How to get your first customers: find people who are publicly struggling and help them first

I'm Mateusz. I build IndiePilot on my own. For a long time it had signups and almost no revenue. The first customer who actually paid came from a playbook that sounds too simple to write up. I'm writing it up anyway, because when you're at zero you usually do the opposite.

Here's the short version: find people who say in public that they have the problem you solve, do something useful for them before you ask for anything, and then follow up like a human.

What didn't work first

It's worth seeing what failed, because these are the defaults most of us start with.

• Describing my product to strangers. Cold emails that led with features got silence. A stranger has no reason to care about your feature list.
• Waiting for free users to upgrade. People signed up, poked around and left. A free plan isn't a funnel if nobody walks people through it.
• Thinking about ads before attribution. I looked into paid ads and decided not to spend. With a tiny budget and no way to tell which channel produced a signup, I'd have been paying blind.

Step 1: Find people who are struggling in public

Founders, freelancers and small-business owners describe their problems out loud every day. A few places to look:

• X / Bluesky: search phrases like "0 paying customers", "launched and nobody signed up", "marketing is the hard part", "how do I get my first users". Use your own problem's wording, not your product category.
• Indie Hackers: look at posts where people share their stuck numbers ("25 signups, 0 paid").
• Reddit: "how do you handle X?" or "is there a tool for X?" threads in the subreddits where your buyers are.

You're not hunting for leads to spam. You're looking for people who have already told you, in their own words, what they need. That's the warmest kind of cold outreach there is.

Filter hard. Skip anyone where your product is a stretch. Skip people who've said they aren't looking for help. A short list of people who clearly fit beats a long list of maybes.

Step 2: Do the useful thing before you ask

This is the part that changed my results. Before writing anything, I spend a few minutes doing real work for that person.

In my case, that means finding the Reddit threads where their customers are asking for what their product does. So the email isn't "here's my tool". It's "people in this community keep asking for exactly what you built. Here's the kind of thread I mean."

Change the specifics to fit your product:

• Sell a design tool? Point out one concrete fix on their landing page.
• Sell analytics? Show them one thing you noticed in their public data.
• Sell a hiring tool? Point them to a community where their candidates hang out.

The test: the message should be useful even if they never reply or sign up. If it is, it doesn't read like spam, because it isn't.

Step 3: Keep the message short and the ask tiny

A structure that works for me:

1. One line showing you actually read what they posted (their words, not flattery).
2. The useful thing from step 2.
3. One or two sentences about what you built and how it relates.
4. A tiny ask, like "reply 'go' and I'll set it up for you", plus an easy way out: "Not relevant? Reply 'no' and I won't email again."

No links in the first message if you can avoid them. No made-up numbers. No "quick call?" Respect the no, every time.

Step 4: Follow up by hand, especially with free users

This is the half most founders skip.

The customer I mentioned, a founder who found us through a cold email like the one above, signed up for the free plan and didn't upgrade. What came next was a short, personal, plain-text follow-up: who I am, what I'd noticed about their product, and an offer to set up their workspace myself. After that, they became a paying customer.

I don't have proper source tracking yet, so I can't prove which touch made the difference. But the pattern is cheap to repeat: a useful first message → free signup → a personal follow-up a little later. I now check new free users every week and write to them myself.

A one-line "want me to set this up for you?" from the founder beats any automated onboarding sequence I've written.

Step 5: Turn it into a weekly routine

• Monday: collect 10–15 people who publicly described the problem this week.
• Tue–Thu: do the useful thing for each one and send a short message. Quality over volume.
• Friday: go through new free signups and send each a personal note with a concrete offer to help.
• Always: write down which channel each signup came from. Set up attribution before you spend money on anything.

Takeaways

• People who are struggling in public have already done your customer research for you.
• Lead with something specific and verifiable about their business.
• Make the message useful even if they never answer.
• Your free plan needs a human follow-up.
• Don't automate the part that has to feel personal.

Disclosure: I built IndiePilot, which does the "find the threads where your customers ask for help" part of step 2 automatically. It watches subreddits and keywords, scores posts by buying intent and drafts a reply that you edit and post yourself.

If you'd like to try it, there's a free plan at indiepilot.app with no card required. Reply to the welcome email and I'll set up your workspace myself.

Reader code READER30: 30% off your first 3 months (or READER30Y: 30% off your first year)

Related reading

• How to find your first customers on Reddit without getting banned
• How to write Reddit replies that don't read as self-promotion
• How to find customers on Reddit automatically