What is Feature Crawl?
A pattern language for pre-launch builders.
When I was initially building WAND, an app that connects customers to housekeepers through and Uber-like model, I had an affliction that a mentor and investor at the time referred to as feature crawl. Feature crawl is the stage after a first version product reaches the point where it would be good enough for market, but the founder or builder continues to tack on feature after feature beyond the point of it being valuable—at least to the point of an ROI.
This may sound a lot like perfectionism, and maybe it is. Both builders in feature crawl and perfection seeking artists are playing the infinite game, because they are reaching for some version of reality that doesn’t exist. The subjective incentives for playing this game however are where I think the 2 differ.
Why not to crawl
When I was in feature crawl with WAND, we had a working app—with some bugs, our initial pool of cleaning providers ready to accept work and even a small customer base. All that said, I was still hesitant to publish the platform in app stores outside of beta—and it wasn’t because I was seeking perfection.
After a point, I had a misguided conception that with each additional tweak or improvement we made, that we would be working towards a greater first impression when we finally moved out of the gates to meet our first real customers. What I didn’t realize at the time was that this was more like being in the kitchen, adding more and more food to a hungry person’s plate, while all they really want is for you to bring out the damn food.
Your customers can only eat so much, and if they like the food, and they can always order more.
Getting back on your feet
2 things woke me up from feature crawl.
Mentor Input
Running out of Money
Let me tell you, #2 was a stronger incentive to release, but better to get out the gate before you’re counting down your runway.
Some other indicators that you’re waiting to long to go live may include:
Frequent contact from key customers waiting on your product. Chances are if they are bugging you, they feel the pain you alleviate and want the solution now. Even if you have room to grow or make that better first impression, take the request for early access as an opportunity to build a tighter bond with an initial customer. First customers are disciples and ambassadors. If you start working together earlier, they can also be QA testers, focus groups and not to mention a source of revenue. While they may catch problems or room for improvement here and there, you will shine all the brighter if you are able to receive and iterate on that feedback, making them feel heard and involved in the process.
Having the feeling that you want to add more but not knowing where to go next. Market feedback will answer that question.
If your LinkedIn work experience has said ‘Founder at Stealth Start-up’ for more than 6 months.
TL;DR
A picture is worth a thousand words:
cc: @dagorenouf
I wrote this up because as guest on a recent podcast, I mentioned this idea of feature crawl and had a few folks reach out to ask about the concept. When I searched the internet for a canonical origin of the pattern language, I did not see one and so here we are.
Next week, I’ll be back to a book summary post for Voltaire’s Candide, Zadig and Selected Stories.
Move fast, take chances.
-Benjamin Anderson