AI copilot for your own research, notes & conversations

AI copilot for your own research, notes & conversations

It’s great to be back, PH fam! I’m Sam, co-founder of Heyday.

My co-founder Samiur and I will be answering your questions for the next 24 hours. Hit us with anything!

As @Samiur1204 mentioned, our decision to focus on executive coaches unlocked a level of traction we hadn’t hit when we first launched Heyday. And we think it will be the reason why we ultimately achieve our goal of building the AI copilot for all knowledge workers.

But like being told to “talk to your users,” being told to “focus on a specific niche” is not specific enough advice to be helpful.

Below, I’ve shared the process we ran to choose a niche. I hope it helps you whether you’re building a startup or hustling on a side project!

👉 Create a criteria of persona we want to serve

Does this person have a hair-on-fire problem? Is it ongoing? Would they pay to solve their problem? Do they have an urgent need to switch to us? Are there a lot of these people or are they very well connected with other groups?

We knew we’d want to be able to answer “yes”, to any persona we selected.

👉 Investigate existing user base for “weird” behaviors

We looked into our database to see who was using Heyday in extreme ways. This pointed us to folks who were motivated to solve a problem, many of whom had a profession that we didn’t think of as one who might need our help.

👉 Interview folks to learn about their problems

With quantitative usage data and no additional context, we wouldn't have understood the problems folks were trying to solve. Our conversations shed light on this.

👉 Select a few personas to run an initial experiment

After learning about the problems of ~50 people and whether they met our criteria, we had a shortlist of personas to work with - Coaches, Equities Investors, Consultancy CEOs, and Writers.

👉 Manually super-serve people in the target persona

We chose 3-5 people from each person and started working for them as consultants. Before we wrote a single line of code, we wanted to be sure we understood their problem deeply and identified a solution that would make them ecstatically happy when delivered.

👉 Identify 1 user persona that we are making ecstatic

What does ecstatic look like? We looked for two signs: 1. They spontaneously told friends how we were helping them 2. They said they would happily pay $100/month for the service we provided. Coaches were ecstatic about how we were helping them.

👉 Build tools for ourselves to serve 10 times the number of users

By this point, we were confident that we understood the problem and desired solutions for this small group of coaches. But with such a small group, we had to be sure that a larger group would feel similarly excited.

Instead of trying to nail the product experience at this point (costs time + $$), we built internal tools for ourselves and consulted 10x as many coaches.

👉 Incorporate the features into the product

When 60% of coaches in this larger group agreed to pay us $100/mo, we knew we were ready to flesh out our product and bring more coaches on board.

That’s our process for choosing a specific persona.

Let me know if you have any questions about this process or Heyday in the comments below!