• IdeaHub
  • Posts
  • 🧠 How to find out what features to build next

🧠 How to find out what features to build next

How to talk to customers, who to ignore, what questions to ask and a mock interview to get you started...

Happy pre-Friday 🎉

Talking to customers is arguably the only way to decide which features to build for your SaaS.

But it can be a minefield.

It’s easy to get lost following advice from the wrong customers and lose sight of your vision.

This week in IdeaHub…

  1. What kind of outreach should you do? Not all feedback is created equally, here’s what actually works.

  2. Which customers should you listen to? How to prioritise customers based on ‘opinion value’.

  3. What questions should you ask?: How to ensure you get reliable feedback without ‘leading the witness’.

Let’s jump in… 👇

(Before we do, if you’re struggling to get your first users, Reddit can be your goldmine. This guide lays out how to use Reddit to find SaaS customers - it’s worth a look…)


🧠 The Newsletter For SaaS Founders

Every week discover the best growth, marketing, tech and product development strategies for your SaaS.

Join 1,000+ SaaS founders, indie hackers and tech entrepreneurs.

1. What kind of outreach should you do?

There are lots of ways to collect feedback from customers.

But surveys, emails, and phone/video calls are most commonly used. Each sits somewhere on this scale:

Types of usr outreach

Outreach strategies

At one end, surveys are easy to set up, involve little management and give you data that’s simple to analyze.

However, because of this, the data quality is limited by the assumptions you made when building the survey.

Your customers have little opportunity to express their true thoughts and feelings as they can only answer the questions you give them.

Phone and video calls are the opposite of this.

It takes much longer to collect the data as you’re interviewing one customer at a time, and the results require more interpretation.

However, the data quality is far better than in any survey. You’re getting unfiltered expressions of customer’s true pain points.

My advice? Take the time and do as many phone/video calls as you can and make sure you record every second.

Plus if you're short on time, you can backfill with automated interview tools like AskMore.


2. Which customers should you listen to?

The reason why not all feedback is created equally is because the people producing it (you’re customers…) are also, not all equal.

Not all users are equal

‘Lurkers’ who are happily using your free trial once a month aren’t going to give the most reliable feedback.

But when asked, everyone has an opinion. And whether they think it’s valuable or not, they’ll happily tell you anyway…

So you need a strategy to decide who to listen to, and who to ‘politely ignore’ to make sure you don’t muddy the waters.

Your ‘power users’ are the ones you should spend the most time talking to. They are likely those paying the most to use your product.

Ideally, you will know their names and have an open relationship with them where issues and feedback can flow back and forth.

You should schedule regular calls with them to go through their thoughts in more detail. As their business changes, so will their requirements.

The rest of your customer base probably sits within the ‘light users’ category, and the level of communication you have with them should follow suit.

3. What questions should you ask?

First of all, 3 questions you shouldn’t ask:

  1. “How would you improve the product?”

  2. “Would you recommend it to a friend/colleague?”

  3. “Do you think the product is simple to use?”

These might sound like good questions to ask but just think about it:

How many people would say to a founder’s face that they would not recommend their product, and it looks as bad as Windows 94?

It would come off as a slap in the face, so they’ll likely just say yes anyway.

But that’s not helpful to you, you want to get their real thoughts and feelings.

Better questions are:

  1. How does the product help you solve [problem]?

  2. How did you solve [problem] before using the product?

  3. Why is solving [problem] important to you?

These questions get under the skin of exactly what customers are doing on the platform, and a response will naturally throw up gems you can follow up on.

To demonstrate this, I told ChatGPT to act as a fake business owner and conducted an interview with them.

Naturally, it started to wander off towards the end, so I cut it short! 🙃

But it is representative of how I would go about conducting an interview.

After you’ve done this with a sample of your customers, you’ll hopefully start to notice patterns in what they need.

This is where you can start manifesting ideas for new features.


👉 You’ll want to see this…

Interview Tips from YCombinator - A partner at YC shares his thoughts on interviews when validating ideas for new products. 🟧

Automating Product Discovery - Maze helps you manage the outreach process with the help of AI. 🤖

How He Grew His SaaS to $74K MRR - Tim B shares his story on how he built Pally into a classic indie hacker success. 📈

See you next week 👋

You better bring your costume! 🧙‍♀️

- Charlie

🧠 The Newsletter For SaaS Founders

Every week discover the best growth, marketing, tech and product development strategies for your SaaS.

Join 1000+ SaaS founders, indie hackers and tech entrepreneurs.

Join the conversation

or to participate.