YC W23 Week 1: Go smaller, not bigger

You will often be asked “How big can this get”. But to go big, it may be wise to first get really small.

Intro

This is the first of a series of posts to document the journey through YC W23. For those not familiar, YCombinator is a startup accelerator based in SF that has an incredible track record of success and as a result, one of the biggest networks in Silicon Valley.

Our Story

My cofounder and I applied to YC with Backprop a couple days before the deadline in September. We got an interview request the week before Thanksgiving and had our interview (and ultimate acceptance) a week later. At some point, I’ll write a more detailed story of how we got into YC but for this post I want to focus on sharing real-time learnings that we’re collecting while on this journey!

The Company

We build tooling to make code review painless.

The problem

Code review for many software developers today is a painful process. Many teams don’t have great code review culture and as a result, the data suggests that upwards of 25-30% of developer time is spent reviewing code.

What is code review?

Code review is the process where teammates review and leave feedback on one another’s code before it is sent to production. This includes confirming that an implementation of a new feature is scalable, that best practices are being followed, or that new bugs aren’t being introduce with the code.

Because codebases for teams can be quite large and complex, code review is a rather scrutinized workflow. Poor code review culture can have lots of downstream consequences including

1. Release delays

2. Production bugs

3. Technical debt build-up

4. Organizational knowledge siloes

Week 1 Learnings

Find a narrow wedge user

It’s common knowledge now that YC’s mantra is “talk to your users”. But as you do this, it’s incredibly important to nail down who your initial user should be. And common temptation is to think “every X person is my customer”. But it’s better to get small.

What type of company experiences these problems the most painfully? Who will be most passionate about your solution? It’s easier to get your first few users if they’re near perfect fits for the product. People who will absolutely love the product. This will help when finding and engaging with growth channels.

Most growth channels are saturated.

This past week, we sent out dozens of cold emails and made some posts on social media. While we received some engagement, it’s not what we were hoping for. We discussed with one of our YC partners, and it became clear that we needed to hyper personalize our reach out.

Your customers are busy, and they have lots of information being thrown in their face. It’s important for them to know why you are reaching out to them specifically. Mild personalization is not enough, you need to be hyper-personalized.

Our initial emails were targeted to engineers who worked out our previous company (big tech), who moved to smaller startups. But turns out that was only mild personalization.

Lean in to some very small overlapping circle that you have in common. Maybe you worked together on the same org at your previous company, or maybe you have a mutual connection, or you’re both members to a niche online community.

Finding the right people is hard and if you start with too broad of an audience, your outreach will be very diluted and your response rate will be low. You will put a lot of effort in and get little out.

Be creative and minimize time to value

You have to catch someone’s attention in as little time as possible. Have a demo ready, maybe even with example data that fits their specific use case. This week we’re going to build a sandbox environment for our product where people can try the product with as little upfront commitment as possible.

Add value wherever you can

You’re either receiving favors or offering value. Offer value wherever you can. What is your edge? Where is your expertise. Again, smaller can be better than bigger. Maybe you can offer someone a useful intro. Or maybe you can offer product feedback. Provide organic value where you can, and you over time increase your surface area of opportunities for favors. Same goes for customers. Understand their problems deeply and try to offer value. Same goes for social media. What can you contribute to the conversation.

Run small experiments

It’s tempting when building a new product to let your imagination run wild. You visualize the end state of your product over and over, and you want to move as quickly to that vision as possible. This is the mindset of the artist. Instead, have a mindset of a scientist.

1. Do a deep literature review. Understand how this problem statement started and how it has evolved. Understand the solutions that have been tried. Become an expert. Research.

2. Develop hypotheses and run small experiments. Think of a new audience to target? Try reaching out to 10 people in that audience to see if they’re willing to have a conversation. Need better landing page conversion? Try adding a video demo. See what happens. Iterate not only on your product, but on the growth channels too.

3. Interpret data. Does this data prove the hypothesis wrong? Then try a different one. Does it show good results? Scale it 3x and see if it holds up.

4. Don’t get emotionally attached to your ideas. Test them, get data and be okay with being wrong. Being emotional will guard you from correctly interpreting your data.

That’s all for this week. If you want to follow our journey, subscribe to the newsletter or follow me on twitter @eagrwl!