Back in September, we were iterating on many ideas. We had come up with a to make sure that we were not wasting too much time on bad ideas. 2-day invalidation process Finally, we had an idea that we couldn’t invalidate after 2 days. The idea was the following: that would be free (we would make our money through the interchange fees of our credit card). an integrated banking, accounting and tax service for the workers of the gig economy The market size was huge, with 57 million independent workers in the US. The issue we saw was that the market was scattered across a lot of verticals, and it would be hard to have consistent messaging across all of them. We decided to start with drivers (Uber, Lyft, Amazon, Postmates, etc.), as there was a big opportunity in tax cuts through the mileage deductions, and there are a LOT of drivers to address. Our first YC application After a full week of trying intentionally to disqualify the idea, we decided to apply to YC, as it was the end of September, and it was the deadline for YC application. We continued our invalidation process, taking some Ubers and Lyft rides to ask drivers about how they did things. At that point, we were offered a YC interview. We started talking with people from the industry, too, and one day we received 2 disheartening pieces of information from two different sources — an Uber insider and a Lyft insider — on the same day. Both were working on similar fintech products for their drivers and were planning to release them for free as well, with the exact same business model we had planned. With this information, we faced a decision whether to continue with this idea. We tried to find some other verticals that would have made as much sense, but couldn’t. So we decided to withdraw our YC application 3 days before our interview. Our second YC application At that point, we were back to square one. So we continued to iterate on our ideas, trying to invalidate them. One idea led to another. We finally had one idea that we couldn’t disqualify. We wanted to take on all the operational complexity involved in any data exchange between two parties. We were targeting dataset exchanges between a data provider and a data consumer. After three weeks, we decided to apply again; we got invited for the interview the next week and got accepted. Note that we got accepted into YC with a completely different idea than too :). Dataline Why would YC let you do that? YC calls companies that don’t have a product yet “pre-idea projects”. By that, they mean that there is a very high chance that your idea will evolve during YC. Essentially, you might pivot a few times during the three months of YC. We were definitely one of those who did. And there were many of us in the batch. Why would YC accept projects with a high chance of pivoting in their batches? Our understanding is that they don’t really care about the idea you apply with per se, but more about: the team (experience, ability to build the product in an autonomous way)how we think; if we tend to have uncommon insightshow we iterate and learn fast In our case, applying twice wasn’t actually a bad sign; we were showing our perseverance and our ability to iterate fast. Plus, in the second YC application, we wrote a few of the ideas we iterated upon. That’s actually why they ask this question in their application: “ If you had any other ideas you considered applying with, please list them. One may be something we’ve been waiting for. Often when we fund people it’s to do something they list here and not in the main application”. If you’re very early stage, this is an important question. And it could show the velocity of your team. We’ve seen teams that were pivoting all along the batch — even in the first week, because their partners told them they thought that their team could do better in terms of project potential. So don’t despair about YC. You can get in with a three-week-old idea, but you need a good team that has the skills to implement the said idea and is convincing enough with their abilities that they will want to bet on them. Originally published at https://dataline.io on March 27, 2020.
