have to manage the candidate’s expectations and keep them happy and interested throughout the process. 8. ALWAYS IMPROVE After your hiring cycle ends make sure you analyse whether the test turned out to be a good predictor of performance and employee longevity. As the company passes through fundraising stages this data will become better and better. If correlation is low or negative, make changes to your test. 6262

INTERVIEW TECHNIQUES & QUESTIONS FOR TECH NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT Having touched on general best practice for structured interviews and dug deeper into skills tests, now is the time to think about what to ask candidates during the actual interview. Here we focus on the general themes that are crucial to cover during your questioning (either over the phone or face to face). There are plenty of online and offline guidelines on example questions, so take time to peruse those before you start your process. As a general rule: slowly increase the difficulty of the questions. This gives the candidate a sense of progression and will help contextualise notes and ease scoring immediately after the interview. 6363

BACKEND Themes to explore: Does their skill set fit your existing tech stack, do they know computer science first principles, are they good problem solvers and can perform quick critical thinking. QUESTION ANSWER What is your favourite programming When it comes to backend developers language and why? - particularly experienced ones - this is a “safe” opener, as it is a fairly simple question and a nice way to ease into the conversation. Furthermore, it gives the developer a chance to freely talk about their work experience and preferences. You should judge how enthusiastic they are and whether their skill set fits with your technology stack. NONO TION TTION T How would you describe the software The software lifecycle often fits lifecycle at your most recent position? What somewhere between Agile and the did you enjoy the most? What would you previously popular Waterfall approach, have liked to change? which is more of a progressional flow, ALENT AALENT A where each stage is fully completed before moving onto the next. Therefore it is super SS relevant to ensure that the person you are SESSES interviewing will be a right fit not only for your organisation, but also within the SMENTSMENT development team by ensuring work flow compatibility. What is your understanding Asking this question will allow you and experience with object-oriented to delve deeper into the mindset and programming (OOP)? routines of the developer you are talking to. Object-oriented programming can be a hard topic to explain in just a few lines, however understanding the interviewee’s relationship and experience with OOP is highly relevant in order for you to get a better understanding of what kind of software developer you are dealing with. Since the late 90’s OOP has been the general practice when organising code, which is a way of generalising and objectifying certain groups of code and making it possible to apply them to various situations. 6464

Imagine that you have a large dataset on External sorting algorithm. NONO disk (or in the cloud like S3), and you have TION TTION T limited memory on your servers but you want to sort and manipulate the data. What is the best algorithm/sort to use in this situation? ALENT AALENT A SS Given an array what is the longest This is a classic dynamic programming SESSES contiguous increasing subsequence of problem and tests their knowledge of elements? What is the longest increasing computer science principles. The candidate SMENTSMENT subsequence? needs to find an optimal substructure (which is basically recursive in nature) in the problem so as to apply an iterative solution to the problem. Give them an array with integers (positive, negative, zero) and they are supposed to find the maximum contiguous sum in this array. If the given array is: 5, -6, 7, 12, -3, 0, -11, -6 The answer would be, 19 (from the sub-array {7,12}). If they go on to form all the possible sets and then finding the one with the biggest sum, their algorithm would have a mammoth-like time complexity. They should use dynamic programming and finish quickly. More specific Java, PHP, Linux questions are available at Github and Katemats.com. 6565

FRONTEND Themes to explore: Organisational capability, user focus, does their process fit yours and are they able to communicate in both technical and non-technical language. QUESTION ANSWER Describe your work flow when you create Each developer will share a different a web page? unique work-flow which will give you valuable insight into their organisational patterns and general technical preferences. Look for a match with existing systems and processes. Assume you arrive at a new company that This question drives at problem-solving, has 3 competing style sheets, how would communication and teamwork skills. There NONO you best integrate them into the site? should only be one style sheet within TION TTION T projects. Have you ever used an MVC? If so, which MVC stands for Model View Controller. ALENT AALENT A one and what do you like or dislike about it? MVCs typically organise web applications into a well-structured pattern, making SS code easier to maintain. Their preferences SESSES should be revealing. Candidates who are able to articulate why they use one MVC SMENTSMENT over another, show that they are engaged in what they do, care about the technology they use, and have considered different options. What are your favourite features of HTML5 It reveals whether the front end developer and CSS3 and what would you change? being interviewed is also up-to-date with new features within the core technologies. Look at the design of our website. Talk Test their ability to clearly articulate their me through the features that draw your creative preferences. Generally, developers attention. What do you like, what do you would need 20–30 minutes to look at the dislike? What would you change? page and underlying implementation, so tell the candidate about this question before the interview - but not too long before. Github has a great mix of language, general and interesting interview questions. 6666

DESIGNER Themes to explore: Understand their creative process, how they seek ideas, how they get inspired and whether they are visual and able to communicate in both technical and non-technical language. NONO TION TTION T QUESTION ANSWER What is UX design? Why does it matter? Each designer will share a different unique ALENT AALENT A How would you explain the UX design work flow which will give you valuable process? insight into their organisational patterns SS and general technical preferences. Look SESSES for a match with existing systems and processes. SMENTSMENT What’s your process for working with other Get a sense of their working style. Like most designers, developers, or product managers? things it’s a team sport. • Developers — For effective collaboration with engineers they should show understanding for the technology stack and have a good sense of constraints and opportunities. • PMs — To work well with product managers, they should translate prototypes into proper specifications, usually in the form of detailed stories. • Designers — Designers typically have continuous cycles of pairing and siloing, followed by structured critique sessions. Having self-awareness about the way they work and demonstrating flexibility is the key to success. 6767

What is your design process? Describe the Their projects should follow a typical NONO design methods that you follow. story arc. Ideally they will describe how TION TTION T they have done it in the past. Their approach should be user centered and involve managing external and internal stakeholders. This builds on the last ALENT AALENT A question so make sure they extrapolate on the process and how they go through each SS part of the process. They should be specific. SESSES SMENTSMENT How would you decide which features to Before drilling down on specific features, add to your product? they should clarify who the user is, what the user’s goal is and what problem the product is supposed to solve. They should mention gathering user-generated data. Where do you draw inspiration from? Look for an answer that contains blogs & books, visual content, other good products, industry leaders, etc.. Other than that there isn’t much guidance here. What would you say is the next big trend in Do they know what is coming around the UX design? corner and can they implement it for you? Does the subject matter itself inspire and motivate them? This uncovers whether their job will be intrinsically motivating, which is the strongest source of motivation and a great plus. 6868

EXPERT INTERVIEW: STEPHEN WILCOCK NO TION T ALENT A S SES SMENT Stephen Wilcock is the CTO of Apperio. Before that he was Director of Engineering at Huddle with a team of 50 engineers and VP Engineering for OpenBet, where he grew the division from 70 to over 250 engineers. 69

_ WHAT’S THE POINT OF USING TECHNICAL TESTS IN THE HIRING PROCESS? I use tests in order to both make a qualitative and a binary decision. Tests should eliminate candidates from your pipeline and they should allow candidates to differentiate themselves. In the end though, technical tests are a method for founders and managers to know what they are getting for their money, which is two things: 1) problem solving and NONO intellectual fire power, 2) knowledge of tools and principles. I want it to be possible to fail TION TTION T and pass, but also pass with a clever and innovative solution. I use tests partly to reaffirm CV skills and experience, but mostly to uncover problem solving capability, intellectual fire power and knowledge of tools. For graduates I am looking ALENT AALENT A for the former. For experienced hires I am looking for both. SS WHEN SHOULD YOU USE TESTS? SESSES Founders often use tests as a separate stage in their candidate funnel. It’s important to consider some of the negative reactions tests can have with candidates. They can be SMENTSMENT time-consuming, not relevant to the role or too complex. This means that I have to consider whether it’s a buyer or sellers market for talent. If there is a sellers market I cannot risk putting candidates off with up front code tests. The decision of when I use tests therefore depends on our ability to fill the top of the funnel, which relies on macro conditions, our brand, our channels and the role I am hiring for. NO ONE SIZE FITS ALL Companies at different stages will be able to put aside different amounts of time to build and manage a test process. For startups that are raising money there is a natural cadence. A funding round might last two years and the first six months post-raise will be packed with intensive hiring, followed by 18 months of relative stability. My rule of thumb is that if I am hiring more than four or five developers in a short period of time then I have to take time to create a set and structured test that is used for the entirety of that hiring cycle. Founders and technical leads might not have time to create tests and come up with problems from scratch, but there are technical assessment tools like HackerRank, Codility and CyberDojo that help with this process. However, using automated assessment tools does limit the ability to assess non-machine attributes. Any automated tool I decide to use works by assessing the characteristics of code, not of people. 7070

TIMING AND PURPOSE AFFECTS FORMAT If the candidate funnel is filling at an unsustainably high rate and founders preferably want a “Yes/No” to the two aspects mentioned earlier, an early take-home test would do the job. If the talent market is less kind and a more qualitative understanding of the candidate is preferred, have them perform a test while they are in for interviews. Make sure to then add an element of teamwork to see how the candidate takes feedback, explains tasks and implement improvements. For example: I give the candidate the test and 10 minutes in private. Then I go back in for 10 minutes and go through their thinking (and potential solution) with them. Allow them 5 minutes to improve on the solution. Programmers will not work in a dark room alone. It’s really important to see how they go about solving the problem and how they interact with others. NONO PASSION FOR SPEED TION TTION T There are differences of opinion on speed. Some believe it can get in the way of creative programmers. For me speed is about expectation. At Apperio the tests are logical and not about competency. This is a key distinction which helps uncover problem solving capability and intellectual fire power, rather than whether someone can simply “work fast” or already ALENT AALENT A knows the answer. The tests are then more indicative. Given that founders and senior managers often do not have time for large tests, they rely on observing that candidates can SS solve smaller simpler, common problems very fast. SESSES If you cannot do the simple things quickly you will find solving the complex, larger SMENTSMENT problems very difficult. ASSESSMENT I make sure a minimum of three people interview the candidate. Preferably there should be some overlap between them so that one interviewer can just take notes and observe before they leave. Have a minimum of 2 hours for the interview and include CTO, lead developer and a peer. Keep notes and scores separate. Make sure the same person reviews all the tests and is the one that either works with or observes the candidate during the 10 minute teamwork part. It’s crucial to avoid group thinking, so no conferring. Everyone should reveal their score and decision independently. At Apperio we achieve this by getting everyone that spoke with the candidate in a room and doing a rock/paper/ scissors style reveal of the score after the interview: 1. is “definitely hire” 2. recommend hire 3. recommend no 4. veto After the reveal we discuss. Founders must protect against group thinking. There must be absolutely no conferring of any sort, smiling, winking, rolling eyes. Everyone must be poker faced. 7171

STEVE’S DO’S _ 1 Always get them to write some code, potential. You have to spend the first part of NONO even if it is small. There is a tremendous the interview putting them at ease if they TION TTION T amount of subtle signal to observe in how are nervous, otherwise you will get very someone writes code. How quick are they in little, or they will clam up. So get them to the integrated development environment talk about what they have done, ask them (IDE)? Do they touch type? Do they know some easy questions until they relax a bit. ALENT AALENT A keyboard shortcuts? Do they close their brackets as they open them? Do they have 4 SS certain rituals of laying out code using a Incorporate some open questions. Like SESSES process? Are they careful with indentation? “What happens if I type www.google.com All of these help you build a picture of the into the browser and hit enter, in as much SMENTSMENT candidate's experience and confidence. depth as you understand.” 2 5 Split the language/framework Ask them for their opinion on knowledge from general purpose problem something they mention. “Ok, I see you solving/computer science concepts. For follow TDD at your last place? What is your example, you might want evidence that the experience of that like?” candidate understands computer science concepts like recursion, indirection, trees, 6 hash tables and so on. You could test these Let them ask questions. If they don’t with an abstract problem solving question. have any questions about process/tools/ Language/framework knowledge testing is expectations that is a warning sign. more straightforward. 7 3 Always make notes on the answers to Build their confidence and get the questions, so you can justify your decision interaction to conversational level if if the need ever arises (discrimination / possible - this is particularly key with unfair claims / hiring discussion). graduates. You may have a good candidate who, because of nerves, cannot show their 7272

STEVE’S DON’TS NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT _ 1 2 Adversarial and confrontational Playing ‘stump the programmer’. “Do hiring - the kind you see popularised on the you know what does?”. Apprentice - is a very, very bad idea with Lots of insecure interviewers will try to programmers. If they nail the interview, make themselves look smart by stumping they will run for the hills. candidates. 7373

SKILLS TESTING FOR SALES AND MARKETING NO TION T ALENT A S SES SMENT 74

“The questions don’t really matter. What matters first is a clear understanding of what you need.” – Soham Mehta 1. DEFINE THE ROLE There is less variation in sales and marketing than tech, but your role should distinguish clearly between the two. Sales Internal sales, SDR, BDR NONO Marketing Digital / above the line, partnerships, brand/content, growth hacking TION TTION T 2. DEFINE THE KNOWLEDGE Decide on what specific experience/tools and knowledge. These elements should be part ALENT AALENT A of the test. SS Sales CRM (Salesforce), Prospecting tools (Duedil, LinkedIn), Deal SESSES sheet/productivity tech used SMENTSMENT Marketing Analytics (Kissmetrics, Google Analytics), CRM (Hubspot, Salesforce), CMS, SEO, PPC 3. DEFINE THE ATTRIBUTES Decide on what intellectual / mental skills or attributes you require for the role. The central challenge of the test should enable candidates to showcase these. Sales Persuasive, empathetic, good learner, talker/listener, competitive, optimist Marketing Analytical, good writer, organised, creative, trend aware 4. DEFINE ‘OTHER’ SUCCESS FACTORS If there are skills or traits that you kno w increase the likelihood of an employee being successful at your company, make sure you include a way to check for that skill or trait. It could be something so simple as ‘their ability to work in a team’, ‘genuine product interest’, or ‘they are not coachable’. These skills and traits often transcend the role and relate to company culture and values. 7575

5. CREATE THE TEST SALES MAKE THEM SELL This will help you see if they are more than just talk and whether they are actually any good at selling, but also NONO selling in the right way (aligned with the type of customers, TION TTION T sales peers, company values, etc.). Make sure the test fits your sales process. If your SDRs sell over the phone make them actually call you. If it is face to face sales then sit in the room with them. ALENT AALENT A They can sell you the previous product they were selling SS for 5 minutes - this allows you to see how they sell SESSES something they know. Then switch to selling your product in 5 minutes. This will allow you to see how they navigate SMENTSMENT the conversation when they don’t know everything. WRITING TEST Sales is a lot about following up and writing emails - either cold or hot. Explain who your ideal customer is, explain your product and tell the candidate to: research a company that fits the profile, find a person to contact, write an email to that person. Explain why they chose that company, person and wrote the email they did. This will tell you a lot about what type of person they are, what kind of experience they have and what their ability and talent for this job is. CALL YOUR Make your candidates do research on your competitors (you COMPETITORS can give them a list). Make them familiarise themselves with their products, sign up for a trial or a demo and get on a phone call. This might be difficult to do during the interview so it could be a home assignment. You should gauge how good they are at being a prospect. Do they ask a lot of questions? Do they get all the information they want? If they don’t know how to be a prospect, they will not know how to be a good salesperson. 7676

_ SALES Themes to explore: Are they coachable, does their tone and style suit your sales method and process, are they strong salesperson, are they driven, optimistic and independent. NONO TION TTION T Why sales? Motivation is crucial to performance and this question gauges whether they are intrinsically motivated by sales. ALENT AALENT A Furthermore, if their drive comes from processes/results or other things that you will not be able to meet/fulfil this is SS good to know at this stage. Interviewing candidates is about SESSES assessing alignment. SMENTSMENT Why do you want to sell this Do they have motivation for selling what you are building? product? If they don’t have that affinity it might be difficult for them to speak with conviction to prospects. What do you know about our This can be taught, but it is still a fundamental question to customers? ask. Did the candidate do their homework and are they able to perform research? Even if they do not answer correctly, if they have put in effort that is a foundation to build on. Furthermore, does this customer align with who they were previously selling to? Do they have the tone, ability and experience to sell to your customers? What was the last (or best) This will allow you to gauge whether they are industrious, sales tactic you have learned? coachable and driven. The best salespeople are lifelong How did you learn it? learners and grow continuously. It does not matter if the last tactic they learned did not work out or if they had to iterate and change it. 7777

NONO TION TTION T ALENT AALENT A MARKETING SS SESSES CREATE A CAMPAIGN Set up a digital ad campaign with ad groups and keywords SMENTSMENT around an aspect of your products. Let them select which channel they use (social, adwords, media) and have them explain why across both creatives and channel selection. AUDIT Ask them for an audit of your website usability (using Google Webmaster Tools, UX approach). DESIGN Design the wireframe, copy and workflow for a landing page. Test the creativity and ability to convey the value proposition. BOTH SALES & MARKETING ROLES Test coachability . No matter how long the interview process, it isn’t really sufficient to see how someone handles criticism and direction. Sit down with the candidate and critique their answers and give feedback. Be both positive and negative. Do they take onboard what you say? Also, don’t be too specific. Then ask them: how would you improve on this. If they have an answer it means they understand what you have said, internalised it and not merely blown it aside. 7878

‘The whole process is a great way to see if someone will proactively ask for next steps instead of waiting for you to tell them. If they are not going to close their own job, it’s unlikely they will close deals for your business’ – Steli Efti, CEO and Co-Founder of Close.io NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT 6. ASSIGN INTERNAL ROLES Pick who will be involved from your hiring team. Get separate people to a) score, b) assist the test takers, c) provide feedback to the candidate. Remember, avoid group thinking so keep your notes and thoughts to yourself. 7. IMPRESS THE CANDIDATE Make sure you tell the candidate early in the process and in clear terms what is expected and how the test process works (even that you will provide detailed and curt feedback). Hiring is a two-way street and you have to manage the candidate’s expectations and keep them happy and interested throughout the process. Continuously provide the candidates with feedback on how they are performing in tests and how they can improve. 8. ALWAYS IMPROVE After your hiring cycle ends make sure you analyse whether the test turned out to be a good predictor of performance and employee longevity. As the company passes through fundraising stages this data will become better and better. If correlation is low or negative, make changes to your test. 7979

INTERVIEW TECHNIQUES AND QUESTIONS FOR SALES AND MARKETING NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT We have shared how founders can think about structuring their interview process. We have explored how elements of testing for technical hires can be applied to sales and marketing hires in an era where a CV is not the only point of reference in hiring. Here we focus on the general themes that founders should cover during the other parts of the interview process (either over the phone or face to face). 8080

NONO TION TTION T _ ALENT AALENT A SS SALES SESSES When was the last time you Never = either they are sugarcoating or they are not taking SMENTSMENT lost a deal? enough risks with their prospects. If their answer indicates that it’s “too many to recall” that is another warning sign. What you really want here is for them to analyse themselves. If they are good at analysing what happened and why the deal fell through it shows that they are mature and indicates that they will be coachable. When was the last time you You want to know how people act under pressure. Is the were stressed and how did situation relatable or trivial? Do they seem strong minded you deal with it? or will they be easily thrown off by minor setbacks? Most sales candidates will not prepare for this answer so you will uncover real personal traits. When was the last time you Is the candidate a risk-taker or not? They should be, to took a big risk and it didn’t some degree. They have to be comfortable pushing their pan out? own boundaries. The risk itself doesn’t have to be related to selling, it can be anything in life. You just want to know that when this person sees an opportunity, they’ll go for it. Taking risks speaks to a killer instinct and a confidence that will translate really well in selling. 8181

_ MARKETING Themes to explore: are they data-driven and analytical, are they able to be creative, do they know the concepts and principles of digital marketing? How should you create a This tests a candidate’s quantitative abilities, and you would lead score for inbound leads? only ask it to people applying for certain marketing roles. NONO How does the candidate think about analysing data and TION TTION T what is their data sophistication level? Most won’t get very far and are either unwilling or unable to look at more than one variable at a time, or understand how to analyse a lot of data in a simple way. ALENT AALENT A You want to find candidates who: SS 1. Look at the leads that were closed in one group and SESSES compare them to the leads that did not close; 2. Look at multiple variables at a time; SMENTSMENT 3. Use statistical functions in Excel (or language to that effect), like summary tables, pivot tables, and so on. Between videos, ebooks, The best candidates will know that doing it all is not blog articles, photos, viable. They should sort the content by importance to your podcasts, webinars, prospects and customers. They should have a way to figure SlideShare, Facebook, out this information. Twitter, LinkedIn, and Pinterest ... there’s a lot of potential content our team should produce for inbound marketing. How do we do it all? What’s an example of a Don’t expect the candidate to know how your business lead-generating campaign generates leads. The ideal answer is one that demonstrates you’d be excited to work an awareness of your customers and a thoughtful approach on here (or that you have on the buyer’s journey. Furthermore, they should mention worked on in the past)? channels and mediums that will be key to the campaign. 8282

NONO TION TTION T ALENT AALENT A SS Draw a funnel on the Everyone in marketing should understand the funnel and SESSES whiteboard showing think about how to optimise it. If their role is more focused 10,000 visitors, 500 leads, towards the top of the funnel it is ok if their focus is there. SMENTSMENT 50 opportunities, and 10 They should be able to calculate conversion rates with ease new customers (or any and have an intuitive sense of what good and bad conversion other numbers you think rates are. More importantly, you will test their creativity are interesting). In order when it comes to improving particular steps. to improve these numbers which areas of the funnel would you focus on, and what would you do to change these results? We have two potential How does the candidate approach a conflict of interest? Do designs for the homepage they care what people think, or do they go to the data for of our website supported their answers, such as through A/B testing, user testing, by different parts of the and customer interviews? The best candidates introduce business. How do you logic and marketing methodology into their answers, while decide? removing opinions. Another plus will be candidates who say you should be constantly testing, tweaking and improving the website, rather than always doing a complete redesign every now and again. 8383

WILL THEY DO THE JOB? NONO TION TTION T ACCORDING TO BOTH US AND GLOBAL GALLUP POLLS, ABOUT 70% OF WORKERS ARE NOT ENGAGED AT WORK. A recent Deloitte Insights study also showed that over 87% of America’s workforce is ALENT AALENT A not able to contribute to their full potential because they don’t have passion for their work. Founders may think that data - which includes American manufacturers in the rust belt SS - might not apply to the realities of a fast moving startup because there is more scope for SESSES ‘natural’ motivation and ambition in fast moving, exciting startups. SMENTSMENT But the data does speak to a truth about human psychology and therefore the importance of gauging (and later nurturing) motivation, ambition and potential when hiring for any people. In his book Alive at Work , Organisational Behaviour Professor Dan Cable of London Business School, notes that activating people’s seeking system makes them more motivated, ambitious and engaged at work. So, how can founders gauge whether their candidates have high levels of intrinsic motivation, and the ability to be engaged? YOU NEED TO ASSESS THE CANDIDATE’S DRIVERS OF MOTIVATION AND AMBITION It is difficult to test for motivation and ambition so this has to be uncovered using questions. Most companies and founders value different drivers and sources of ambition and motivation. Do you want them to be driven by a chip on their shoulder? Or perhaps a desire for a better world? Make sure you include specifics in your answer guide, because it is fairly easy to score things like whether the answer is credible or passionate or not. First ask theoretical questions and then ensure you get examples of how their ambition and motivation has affected their actions and execution. 8484

MOTIVATION: INTERVIEW TECHNIQUES AND QUESTIONS WHY IS MOTIVATION IMPORTANT? There are two different types of motivation. Extrinsic motivations ar e those that arise from outside of the individual and often involve rewards such as trophies, money, social recognition, or praise. Intrinsic motivations are those that arise from within the individual, such as doing a NONO complicated crossword puzzle purely for the personal gratification of solving a problem. TION TTION T Intrinsic motivations are the ones you should try to uncover in the interview situation and, crucially, see if there is a match between the candidate’s motivations and what the the job can offer. ALENT AALENT A According to Daniel Pink, a writer on business and management, there are three intrinsic elements of motivation: SS SESSES SMENTSMENT AUTONOMY PURPOSE MASTERY The key for the motivational interview is to determine the candidate’s preferences and honestly assess whether their preferences overlap with what your organisation provides. For example: what are the time, techniques, team structure and tasks that motivates the candidate (in autonomy), and do these match how your organisation operates? 8585

How do you motivate a While this might seem like a leadership question it will colleague? actually reveal what motivates them. When candidates imagine what others are motivated by they are highly likely to mirror their own motivations. This is therefore a slightly backhanded way to ask “what motivates you” and is therefore more likely to get an honest reply. Take note of whether the candidate is describing intrinsic motivational aspects or extrinsic. What demotivated you in Don’t dismiss negative motivation! This will actually your last job? highlight deeper motivational triggers that candidates have. Try to generalise their answer even though it will be specific to their last job. Was it about the people? Was it about processes? Something else? Check back and see if what made their blood boil is something they are guaranteed to encounter in your organisation. NONO TION TTION T What motivates you? There will be two types of answers: extrinsic and intrinsic. Are they working for external rewards, a paycheck or to facilitate non-work activities? Are they working because they are inspired by colleagues, like the tasks and jobs ALENT AALENT A they are doing? A combination might be natural but the candidate should preferably be predominantly in the latter SS category. SESSES What has been the most What types of tasks and what type of colleagues/people SMENTSMENT exciting project that you makes this person tick. Be honest with yourself and have been involved in? Tell assess whether these tasks and people are present in your me about the project and organisation or not. who you worked with? Why did you choose your This question reveals how people approach everyday work. last job? Do they look for new ways of doing things or do they prefer well-established procedures? One answer type should reflect that a person is optional – the y will answer with criteria and opportunities, presenting options. The other type is procedural – they will tell a story and present events in a sequence using procedural language “first”, “second”, “last”. The way of answering this question can tell a lot about the way a person likes working: do they keep their options open or are they procedural. This in turn will be a match or mismatch for the role and your company. Why would you like to Are they motivated by the job, the tasks, the industry, the work for this company? trajectory you are on? Does their expectations match with your reality? 8686

AMBITION: INTERVIEW TECHNIQUES AND QUESTIONS WHY IS AMBITION IMPORTANT? According to academics, ambition is mostly about striving for status and achievements. NONO “It’s not just about working hard or not working hard, it’s about achieving things and the TION TTION T status that goes with reaching those goals” says John Kammeyer-Mueller, an Associate Professor at the University of Minnesota’s Carlson School of Management. Ambition does not have to mean that a candidate has one big, huge goal. In fact, ALENT AALENT A as a company it might be easier to satisfy candidates who show determined pursuit of short term, smaller goals. There is no right or wrong in this context, but founders should SS ensure that the type of ambition (big and rare or small and plenty) candidates have, match SESSES how their business operates and the capacity the company’s leaders have to attend to employees’ goals. SMENTSMENT Where do you see yourself 5 An answer that indicates that the candidate is going to be years from now? somewhere else might be ok. Don’t dismiss them because of this. Your company is growing rapidly and they might be left behind and that is fine. However, if you are hiring someone you are going to need for a long time (or you are hiring someone for their potential) then you should assume they should want to stay. What are your goals for Are they mainly personal or professional? One of these will the future (personal and/or be outside your control so make sure that there is at least professional)? a balance. Here you can gauge whether they are a person driven by one big goal or many small ones. How did you form and Is it possible to inspire them? Are they susceptible to great define your ambition(s)? leadership or does it all come from within. As a startup founder you might not have time to be an inspiring and personal leader to everyone, so self-motivators and people who formulate their ambitions internally will be crucial. 8787

POTENTIAL: INTERVIEW TECHNIQUES AND QUESTIONS NONO TION TTION T WHY IS POTENTIAL IMPORTANT? Founders are building companies that will grow exponentially, while people generally ALENT AALENT A grow linearly. SS As companies scale it is invaluable to have people who can grow with the company SESSES and not be left behind. This does not mean pegging all hopes on potential - there still has to be a minimum skill level. But it might mean foregoing some tried and tested experienced SMENTSMENT hires for others founders gauge to have less experience but greater potential for learning and development. To determine a candidate’s potential to succeed, we recommend that you start at the end and work backwards. If you hire someone and they are successful, what does that success look like? The answer may be different between junior and senior candidates, but the strategy is the same. Consider what they need to accomplish in their first six months to be deemed successful. Chances are it means acquiring skills and knowledge that they currently do not possess. The universal identifiers of potential are extensive learning capabilities, flexibility, a positive attitude, a desire to grow, high performance, high emotional intelligence, honesty and a vision for the future. As a final note: try to notice if the candidate has the ability to say “I don’t know”. If they can’t say “I don’t know,” they probably won’t ask for help, and their potential is limited. As the interviewer, you should be able to come up with a question about some esoteric but interesting piece of knowledge. They should show curiosity as to what the answer is. 8888

QUESTION & ANSWER EXAMPLES Tell me about your Asking about candidate’s working overtime experience is not experience of working only used to check their working capacity, but also to check overtime? their willingness to go the extra miles that will inevitably be required at times in the startup environment. Instead of asking directly about their commitment to work overtime, you can draw a conclusion from their previous experience. However, while having employees who are ready to work overtime is a beneficial, choosing individuals who prefer working faster than working overtime is better as you do not have to allocate additional overtime pay. What will you be Growth-minded candidates leave an imprint long after their remembered for in your last departure. Check for people who changed or developed role? programs that will ‘outlive’ their existence. These are NONO ‘structural’ improvements. TION TTION T Why are you better now at Determine how they have increased their potential. The what you do compared to best answers are about being wiser. Acquiring knowledge/ six months ago? skills is the first step. Leveraging that knowledge/skill is the ALENT AALENT A second. Converting knowledge and experience to wisdom is the final step. Learning from one’s failures can also make SS someone better at what they do. If you don’t learn from your SESSES mistakes, you’re doomed to repeat them. Being better isn’t just about successes. SMENTSMENT What are you currently Knowing their biggest weakness only matters if they are working on getting better doing something about it. The second part to this question at? is asking how it will make them more effective. They should be able to articulate how their efforts will have an impact. Tell me about the last See if they are leveraging the startup community to increase article/video/talk that really their potential. This should reveal curiosity, initiative, and intrigued you? proactive learning. What do people come to you See if they are increasing the potential of others. For junior for help/advice about? level candidates you could ask what they go to others for. Being able to ask for help/advice is a skill often overlooked. Why is your team better A team that collaborates has high collective intelligence because you are on it? and thus high collective potential. The best answer will be about how they contribute to the team rather than how they personally elevate it. Junior level candidates do make a team better; otherwise why are they on the team? At the very least they increase the capacity of the team. 8989

DO I WANT THEM TO DO THE JOB? NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT CULTURE AND VALUES Culture and values lie at the heart of what founders do and relate to every part of the business, particularly talent acquisition and retention. Below we explore how founders can assess a candidate’s cultural fit. And why should we care? A 2005 meta analysis in the Journal Personnel Psychology revealed that employees who fit well with their organization, coworkers, and supervisor had greater job satisfaction, were more likely to remain with their organization, and showed superior job performance. While values can be defined by founders and leaders, culture can often develop on its own and does not work in the same top-down fashion. Testing your company culture through existing employees can be a valuable exercise which will give you clarity for answer guides and the ability to actually test for cultural fit. Workable, a hiring software tool, recently partnered with Sa berr , a company which surveys existing employees for cultural factors and elements, to implement a 15-minute survey that candidates must complete to indicate the candidate’s cultural alignment with the team and how well they will work with specific team members. However, be wary that tests like these embody our innate biases and should therefore be sense-checked. 9090

CULTURE AND VALUES: INTERVIEW TECHNIQUES AND QUESTIONS Using culture and values as an assessment tool can be tricky and founders should be mindful of not only hiring carbon copies of themselves. This is how unconscious bias can fairly easily leak into the hiring process and be legitimised behind “they share our culture and values”. The best way forward is to look for cultural and value fit. This does not mean that someone shares your values verbatim, but that their personality and values fit in under the culture and values that your organisation has formulated. It is difficult to provide a comprehensive guide to good answers on culture, but we will outline some general best practices when it comes to candidate’s answers. NONO TION TTION T Tell me about a couple of This question helps uncover whether the candidate has times you went out of your actively acted on their values (whatever they might be) and way to show [include a relates their actions to your company culture and values. company value here]? ALENT AALENT A What type of culture do you Does the response reflect your organisational culture? Both SS thrive in? the actual culture your employees have described and the SESSES values you have defined. SMENTSMENT If you could add anything to Culture is not a static thing. Hiring is not just about our company culture, what assessing fit, but how a candidate can positively add to the would it be? company culture and bring their best selves to work. Their answer should not be a copy of a current culture and make sure that they give a reason for why that specific addition. Their logic should be that it will eventually positively affect the bottom line, not just be a nice to have. Preferably they should have experience with the cultural element working well in the past. Describe the working Don’t look for confirmation, look for red flags. Are there environment where you are elements that are simply “no go” in your organisation. It most productive and happy. can be elements of relations between peers, supervisor relations, processes, team structure and more. Since culture and values are such unique parts of companies it will be more useful to em- phasise that you should ask questions directly related to your values. Ask both theoretical and behavioural questions in order to get both a sense of how they think about your and their values, as well as how they have embodied them. 9191

5. NO FROM TION T CANDIDATE ALENT A S SES TO SMENT EMPLOYEE 92

NO TION T Offer stage | Compensation benchmarks ALENT A S How much should you pay mid and junior SES SMENT level staff? This varies greatly by role and market dynamics but a rough guideline based on gathered benchmarks is shown below. The numbers are based largely on three surveys: Major Players 2018 survey (3000 surveyed), Reco Group (1000 surveyed and 210 surveyed online) and Propel’s Digital Salary Survey (1261 surveyed). 93

TECH NONO TION TTION T Graduate Junior Mid-Level Senior Architect/ ALENT AALENT A Tech Lead SS SESSES Full stack £32,000 £55,000 £80,000 £95,000 SMENTSMENT £25,000 Java Front End £20,000 £30,000 £50,000 £65,000 £80,000 JavaScript Infrastruc- £20,000 £28,000 £35,000 £50,000 £70,000 ture Automation £20,000 £30,000 £50,000 £70,000 £90,000 test engineer C++ Full £20,000 £30,000 £50,000 £65,000 £75,000 Stack Full Stack £40,000 £50,000 £70,000 £90,000 £100,000+ Node/React Data £35,000 £40,000 £55,000 £90,000 £100,000+ Scientist 9494

MORE GENERAL TECH BENCHMARKS NONO TION TTION T Average Minimum Maximum ALENT AALENT A SS Junior Front-end Developer £25,000 £22,000 £28,000 SESSES Midweight Front-end Developer £40,000 £35,000 £45,000 SMENTSMENT Senior Front-end Developer £65,000 £50,000 £75,000 iOS Developer £55,000 £50,000 £60,000 Android Developer £50,000 £48,000 £57,000 Full Stack Developer £60,000 £55,000 £65,000 Programme Manager £67,000 £60,000 £75,000 Technology Director £80,000 £75,000 £85,000 Product Manager £60,000 £58,000 £65,000 Test Lead £52,000 £45,000 £58,000 Head of QA £80,000 £75,000 £90,000 Solutions Architect £70,000 £65,000 £75,000 9595

MARKETING NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT Average Minimum Maximum Marketing Executive £28,000 £25,000 £32,000 Senior Marketing Executive £34,000 £32,000 £38,000 Brand Manager £42,000 £38,000 £50,000 Marketing Manager £38,000 £32,000 £45,000 Campaign Manager £40,000 £35,000 £45,000 Senior Marketing Manager £58,000 £50,000 £65,000 Online Marketing Executive £30,000 £25,000 £35,000 Online Marketing Manager £45,000 £35,000 £55,000 9696

DESIGN Average Minimum Maximum Junior Digital Designer £25,000 £20,000 £28,000 Midweight Digital Designer £30,000 £28,000 £35,000 Senior Digital Designer £50,000 £40,000 £60,000 NONO TION TTION T Junior UX Designer £30,000 £25,000 £35,000 Midweight UX Designer £40,000 £30,000 £45,000 ALENT AALENT A Senior UX Designer £58,000 £50,000 £65,000 SS Junior Motion Designer £27,000 £25,000 £32,000 SESSES Senior Motion Designer £32,000 £35,000 £40,000 SMENTSMENT Digital Design Director £75,000 £60,000 £85,000 SALES Average Minimum Maximum Business Development Executive £26,000 £23,000 £28,000 Business Development Manager £40,000 £30,000 £45,000 Senior Business Development Manager £47,000 £42,000 £52,000 Business Development Director £65,000 £55,000 £75,000 9797

NONO TION TTION T ALENT AALENT A SS SESSES SALES SMENTSMENT 0 - 2 Years 3 - 4 Years 5 - 6 Years 7+ Years Sales Management * £40 - £50K £50 - £70K £70 - £150K Business Development - £25 - £30k £30 - £45K £45 - £65K £50 - £120K Field Sales Account Management £25 - £30k £30 - £45K £40 - £55K £55 - £100K Key Account Management * £35 - £45K £45 - £70K £70 - £95K Channel Management £25 - £30k £30 - £45K £45 - £70K £60 - £80K Pre - Sales * £30 - £40K £45 - £60K £55 - £70K Business Development - £20 - £30k £25 - £35K £35 - £45K * Internal 9898

MAKING AN OFFER REMEMBER: TIME KILLS ALL DEALS Losing a great candidate at the finishing line is not a great feeling for any hiring founder/ manager. It is therefore key to ensure three things: 1. Speed 2. Alignment with candidate motivation 3. Excitement SPEED Co-founder and CTO of Slack, Cal Henderson: “Make an offer within the day or the next day of the final interview. A structured interview process will yield a clear yes or no - either way the result should be shared with the candidate immediately.” ALIGNMENT According to Adil Ajmal, an experienced engineering leader with nearly 20 years at companies like Amazon, TenMarks Education, Intuit, Posterous and Homestead, “the secret NONO to closing isn’t offering the most equity or saying just the right thing during an offer call. TION TTION T It’s closely listening to and noticing what motivates each candidate, and then explaining precisely how your company will serve those needs.” EXAMPLES: ALENT AALENT A If you’re trying to close a problem-solver , you might continue to share some of the thornier goals or features your team is currently working on. You want to give them as SS much of a window into the tough challenges that actually exist at the company as possible. SESSES If you’re trying to close someone who’s mission-focused , you might want the candidate SMENTSMENT to see themselves reflected back across the table so that they know they’ll be able to make the impact they’re seeking. EXCITEMENT Gusto, the online HR company, does group offer calls. Instead of having the recruiter make the offer call, Gusto gathers everyone who interviewed the candidate in the room, to make the offer as a group. They start the call as a celebration, with applause and cheering, then go around one by one, each person sharing a few sentences about why they personally are excited to work with that particular candidate. This makes the candidate feel special and more excited about accepting an offer. Gusto goes one step further, after the call every person who interacted with the candidate is told by the hiring manager to send a follow-up email. The goal is to show the candidate explicit interest from your company, reinforcing your enthusiasm. If the candidate is senior, Gusto will go one step further, asking the C-suite to email as well. Don’t forget to ensure that the hiring manager/recruiter is providing the candidate with the essential details. All the excitement and speed and alignment count for little if you don’t also ensure that the offer is clear . 9999

ONBOARDING Onboarding is crucial because it is basically the first impression of your company. You might think that the interview process is the first impression. The interview is the candidate’s first impression. Onboarding is the employee’s first impression. It helps employees be better and it helps them stay. According to Perkbox, employees who attend a structured orientation program are 69% more likely at a company for up to three years. NONO TION TTION T A good onboarding has two elements: 1. Structure and relevancy 2. Individualisation to encourage people to bring their best selves to work. ALENT AALENT A STRUCTURE AND RELEVANCY There are lots of things you can do here: SS 1. Prepare a welcome kit. This familiarises new hires with the company brand and SESSES whatever else you put in the pack. 2. Give them a t-shirt/jumper and a personalised name-badge. This makes them SMENTSMENT instantly part of the culture and allows people to quickly spot the new person. 3. Make them the focal point of a company tradition. At Perkbox the newest hire pushes around the beer trolley on Fridays. 4. Departmental bootcamp. Set aside however long they need to get to grips with your current system, operations, inputs and outputs. Make them produce something on their first day. INDIVIDUALISATION Individualisation makes the employee feel valued and safe, which in turn is crucial to building trust. Trust is a prerequisite to enabling employees and fostering innovation. Starting a new job also makes people feel emotionally vulnerable and individualisation mitigates this. 1. Talk to them about their specific role Their manager should explain expectations and processes 2. Show them what success looks like A star employee can introduce the role and the company 3. Give them personalised swag Name badge, computer with a sticker, etc. 4. Best self activation Either get friends / next of kin (relational) or themselves to write about the employee’s best characteristics and share it 100100

NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT Organisational Behaviour Professor Dan Cable has researched the power of individualisation. His experiment used 605 new Wipro employees in India. The results showed that Wipro hires who received a jacket, a personalised name badge and who were made to pr esent their best self to a gr oup of fellow hires showed a 32% improvement in job retention. They also generated an 18% improvement in customer satisfaction compared to the control group (they were call center employees). Founders might think that these things are difficult to define and go through with. However, study after study has shown that an individualised approach to onboarding, where newcomers write and share stories - or stories about them are shared - about their best selves with others, leads to greater performance and retention. Julia Lee, a Professor of the University of Michigan, found that those who undergo relational best-self activation experience significantly less anxiety, negativity and over 200% creative problem solving enhancement. Onboarding goes at the heart of the new war for talent - not wooing employees away from competitors, but unleashing the enthusiasm that is already there within employees, but dormant. 110101

TRIAL PERIOD “18% of new recruits fail probation.” – Opinion Matters survey (2014) NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT The trial period is actually one of the most important parts of the hiring process. You wouldn’t marry someone after three dates (basically interviews) so why should you make a binding decision on who to take into your company family after just three interviews? A startup faces multiple risks, especially when hiring. One way of lowering the risk inherent in hiring is to have a rigorous trial period. Most startups use trial periods, but we want to emphasise that you have not actually hired candidates until they pass their probation. This makes the probation period one of the most important ingredients in your hiring process. Remember that the probation period is also a chance for them to suss you out as well, hence the importance of onboarding as discussed earlier. If you are running a company where 100% of new recruits get through, you should take time to evaluate whether you are taking the trial period seriously enough and whether you need better processes in place to track candidates’ performance and assimilation during this crucial time. 110202

NONO BENEFITS CHALLENGES TION TTION T It takes some time for candidates to adjust Does it limit the number of people you end socially and professionally up hiring? ALENT AALENT A Few hires “hit the ground running” and Yes, but the company is better off you should not expect them to. A trial not hiring a person at all than getting SS period allows you to check candidates’ someone on board that’s not a good fit. SESSES actual cultural fit and their capabilities. It is often very difficult (emotionally and otherwise) to let people go if they SMENTSMENT Candidates will show their true colours are below expectations, but a trial Candidates can show themselves period gives you a process /mechanism from their best sides during tests and for that. interviews, but as they get down to business you will be able to see how Is it viable in a seller’s market? they work with others and how they The best people often like having the deliver. ability to test out the company (as much as you want to test them out). This Protects your equity gives them a way to do that without Though equity is part of most locking themselves in after hearing compensation packages, ensuring that what amounts to the CEO/founder’s even though shares vest from the start sales pitch on why the company is such date any termination before the end of a great place to work. They get a chance the trial period will return the shares to to find out for themselves. the company. What about referred candidates or candidates that have been courted? Same rules apply. Though you should still allow the referring person to be the “evangelist” after the trial period. 110303

NONO TION TTION T _ KEY ELEMENTS OF A GOOD TRIAL / PROBATION PERIOD? 3-6 months long Make sure you understand the trade off here. Extending the trial period might lead to ALENT AALENT A lower acceptance rate from candidates and it will place a bigger burden on your team to keep tracking their progress. In return you will get more data to make your final SS decision. SESSES Onboarding has to be thorough and ensure they can contribute early SMENTSMENT In order to ensure that the trial period is a fair contraption you have to give candidates the opportunity to start adding value as quickly as possible. This means putting in place a good onboarding process that focuses on both the social aspect and tasks of the job. Ensure that you communicate to them that the trial period is there to be used It has to be clear to them that they are still a candidate. If they don’t start vesting shares before they have passed the probation period, you should also let them know that. It has to be very clear to them that you take this period seriously. Be clear about expectations and how they will be reviewed Make sure they have realistic goals and that both their and your expectations are communicated. This ensures the best possible outcome. By the end each candidate should have at least two evangelists that want them to stay. An evangelist can be any other employee in the organisation. You want to make sure that the employee has made a sufficient impact to have supporters within the organisation. Full review at the end with a re-scoring of the candidate. 104104

PAY TO QUIT NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT This program was pioneered by Zappos, who offered employees in their onboarding stage $3,000 to quit. The idea is that they only wanted people who really bought in to the company and who took a long term view. According to Tony Hsieh, CEO of Zappos, only 2% to 3% of people take the offer. After Zappos was acquired by Amazon in 2009, Jeff Bezos modeled the Pay to Quit program for employees at Amazon fulfilment centers on Zappos’ programme. They key is to understand that disengaged employees impose financial and emotional costs on an organisation. For founders who want to do things differently it might be worth considering incorporating a pay to quit programme either on a decreasing, continuous basis like at Amazon, or as part of a trial period like at Zappos. Why focus on engagement? Gallup’s research found that, in 2011-2012, publicly traded companies with 9.3 or more engaged employees for every disengaged employee had 147 percent higher earnings per share than their competitors. Meanwhile, companies with little engagement - 2.6 engaged employees for every disengaged employee - had 2 percent less earnings per share than competitors. This will be particularly true for SaaS companies where people are the real product. 105105

6. NO RETAINING TION T ALENT A S YOUR SES EMPLOYEES SMENT 106

BRIEF INTRO TO KPI VERSUS OKR NONO TION TTION T ALENT AALENT A SS SESSES KPI OKR SMENTSMENT KEY PERFORMANCE INDICATOR OBJECTIVE AND KEY RESULT A type of performance measurement, The Objective tells you where to go and aimed at evaluating the success of a process a Key Result lets you know whether you’re or activity. there or not. OKRs provide the missing link between ambition and reality. They help you break out of the status quo and take you into new, often unknown, territory. A KPI, on the other hand, measures the success, the output, quantity, or quality of an ongoing process or activity. They measure processes or activities already in place. Very often, a KPI that needs improvement will be a starting point for creating an OKR, and it will become a Key Result of an Objective. The simplest way to explain this comes from the OKR management company P erdoo: Comparing an OKR vs. KPI is a bit like comparing a fruit salad with an orange, they both contain fruit, but one is a combination that contains the other. KPIs are what allows you to define Key Results. 110707

SETTING YOUR KPIs & OKRs NONO TION TTION T ALENT AALENT A HOW TO DECIDE ON YOUR KPIs? According to the KPI Forum they should: SS SESSES 1. Provide objective evidence 2. Measure what is intended SMENTSMENT 3. Offer a comparison over time 4. Track organizational progress 5. Be valid, verifiable and balanced 6. Reflect and align with culture and values Your KPIs have to have the ability to cascade all the way down the organisation. In a purely practical sense the KPIs you chose ha ve to be changable through the actions of your employees. Nothing is as demoralising as a KPI that no one knows how to affect. Make sure there are minimal overlaps. For every KPI you define there will be 4-6 KPIs one layer down in the organisation. KPI 1 KPI 2 C-SUITE KPI 1.1 KPI 1.2 KPI 1.3 KPI 1.4 KPI 1.5 KPI 2.1 KPI 2.2 KPI 2.3 KPI 2.4 KPI 2.5 MIDDLE MANAGEMENT 108108

HOW TO DECIDE ON YOUR OKRs An OKR should be: 1. Ambitious. It should feel somewhat uncomfortable. But keep it realistic and definite. 2. Measurable. Objectives are quantitatively measurable by their key results. 3. Public. It is necessary to help you know what matters and what everyone else is working on. 4. Aligned. To keep everybody focused and moving in the same direction. YOUR OBJECTIVE SHOULD BE DELIMITED BY TIME (USUALLY YEAR, QUARTER OR MONTH) NONO It has to be attainable, but should be a stretch. Be careful with stretching it too far TION TTION T though, this will effectively serve as a handbrake by lowering morale and motivation. Each objective should have 3-5 key results, which are defined by your KPIs. Make sure the key results are diverse and rely on multiple departments of the business. Each department could have their own OKR for example, which feeds into the company’s OKR. Think of it ALENT AALENT A as a pyramid. A young company doesn’t have to have this many layers, but there should at least be two OKR layers. SS SESSES SMENTSMENT Company’s mission Company’s OKR Department’s OKR Team’s OKR OKR As a communication tool, OKRs bring two key things to an organization: • Easily digestible direction such that every member in the organization understands how they contribute to the mission; aka focus • Expectations amongst teams and their individual members; aka accountability 109109

_ COMMUNICATING AND IMPLEMENTING YOUR OKRs Ensuring buy in, as with anything in business, is important. Whether you have produced OKRs through a top-down or collaborative approach, make sure you have a ‘launch event’. Explain the OKRs and take questions. A company-wide email is not a launch event. NONO As with most things founders do simply communicating is not enough. There have to be TION TTION T fundamental changes to how you operate and there have to be visible changes to processes for employees to understand that the company is being serious about this. APPOINT AN OKR MASTER ALENT AALENT A The OKR master is someone on your team that is in charge of making sure the OKRs are implemented and communicated. However, they are not responsible for achieving the SS OKRs or holding management accountable. The buck stops with management and it’s not SESSES up to anyone else to hold them accountable. SMENTSMENT The role is similar to being Scrum master. It is a very much role that supports the process, as opposed to a decision-making role. Just as a great Scrum master helps team / scrum to excel at agile software development, great OKRs master helps teams get most out of OKRs. CHARACTERISTICS TASKS They are an employee, not a consultant Ensures that the team follows agreed upon OKRs practices An operational guru Helps team come up with great objectives (executives should not hold this role) and key results Enthusiastic about OKRs Facilitates OKRs adoption Enjoys mentoring and coaching Mentors and coaches team on OKRs process Organized and committed Administrates OKRs tools 110110

An OKR master should be appointed early on in the process of adopting OKRs, as this will be key to a successful OKR launch. One of the biggest traps is to set them and forget them. Appointing an OKR master will help avoid this trap. SET UP A TRACKING SYSTEM FOR KPIs Have a visible dashboard, whiteboard or a company-wide tracker email that tracks the progress towards the Key Results. This can be as simple as a weekly or bi-weekly KPI overview. NONO SET UP A REVIEW SYSTEM TION TTION T To drive performance and accountability, the progress on OKRs needs to be reviewed regularly. Most companies set up short meetings every week (typically Friday) to discuss the progress against objectives and set priorities for next week. It is important to note that those meetings should not be all-hands meetings, but rather each team should have their ALENT AALENT A own meeting. Bi-weekly meetings are fine as well, especially in the beginning, as you want to keep the overhead to the minimum. Remember to send recurring invitations to team SS members. SESSES BEAT THE DRUM IN PUBLIC SMENTSMENT Angus Davis, CEO of Upserve, the restaurant POS and management software, explained to firstround.com tha t the first ‘All Hands’ meeting of every quarter is devoted to talking about OKRs. He quickly runs through results from the previous quarter and then uses the rest of the time to improve how the company uses OKRs and emphasises their importance. This should be an interactive session. MEASURING PERFORMANCE USING KPIs A combination of measuring performance based on meeting quantitative KPIs and on more intangible adherence to culture/values and displays of desired behaviour is a safe option. This ensures that your company does not strangle employees with only numeric KPIs and allows employees to become more alive at work. “We need people to know how they’re doing.” Laszlo Bock, SVP, People Operations, Google PERFORMANCE MANAGEMENT Measuring performance starts with setting targets. Therefore, be honest with yourself: have you and your middle managers defined KPIs and OKRs well enough so that it is clear to employees at any level what is expected from them. Make sure you have gone through the steps outlined above before thinking about performance assessment. 111111

NONO TION TTION T ALENT AALENT A SS SESSES SMENTSMENT A good place to start is to outline the purpose of performance management.What are you trying to achieve? • Are you attempting to create a stronger link between pay and performance? • Are you identifying top performers (as well as under-performers) so you can make more effective promotion decisions? • Do you need to address employee engagement levels? If so, how will you measure employee engagement to ensure your new performance management framework is making progress? • Do your employees need more praise, recognition, and/or clarity of expectations? Are you providing enough feedback? How do you know? • Are you trying to create a more goals-driven company culture? 112112

START BY ADDRESSING THESE FOUR KEY AREAS What How Annual/bi annual Taking a leaf out of Google’s People Operations this performance reviews should include a self-evaluation, 360-degree review (peer evaluation for holistic overview) and calibration of results. Consequently, employee and manager can meet to discuss the outputs. NONO One-on-one meetings Acknowledge that employees may find performance TION TTION T review meetings stressful. Clarify that they should be two- way discussions. Managers should share notes and there should not be any hidden reasoning or information. These meetings should be done every week. ALENT AALENT A Continuous, real-time In order to track people’s OKR even as they make progress, SS feedback which makes the OKRs more real than if they are only SESSES something mentioned every quarter, you can use software such as Perdoo, Weekdone, Lattice or Betterwork. SMENTSMENT Setting objectives Don’t attempt to force goals that won’t cascade perfectly. Goal setting is most importantly, about prioritizing and clarifying expectations. It’s NOT about managing tasks. Discuss the skill development that needs to occur to meet the short and long-term career goals of the employee. Come up with key areas of development with the employee involved. Take into consideration the employee’s strengths and new skills developed through the last performance review cycle. You can read more about Performance Management at Google in Qulture’s Ebook. Remember that the purpose of managers (and by extension performance metrics) is to enable those below them, not control them. If you have spent massive resources on defining a company culture, sourcing, interviewing and onboarding talent, you want to make sure you do not stifle them. This can mean removing managers power from review processes in order to create an honest relationship. 113113

EXAMPLES OF KPIs BY FUNCTION KPIs should focus on action. Work backwards from your goals and emphasise the key levers that your employees can utilise. There are general KPIs that relate to SaaS and to VC fundraising for example. However, there is no consensus on which ones are best to use. There is also a lot of literature on those topics, so we have chosen to focus on departmental KPIs here. DEPARTMENT KPIs • Monthly Sales Target ($) (with goal & percentage for context) • Cost of sales to revenue ratio NONO • Marketing Qualified Lead (MQL) to Sales Qualified Lead(SQL) TION TTION T Conversion Rate • Sales Qualified Lead (SQL) to Win Conversion Rate • Sales growth vs goal • Activity per sales rep. vs wins per sales rep. ALENT AALENT A • Deals closed (#) vs goal • Average sales cycle SS • Average follow-up attempts SESSES • Product performance (list of top performing products/services) SMENTSMENT • Individual Monthly Sales Target ($) (with goal & percentage for context) • Opportunity funnel • Lead response time • Individual Win/loss ratio • Individual Sales activity (# calls/emails/meetings) vs efficiency) • (#) Qualified opportunities (with goal & percentage for context) • Value of closed deals this month ($) vs target MARKETING KPIs • Brand Recall • Branded Search Traffic • Cost Per Acquisition (CPA) • LTV:CAC Ratio • Marketing ROI • Net Promoter Score (NPS) • Press Clippings • Social Media Mentions • Viral Coefficient • Website Conversion Rate • Webite Traffic Growth 114114

SOFTWARE DEVELOPMENT KPIs • Cycle time (amount of time working on an issue) • Lead time • Velocity (delivered value) • Throughput (total work delivered, not just features) • Open/close rates • Work in Progress • Backlog burn rate • Escaped bugs • Defect removal efficiency • Open pull requests • Mean time between failures (MTBF) • Mean time to recover/repair (MTTR) NONO • Application crash rate TION TTION T • Endpoint incidents TALENT ACQUISITION KPIs • Application Completion Rate ALENT AALENT A • Cost per Hire SS • Employee Net Promoter Score SESSES • Team diversity Ratio SMENTSMENT • New Hire Turnover Rate • New Starters per Month • Permanent to Freelance Staff Ratio • Source of Hire • Time to Fill • Training Expenses • Yield Ratio • Retention • Employee Lifetime Value (ELTV) PRODUCT KPIs • Number of new users (daily, weekly, monthly) • New users as a % of your total user base • % User base growth (daily, weekly, monthly) • Active users (daily, weekly, monthly) • Active users as a % of your total user base • NPS • % of customers who repeatedly use the product (cohort analysis) • Customer churn rate • Customer retention rate 115115

NONO TION TTION T DESIGN KPIS ALENT AALENT A • Session duration SS • Average time on page SESSES • Bounce rate SMENTSMENT • Event action • Exit rates and Exit points • Lead time per project FINANCE KPIs • Cash Runway • Current Accounts Payable • Current Accounts Receivable • Gross Margin • Quick Ratio CUSTOMER SUPPORT KPIs • Conversations Per Teammate • Customer Satisfaction (CSAT) • First Response Time • Most Common Issues • Ticket Backlog • Ticket Volume • NPS score 116116

notion.vc © 2019 Notion. All rights reserved

CONTRIBUTORS NOTION Stephen Millard - Chief Platform Officer Maddy Cross - Head of Talent Erling Amble - Analyst NOTION PORTFOLIO Georgiana Barbanta - Talent Acquisition Manager, Paddle Stephen Wilcock - CTO, Apperio Gemma Lockheart - Talent Acquisition Consultant, Triptease & Hazy PARTNERS Abbie Pugh - Partner, Multiple Matt Casey - Co-founder, DOthings

APPENDIX ONLINE PLACES TO ADVERTISE LinkedIn (Sales/Marketing) Github (Devs) StackOverflow (Devs) White Truffle (D evs) AngelList (Tech) Hacker News Who’s Hiring (Tech) Reddit London For Hire thread (Tech) Unicornhunt.io (Tech/Design/Marketing) KD Nuggets (Data Scientists) WorkInStartups (Tech/Design/Marketing/Interns) London Startup Jobs (General) Escape the City (Non-tech) Intern Avenue (Interns) Glassdoor (All) Indeed (All) Reed (All) TotalJobs (All) Monster (All)

APPENDIX EVENTS/COMMUNITIES TO EXPLORE General startup stuff Tech City News events directory Campus London events directory Silicon Milkroundabout – a semi-annual jobs f air Silicon Drinkabout – a w eekly happy hour Developers Gin&JS JSConf Europe HybridConf London Ruby User Group London Python Group PyCon QCon London British Node Conference NodeConf Europe Front End London 12Devs Monki Gras BACON Hacker News London Erlang Factory Geekettes BarCamp London Pub Standards Rails Girls Edge Conference Tomorrow’s Web Node Copter Designers Design and Banter Creative Mornings talks BuildConf Interaction Conference WebVisions Future Insights Product Managers Mind the Product