I’m back to The Valley newsletter. I’ve been swamped on our fundraising in the past months, and if you think it’s hard to raise capital for a startup, try to do it for a VC fund in its early days :). I could say a lot about that, even if I invested in the seed round of multiple unicorns in San Francisco in the past. But this is another story, and today I want to publish an interview with a great company founder, hoping to help many of you do the same.
A few weeks ago, I asked FlutterFlow CEO Abel if he wanted to tell me how a batch of YC works from the inside.
FlutterFlow was founded by two former Google engineers - Abel Mengistu and Alex Greaves - who wanted to make it easier for designers, developers, and entrepreneurs to create mobile apps. Using FlutterFlow’s simple drag-and-drop interface, anyone (yes, anyone!) Can create a fully functional app in under an hour.
Disclosure: we invested in FlutterFlow during W21 Demo Day.
This is truly the uncut version of the interview.
Massimo: I would like Italian startups to understand how it works when you are accepted in a YC batch. Probably things were different before the pandemic; probably now the batches have a different schedule and different organization. What is your story?
You applied, and then what happened?
Abel: So, our story is probably a little different than most. We applied after the deadline, and this was the second time we applied to YC; we applied a year before with another startup. That time we applied on time, and I remember that just applying was a huge learning experience: answering all those questions and having great answers for them is a good part of the YC experience, I think.
So, with the first company, when we applied to YC, we kind of struggled answering some questions. There we started seeing that we were trying to force a solution to the problem, the market wasn’t really there, the problem wasn’t very strong, there wasn’t a huge need.
We kept working on that company, which eventually failed, but while we were building the app we realized that the app-building process was extremely inefficient. So there we kind of started thinking about FlutterFlow, it was September 2020, and the YC deadline was mid-September. But we weren’t really thinking about YC yet; we were just discussing the idea.
Then in early October, we gained even more conviction for the idea after researching it, but we were already 2 weeks past the YC deadline. One of our friends, who is a YC alumna, told us “Just apply, it doesn’t matter, I’ll send an email to one of the partners and maybe they can get you an interview if they haven’t closed their selection yet.”
So we did that, and we applied for the interview. Just the application took us half a day of work. But I have to say that compared to the first time we applied with the previous company when the problem is out there and the need is clear it is much easier to fill the application.
At that point, we didn’t have anything, just the idea. In the following week or two, we didn’t do much, we weren’t really sure about what to do if to turn it into a company or not. Once the day of the interview approached we thought “We should at least get in with something, whether or not we are going to do a startup”. At this point we were very emotionally drained from the last startup, we weren’t sure if we were actually going to redo this. So we thought “Let’s quickly validate this idea”. And this is where we did the Reddit test and we got some traction. With that, we went into the interview with the certainty that this problem was real.
Massimo: Let me clarify, still no product.
Abel: Zero, no product, yeah. Just a landing page, some UI mocks. Which I think is not typical among startups applying to YC. I think the majority of YC companies have some traction, at least from what I saw in the batch, some companies even had $1M in ARR.
Massimo: Yes, it depends, sometimes we find companies that just started and companies having great ARR already, there is no rule from what we saw.
Abel: The interview was also very peculiar. We knew all the numbers and we had really precise and complete answers to their rapid-fire questions. Questions like “Number of apps that are developed on average in the world? The median cost? Why Flutter? How is that growing? What is the ecosystem like? Why are you the ones who are going to solve this problem?”
Then on the traction part, we could show them the interest we collected. At the end, they asked “Can you show us a prototype?” and we said, “We haven’t started building it (laughs). But we can get you a prototype soon”. So they didn’t accept us actually there, at the interview. They gave us 4 weeks to do a prototype. So November was basically
“No-sleep November”. We just worked non-stop in my apartment, me and my co-founder in an empty room most of the day, just working every day. We got the prototype ready and we showed it to Jared Friedman, who’s one of the partners. He was very impressed that we went from nothing to the early version of FlutterFlow. It was still very early, but it worked. And then we got accepted.
Massimo: Ok, that I think is atypical, it’s not the usual way to get accepted to YC, but it shows that you can be part of a YC batch in many different ways. So the message here is to not give up, to push hard on your goals, and try hard to be in that small group of people that will be part of a YC batch.
Then what happened when the batch started?
Abel: As soon as we got accepted we got into the system. We got an onboarding email and we got access to Bookface. Bookface is the internal system that they have for the founders to communicate among each other and with all the alumni. It’s very cool, it’s like an internal forum. At that point, we were incorporating and I looked up some of the previous threads on that topic on Bookface: I found posts from 2009 written by the Heroku founder that was very detailed and were answering my questions. It was very cool to go back and see comments from former YC founders asking and answering each other’s questions.
Before the batch actually started we went through a lot of this material and we were introduced virtually to the other companies forming the batch. I feel this is the largest batch they ever had (350 companies).
The batch started in early January, the first week was mostly content: enterprise sales, product analytics, development iterations, a lot of different talks from former founders, from YC partners. So mostly informational. I think it’s the first time they did it this way, typically, during the in-person program they did a live meeting, kind of a conference. This time it was all recorded, and I really loved that: there are some topics they talked about that weren’t really relevant for us at the beginning (in January), but that are relevant now (e.g. enterprise sales), so now I can go back and look at that content.
Massimo: So, this part is not a proper “training”, is just a set of talks, correct? Founders, investors, professionals giving you their experience. Also, were these short talks, like 1 hour, or longer like half a day?
Abel: Yes, slides and talks, exactly. It was a 3-day Bootcamp (Wednesday to Friday) with multiple sessions a day, from 9 am to 2 pm. I have to say, I feel like a lot of this kind of content is already out there somewhere in the world (like YouTube videos or blog posts), so there wasn’t anything very new there. But it’s structured and it’s easier to learn in that kind of structured environment.
Massimo: Was it pre-recorded or was it interactive?
Abel: It was interactive. People were asking questions at the end. Typically it’s a 1-hour talk and then 15-30 minutes for questions.
Massimo: Small groups or all the batch together?
Abel: They were giving 3 to 5 talks at the same time, kind of like a real conference where you choose the topics you prefer to follow, and you just join whichever you prefer. And then all are recorded and made available.
Massimo: Ok. After that you got back to work I suppose.
Abel: Yes. So because it was remote, one of the biggest parts of the experience was the Slack group. There was a W21 Slack Batch group where all of the companies were, and then the batch was divided into 4 groups, with 90 companies on average per group. Each sub-group had about 12 Sections, so we ended with about 8-10 companies per Section that were kind of similar (dev tools, consumer, etc…). So you have Batch > Group > Sections.
In your Slack, you had the whole Batch channels with random questions, or announcements, or engineering questions. And then you had the Group level, the Session level, and also a private channel where, for FlutterFlow, we were only me, my cofounder Alex and the YC partners working with us (Jared Friedman, Surbhi Sarna, and Tim Brady). So the 5 of us had our private Slack channel where we could go and just ask them “Hey partners, this happened, what should we do?” and then just brainstorm quickly with them.
Outside of the Slack experience, week after week, there were 3 main things. Section meetings, Group meetings, and Batch meetings.
The Section level meeting was every Monday for 12 weeks. We met weekly with the 9 companies in our Section for about 1 hour. Half the time was led by one of the YC partners to talk about something specific based on a theme, and then the rest of the time was founder-led, just sharing experiences. Every week there was something specific we were all experiencing: for example one week we were talking about customer acquisition and there also were some intro opportunities among the companies. This way you get to really really know these people. We met with them 1 hour a week for 12 weeks, we know them well and we still keep in touch.
With the Group level and with the Batch level there were weekly 1-1.30 hour talks. The Group level talks are typically done by our YC partners, sharing their stories and their experiences.
The Batch level talks are very nice, for example, the Airbnb founders came and shared their story, or also the GitLab founders' talk was very interesting. These weren’t recorded and a lot of it is off the records. They’re very open and they’re sharing all the experiences they went through, it’s really cool.
Massimo: Apart from these meetings what else did you get?
Abel: One of the things that are really nice are the office hours, which during the batch you are recommended to do weekly. You are supposed to go there with a specific problem. The first office hour we had was after we just launched our very crappy MVP, retention wasn’t good, not surprising, it was very early, but we had a hard time talking to our users. So we went into the office hour asking, “What should we do to talk to our users more?”. And from then on, every week you are supposed to bring in the “next problem.” Going in, you are supposed to know your numbers, your metrics that matter, your KPIs. So you go through the number, the problem, and they help you to unblock the situation.
Another thing that is really nice is the Bookface launch. It’s like a post on Hacker News but only for the YC community (you can choose to launch just for your Batch or for the whole community), and you can launch as many times as you want. We just launched for the Batch first to get some users, and that way, we got 8 or 9 users, and from them, we got some feedback. And then, we did the same for the whole YC level, which is a great way to sell to other companies. We were still a bit early; we started building it in October-November, so by January or February, building an app through our platform for a mature YC company was hard, but we still got 20-30 sign-ups, and we got a lot of helpful feedback.
This is kind of the process before the Demo Day: launch on Bookface, launch on Product Hunt, launch to Press if you’d like (they help you to take your story and put it on Techcrunch), and then launch on Hacker News (which we held, to save it for the future).
Massimo: Did you have the chance to talk to alumni from the YC network.
Abel: Yeah, so that’s actually another thing you get. I reached out to several folks, like maybe up to 10 people through the network, and some companies reached out to us. And people are very open to answering questions, they are willing to help, which was very nice.
Massimo: So the purpose was feedback, getting new customers, partnerships?
Abel: On my side, reaching out to people was mainly for relationship building. For example, the founders of Supabase, this open-source Firebase, reached out to us for a collaboration, and it was really cool. They saw our launch on Bookface, and they thought it was a good opportunity to integrate with a new company. So that’s how we met them, and then they introduced us to some angels who invested in their company, like Ben Tossel, the founder of Makerpad, who invested in us through an introduction from Paul Copplestone, the CEO of Supabase. So these kinds of things are supervaluable.
Massimo: Yeah, we tried to invest in Supabase, but it was oversubscribed! Also, you have thousands of alumni on Bookface, I suppose.
Abel: Yeah, there’s over 5000, I think. I feel like the best thing from YC - among all the great things that they do - is really maximizing your chances at seed round, even just because the “YC halo” is a very real effect (if a company is in YC, investors are just going to be more interested). 70% of the YC companies close their seed rounds fast. Some companies in our Section made this mistake of not compacting their fundraising: the batch companies are supposed to not make any investors kind of relations early in the batch; they are supposed to focus on their products and on their metrics and prove everything until a few weeks before demo day. And what that does it that they will try to show the best possible kind of result to investors during meetings, and planning meetings 2 to 3 weeks before you close, and not more than that, creates a time-bubble effect that is really effective, shifting leverage a little bit towards founders.
If you like this article, share it with your friends.
FlutterFlow CEO talks about his experience with Y Combinator W21 batch