Hey everyone. I recently came out the other side of a tough entrepreneurial journey, having closed my last startup down and laying off the entire team. I decided to write down my lessons. Hopefully, they’re helpful to this community. (It’s a 10min read!) Here goes:When anyone starts a company and puts so much into it, shutting it down is never the outcome you want, nor expect. It was a result that I thought would never happen to us. We set out with an ambitious vision to help content creators make a living doing what they love. We made over $1M in our first 18 months. In 2017, our product exploded onto the scene and we saw hundreds of thousands of users flood in. We paid over $1M to nearly 20k content creators. We raised a total of $4.4M across two funding rounds from incredible investors — GawkBox was a horse you’d bet on, and many people did. Yet after 3 years of trying, we couldn’t find a business model to make it work.Here are some of my key lessons:#1 Get as close as possible to your customer(s) and understand their biggest problemIn my opinion, the single most important indicator behind the potential success of a business is how well you know your customer(s). It is imperative to understand their biggest problems so you can experiment and deliver the right solutions to them. Just as crucial to have them close by as you work through the inevitable periods of product iteration — it’s pretty much guaranteed that you won’t get a product right immediately and customers are a crucial piece to help you get there.My experience these past 3 years taught me about multiple different components to this.The more types of customers you have, the harder it is to understand them at a deep enough level.At GawkBox, our product value proposition was a unique, if slightly complex one — we enabled viewers of live streams to exchange their time playing mobile games (from sponsors) for a monetary donation to a streamer. That’s three different types of customer: the mobile game (the advertiser), the streamer, and the viewer. It takes time to get to know one type of customer. Doing the same with three is close to impossible — and something that bred a lack of focus for us.Data from the wrong type of customer can misguide your strategy.At GawkBox we did a great job of having conversations with streamers — our community team developed long-lasting, valuable relationships within the community across Twitch, YouTube and Mixer. However, we did not weigh customer insight and feedback appropriately based upon their potential impact on our business. We needed to acquire many of the largest streamers in the world to reach the scale needed to make our ads-based model work. While we had strong relationships with smaller streamers (<100 viewers) — they often have very different priorities and goals to larger streamers who have thousands of viewers. This led us to make decisions based upon how smaller streamers view the world and create a product that we didn’t know was a fit for larger streamers — customers that we needed to bring on board to make the business work.Avoid building in a vacuum — draw customer data into the product team.Too often we would spend cycles building a feature that we thought would solve a problem we identified with our customers. We’d release features that not enough people used or worse pissed users off in too many cases. We built too many things in a vacuum. While I believe that you cannot totally remove the guesswork and assumptions associated with launching an innovative product, you can take steps to increase your chances of success by talking to your customer as much as possible and integrating them into your product development process. Towards the end of our company’s life, we kicked off a Product Advisory Board initiative and enlisted a small group of streamers to play a larger part in shaping our product — from the design phase all the way through to testing. Unfortunately we didn’t get far enough to involve them in our product development process.The CEO needs to talk to customers, too!As CEO, I did not talk to enough customers early enough in the life of the company — often deferring conversations on to our community team. As CEO it’s hard to find the time amongst the million other things on your plate, but my biggest takeaway is that this needs to be prioritized above all else. A culture of understanding the customer needs to be set throughout the organization, a culture which starts with the CEO.Takeaway: Identify the right customer for your business, involve them in your design and development cycles. Spend as much time with them as possible — especially important for founders / CEO.#2 Build a product that your customers need.We set out to enable the 99% of viewers that never send cash donations to support their favourite streamer by creating a way for them to do it for free. Our thesis was that ‘free donations’ could grow to be an even more significant revenue generator than direct cash-based donations — mirroring a trend from the mobile gaming world where in-app advertising is now generating more money than in-app purchases in certain genres. For our product to be successful, we needed viewers to want to donate to their favourite streamer.What we learned was that in many cases a streaming audience is not all that altruistic. While some viewers received live shoutouts for their donations, many saw nothing in return for donating to the streamer — which meant that we were too often relying on a viewers’ benevolence to continue to use our product.We found that relying on the viewer’s philanthropy was a tough place to be. We experimented with different features to try and create more perceived value for the viewer, to little positive effect. As viewers’ were not motivated by a clear value exchange to keep using our product, the streamers that relied on them for donations also started to leave. As streamers started to leave, our mobile game sponsorship opportunities dried up.Takeaway: Create a product that solves an ongoing pain so that people will pay for it...repeatedly. We didn’t create a product that viewers saw enough value in to keep using — meaning we were relying on altruism rather than a need to grow our business.#3 Distribution is just as important as productAt GawkBox, we made assumptions in our growth models which meant we didn’t think strategically enough about distribution.Develop a consistent understanding of how many customers you need to reach your key milestones.As we prepared our forecast models to estimate what distribution we needed to reach our goals, we made natural assumptions based upon our early data. Unfortunately, our early data gave us false positives — our LTVs were actually much higher than expected during our first 3 months because our advertisers were paying us unsustainably high prices. This early LTV data led us to build a strategy around smaller streamers under the assumption that we could extract enough value from them.When advertisers' prices dropped to more sustainable ranges, our LTVs dropped significantly — meaning we had to acquire a lot more streamers to reach our milestones. Had we been more responsive to these important data signals in the business, we likely would have made different conclusions about our distribution or product strategy.Develop a distribution strategy for customers that deliver positive unit economics (LTV > CAC).We found that distributing any of our products to individual live streamers was extremely hard. While we did well to acquire close to 20k streamers, it was an extremely time-consuming and expensive exercise — taking 2 years of brute force sales outreach, with an outbound sales strategy (and acquisition cost) more akin to a B2B SAAS company. The difference was — our LTVs were minuscule in comparison.We focused on the wrong size customer.Live streaming has an extremely long tail. According to Streamlabs, there are over 7M active streamers across Twitch, YouTube, and Mixer with an average of under 30 viewers per stream. The majority of streamers have such small audiences that they could only really deliver minimal impact to our business. With 75% of viewing hours on Twitch sitting with the top 5,000 streamers, we would’ve been far better placed targeting the top streamers where the majority of the audience sits. Interestingly, we later found that acquisition and ongoing management costs were similar for streamers no matter the size of their audience — many streamers are continuously bombarded by new tools and promotions, making competition for mindshare fierce. In the end, we never managed to find an effective, scalable method of marketing to streamers and our customer acquisition cost was way higher than our customer lifetime value — our LTV / CAC ratio averaged just 0.14.Always be open to distribution partnershipsKnowing the scale of customers we’d need to acquire to make our business work and the sheer difficulty/lack of scalable channels to acquire them, we should’ve thought more about leveraging partnerships to open up new distribution channels. Aside from the behemoth platforms, only a few companies in the live streaming space have enjoyed truly significant distribution success. However we became too hung up on concerns that they’d be competitive to us instead of exploring potential partnerships — and ultimately our paranoia about competition prevented us from exploring partnerships like this that could have made a material difference to our penetration.Takeaway: Carefully understand how distribution models affect business economics (CAC vs LTV) and consistently test and update assumptions. High LTVs are needed to support outbound sales — low LTVs need scalable channels.#4 Validate people need your product before investing in growth.After we raised our last round of funding we hastily set about putting our headcount growth plans into action — doubling the team to near 20 within 3 months and increasing our monthly OPEX significantly. By September 2017, we were spending an equal amount on sales & marketing as we were on engineering. At that point, our product had shown only poor retention metrics.On reflection, we ramped our sales and marketing spend out of sync with our product development. I believe that we should have kept things leaner and focused resources on R&D to run rapid product experiments alongside our core customers to validate our ideas and improve retention. Instead, we let our focus be diluted. As we added more people and more customers, we became ever more distracted and unable to refocus on validating the right solution for the right customer.Takeaway: Keep things lean (in particular in sales & marketing) until you’ve seen positive retention metrics!#5 Judge your personal success on how much you learn.After coming to the difficult decision to close the company down, I was left asking myself an array of different questions:“If we’d made different calls, would we be in a different position today?”“What did I do wrong?”“GawkBox consumed my life for over 3 years. What am I going to do now?”“Will investors believe in me again?”“Am I a failure?”The feeling of being a failure is a natural reaction in this circumstance. I initially found it difficult not to judge myself by the positive financial outcomes (or lack thereof) that I expected to deliver to my family, team, and investors who believed in us.Yet it is commonly believed that 90% of startups fail — for a variety of reasons, some of which can be out of your own control. While the importance of a successful financial outcome should not be minimized, I learned more during this 3-year journey than anything else I have done professionally before. That experience is invaluable and leaves me better positioned to drive a different outcome next time around.Some recent advice from a friend & mentor stuck with me:Approach every project with a learning goal alongside any financial ones. Judge your success based upon how much you learn.Takeaway: It’s difficult not to be consumed by feelings of failure, but it is important to recognize how much you learn and grow starting a company — often gaining knowledge and skills that are hard to come by in any other capacity. This will set me up well for whatever my next endeavour is.Closing thoughtsFirstly — I appreciate you reading this far!These are just a few of the important lessons I learned as a first-time CEO of a venture-backed business. There are of course many other lessons that I chose not to include in this post so it didn’t turn into a novel.I’m already getting back on the horse and taking my next swing. I’m trying to take these lessons and craft a different outcome in my new venture, Pickaxe, a sales prospecting extension for Google Sheets. Check it out at getpickaxe.com see hubwealthy.com/wealthy
