Adieu, EnVsion
I'm shutting down my company EnVsion. Read on to learn more about why.
I’m currently in the process of shutting down EnVsion. This is of course a difficult decision that I had to make, but it’s one I’ve been pondering on for many months (see this blog post for some background).
The goal of this article is to serve as a post-mortem for the company. I feel I owe it first to myself to try to articulate the reasons why I believe the business didn’t work out.
Along the way, I hope this post will help other fellow founders learn from my mistakes and do better with their own startups. Of course hindsight is 20/20, but sometimes we have to look back and reflect in order to move forward in the right direction.
While I will mostly use the pronoun “we”, I take full responsibility for this failure as the CEO of the company and its last man standing.
Why EnVsion failed
Put simply, EnVsion failed because we built a product that not enough people wanted or cared about.
Part of the reason this happened is because it took us too long to understand and then clearly articulate the problem we were solving. The company went through a series of pivots, and it wasn’t until more recently that the raison d’être for the company crystallised in by mind.
You can see the evolution in our thinking through the successive iterations of the company’s landing page. I found most of the previous versions of the landing page were bad because they did not boil down what we did in simple words.
We’ve done a better job recently; I’d give us B- for the current copy on the hero section of the landing page.
Sadly it was already too late by the time we better understood the problem we were tackling.
Next, I want to go into specific areas where we could have done better.
Market
For too long, we stubbornly stuck to the UX Research market. I felt initially that it could have been a good niche to establish our beachhead, but it turned out to be an arid land for us to land on.
I believe this was a mistake. We should have cast a wider market net, targeting different types of customer-facing professionals, and eventually narrow down our positioning.
The UX Research market is too small. There are far fewer UX Researchers than people in Sales, Product Management, or Customer Success. We were trying to swim in a very shallow pool, and in the end the company perished partly due to lack of water.
I also felt there was a lack of urgency from UX Researchers to adopt a solution like EnVsion. While I sensed that we were solving a pain, it didn’t seem like it was a “hair-on-fire” type of problem.
Finally, I believe this market has low purchasing power. Many of the UX Researchers I spoke with were extremely price-sensitive. Some confided that they had little budget for spending on new tools.
Some others used EnVsion’s then generous free plan plentifully for their research projects and then disappeared into the ether without giving valuable feedback. This was very confusing to me. I stopped searching for answers to this behaviour lest I became overly frustrated.
Marketing
First-time founders focus on product, second-time founders focus on distribution.
I now understand what Justin Kan meant with this phrase. Product is still very important (well to me at least), but marketing is vital.
We simply didn’t do enough upfront and sustained investments in marketing.
Few people knew about us on the Internet, therefore fewer people could try our product. At some point I tried to do more content marketing by sharing my learnings and knowledge about UX Research in order to increase the number of visitors to our site, and up our SEO rankings along the way.
But once again the market was too niche. The low difficulty keywords I was gunning for brought too little traffic, and researching and writing those articles took a huge amount of time. Whenever I did content marketing the product stalled, and I worried the product wasn’t good enough still.
Still, I learned a lot about SEO and managed to climb in the top 3 organic results for some keywords on Google.
I tried a couple of other channels like cold email, but the results were not good there. I also tried multiple experiments on social media networks (Twitter and LinkedIn) but this also didn’t work.
In the end, we never found a channel that enabled us to acquire “good” customers, first consistently, then at scale.
Tech
EnVsion’s tech stack had lots of moving parts. We initially had some contractors help us build the first versions of the front-end for the product. Then I took things over after we ran out of the modest sum we raised from an accelerator a few years ago.
I’m a back-end developer by trade, so my skills on the front-end were weak then. It took me ages to level up in this area so that I could move the product forward at a satisfying pace.
Unfortunately, we also opted for a tech stack that I felt was not conducive to fast iteration.
For instance we used the Rebass component library on the front-end which has not been maintained for many years. It was until a few months ago that I made the switch to Tailwind CSS.
A lot of time could have been saved if we had used frameworks like Supabase and NextJS with TailwindCSS for instance.
As a result of all the above, plus the fact that by then we went back to our full-time jobs, we moved too slowly on product iteration for long periods of time.
Finances
We’ve been funding EnVsion with our own money since 2022, and in 2023 I was left as the only one still funding the company with his own cash.
To date, I have spent around £30,000 of my own money on the company. I even had to sell my Microsoft, Tesla, and NVIDIA shares to fund the company (I don’t want to think about the gains I missed). I believed in what were doing and did not hesitate to put maximum skin the game.
But I lost this game.
This is of course a frustrating outcome, but I’m reframing this money I spent on EnVsion as the necessary tuition fees for my “startup MBA” (I prefer the term “apprenticeship” but I’m not sure it sounds as good from a marketing pov).
Life has changed for me too. I’m now the father of a beautiful 20 months old daughter. Nursery in the UK is damn expensive, and our mortgage payments went up by almost 50% due to the rise in interests rates.
With all these factors in the balance, it would be even more unwise on my part to pour good money into a bad business, even if I enjoyed working on said business.
What I’d do differently
As I said in the opening section, hindsight is 20/20. But let’s pretend we’re time travellers to a period before EnVsion started…
There are many things I would do differently if I were to start again. While I can’t guarantee success either in this parallel run of events, I am confident we’d do much better. Here are the top 3 things I’d do differently:
Tackle a painful problem instead of chasing vague ideas
Operate in a large market where prospects have money to spend on our solution
Start sales & marketing from day one, and if possible pre-sell our product before we ship it
These tips sound unsurprising - you probably read them elsewhere. I certainly have. Yet I ignored those guidelines when I started EnVsion.
I’m now wiser, and hope I won’t forget these principles next time I build something for profit.
Closing notes
I’ve been bootstrapping EnVsion for over 2 years now alongside a demanding full-time job and new parent responsibilities.
Since I have a day job, I’d work on EnVsion early in the morning, often in the evening, and on weekends.
Working on this company often seemed like play to me, which may explain why I could sustain working so much on it with such limited time.
There have been many tough moments in the journey, but these probably yielded the strongest personal growth too.
Things didn’t work out the way I expected, but I’m still proud of the product we put out there, and I’m grateful to the people who believed in us.
Strangely enough, I feel more confident in myself despite this setback. After all, I single-handedly carried this company on my shoulders for a long time.
I’ve learned so much, and while there is still much more to learn and do in my next ventures, I feel know I can do it.
What I’m working on next
I don’t know yet what I’ll be working on next. For now, I just want to follow my curiosity; I feel in 2024 I will be most often in explore rather than exploit mode.
And right now, my curiosity is taking me back to plunging deep and wide into AI, from understanding and implementing techniques like Stable Diffusion to leveraging LLMs for specific use cases.
I’ve decided to learn AI in public and will be sharing more about my journey on Twitter and on my personal blog (coming soon). I’m also going to curate a Github project called Adventures in AI Land with important and interesting notes on AI stuff.
I remain a builder at heart so I’m still planning to ship a couple of projects later in the year.
Stay tuned 🎬.