Adventures in code
Today I’m sad to announce that we’re shutting down our startup Reccoon. Reccoon was meant to be the future of waste management software, but instead it’s going to be the past.
About a year ago Silas Roswall, Jeppe Reinhold, Carsten Bo Larsen and I started building a platform for businesses to get easy trash pickup, discover their environmental impact and how well they were recycling. It was hard to get traction for that idea on two fronts. Most businesses didn’t want to spend very much time thinking about their trash, and the waste haulers weren’t actually able to provide any sort of meaningful data, as they were either using pen and paper, or using very old systems with no way to export the data.
Even in spite of the fact that the businesses weren’t that keen, we believed that the future of waste management was more transparency. We thought that if the waste haulers could give their customers rich data about their waste usage, the customers would want it. So we set out to digitalize the waste haulers’ workflow, in an effort to make that data available.
We expected that we could replace the pen and paper that most of the smaller waste-haulers used, and then when we had battle-proven technology, we could scale up to some of the bigger players.
Here, about 10 months later, we never really managed to nail that properly, so we’ve made the decision to shut down while we still have quite some runway left. We’d rather be in the lifeboats now than in the Titanic when it hits the iceberg.
While we’ve done a lot of things right, we’ve also made a lot of mistakes, and we think we have a pretty good idea of what we could have done differently. A lot of these are common mistakes, but some things you don’t really internalize properly before you’ve been personally scarred.
A lot of startups have the luxury of being in a position where their product can be imperfect or crash, and nothing terrible happens. We didn’t have that luxury. In the worst case, if Reccoon didn’t work, we would leave millions worth of waste-hauling equipment doing nothing, as they would be left without a plan for the day.
Something that changes the entire daily structure of a business is hard to sell. People are understandably hesitant to change all of their daily operations. We also had a mismatch between the price we could charge, and the onboarding costs of a new customer. Waste haulers have a lot of unstructured, hand-written, data lying around: customer lists, when to pick up what, etc. Onboarding a small waste-hauler could take up to a week of full-time work. That was way too much work for the price we were able to charge.
It also turned out (perhaps not unsurprisingly) that many small waste-haulers aren’t particularly keen on technology. It was difficult to find enthusiastic early-adopters. We did find a few, but our product wasn’t 10X better than the alternatives, so it was only a few we were able to convince to give it a shot.
This brings me to our next mistake:
Through our entire process we’ve been very good at mistaking politeness for interest. We had a strong network in the waste hauling industry, and quickly managed to get a few customers that were willing to help us with feedback during the development process. We called around, and found 34 companies that were “interested in hearing more”.
It turns out they weren’t actually interested at all. When someone calls you and asks if you think that NewShiny would be cool, and if you’d want to hear more about it in six months, it’s much easier to say yes than no. There’s no commitment required by you, and even if you don’t really have the problem NewShiny is solving, that’s a problem for future you. So the solid interest we thought we had, was really just people being polite, and us being okay with that.
Because we thought there was a lot of interest, we didn’t spend much time actually observing the users’ current practices to look for problems. Why would we? They’d already said that they were very interested! We practiced user centered design, and interviewed users multiple times, but it was only in relation to our proposed solution, and not in relation to their current problems. Had we spent more time with the users before looking into solutions, we would probably have realized that their problems weren’t that big - meaning the value we could generate wasn’t that big either.
Misjudging the interest had several consequences - we vastly overestimated the market size for our product. It turns out that most smaller waste haulers are perfectly okay writing things in a spreadsheet or on paper notes, and they don’t actually care much to change it.
That just left us with a few more technology-oriented waste-haulers, and the large haulers, which were the ones most prominently being limited by the software they were using. But enterprises are cautious, the sales cycle is long, and the data migration required would be immense.
Slowly, those 34 interested companies dwindled to less than ten, the potential market had more than halved, and our prospects were looking grim.
We were not a team that enjoyed sales. We enjoyed building things. People have a tendency to try and do things they like, and avoid things they don’t. We were no exception to this. As soon as we thought we had enough market feedback, we started building. A lot. We didn’t pick the simple, bare-boned technology that would let us iterate the fastest, and test out things. No, we wanted to Do It Right™
We wanted a polished product with a wonderful user experience. We wanted it to be prettier and nicer to use than anything else out there.
The final result was pretty good, but at what cost? We spent a lot of time building a wonderful product, to find out that nobody really wanted to use it. We told ourselves we did this because “we cared about the users” - but in reality, we did it because it was enjoyable. We did it because while we were building, we didn’t have to do the things we disliked. Startups are often about doing the hard things, rather than the easy - and we weren’t good enough at that.
Creating a startup is hard and exhausting, and closing one potentially even more so. It’s not easy giving something your all for months or years, and realizing that it’s not enough. However, if you’ve been doing it for a while and realized you’re not going to make it, stopping is bittersweet. You’ve failed - but at least now it’s over. Time to take a deep breath, reorient ourselves and figure out what we’re going to do next. Whether we’ll split up or try to do something together as a team - time will tell. Even though we’re down right now, we’re still looking up, excited about the future.
I’d like to thank Silas, Jeppe and Carsten for going on this journey with me, and Oliver Nørregaard, Nils Alm, Torben Falck and Jan Bodilsen for telling us the things we needed to hear, even when they weren’t what we wanted to hear.