I first heard about YCombinator in early 2011. That summer a friend and I had an idea for a startup and started teaching ourselves to code. That November we were sitting in front of Paul Graham pitching him our idea.

It went terribly. It was like running up a steep hill during a mudslide. PG would frustratedly sigh as he made repeated attempts to understand our product. I could feel it all slipping away.

We left the interview dazed. Peter and I knew that we wouldn’t be getting the happy call that night. We knew we’d be getting the dreaded email. It came as we headed to the bar where we were meeting friends to celebrate or commiserate.

PG said our idea sounded interesting and that the partners enjoyed meeting us. However, they couldn’t get their heads around whether people would actually want our product.

We were shocked. Not that we didn’t get funded, but at the reason we didn’t get funded. We knew there were all manner of reasons not to fund us: skepticism about our ability to gather the data we needed, a poorly designed algorithm, ugly design, flawed strategy, a weak founding team.

But to think that no one would use the product? We had talked to hundreds of people who were saying they would love a product like ours. The one thing we knew we had for certain was an idea people would love.

In the months following we cast about for the real reason for the rejection. Was it that we couldn’t execute on the plan? Was it that we couldn’t get the data we needed? Was it our feeble prototype? Eventually we gravitated toward a harsh truth: our founding team was not impressive.

Three years later I think I understand a bit better.

What we didn’t have was any evidence that people wanted our product. We had no users.

You hear this kind of advice all the time: talk to users. I deluded myself that we were doing that. We weren’t sitting at home fearfully coding by ourselves. We had done interviews. We were talking to people every day. But we weren’t actually talking to users because we didn’t have them using the product. The YC partners are experienced enough to know that a few people saying, “I would totally use that!” at parties and meetups isn’t the same as having actual users.

PG and the YC partners were being genuine. They may well have enjoyed meeting us and probably did find our product intriguing. They just couldn’t get their heads around whether people would use it because we didn’t give them anything to hold on to. Looking back it’s all so simple and obvious. We just needed to get users. If you have users that love your product you will have no trouble attracting investors.

At some level, I think all of this did sink in subconsciously though because I made my 2012 resolution to “Get Users” and in July 2012 launched Blockr which still remains the most heavily used product I’ve built. But it feels good to get some conscious closure on the interview and to realize that PG’s rejection email was not an indictment of our team or a reason to doubt ourselves going forward.