Do paid ads work for early stage products?

Generally not. Paid acquisition amplifies a funnel that already converts. Before roughly 100 engaged users you do not know whether your funnel converts, so ad spend mostly buys noise.

Should I build more features if nobody signs up?

Almost never. Low signups usually indicate a positioning or description problem rather than a missing feature. Rewrite the description and talk to people before building.

Why should I talk to every early user?

Because that conversation is the actual value of the first hundred, not the number. Early users tell you what they were trying to do, where it broke, and whether you built the right thing.

How narrow should my product be at the start?

Narrow enough to describe in one sentence. A tool that completely solves one problem for a specific group spreads by word of mouth far more easily than a general platform.

Getting Your First 100 Users for a Design Tool

DDezignhunt Editorial7 min readLaunch & Growth

Last updated: August 2026

The advice that works at 10,000 users is useless at zero. Paid acquisition, SEO and content marketing all need something you don't have yet — budget, authority or an audience.

The first hundred come from manual effort, and that's not a temporary inconvenience. It's how you find out what you've built.

The best new design tools, once a week.

One short email. No noise, unsubscribe anytime.

Key takeaways

Talk to every one of the first 100 individually — that is the actual point
Communities beat launches: sustained presence outperforms one spike
Give it away to people whose feedback you want
Solve one narrow problem completely rather than many partially
SEO and paid ads do not work at this stage; stop trying

Where the first 100 actually come from

People you already know. Designers you've worked with, former colleagues, people who follow you. Least scalable, highest conversion, and the fastest route to honest feedback.

Communities you're already in. Design Discords, subreddits, Slack workspaces, Figma community spaces. The rule is participation first, promotion much later — six weeks of being genuinely useful earns a share that would otherwise get you removed.

Direct outreach. Find twenty designers who have the exact problem you solve. Message them individually, referencing something specific about their work. Offer it free. Ask for fifteen minutes.

This feels uncomfortable and works better than anything else on this page.

One good launch. Product Hunt, Show HN, Peerlist. Worth doing, produces a spike, does not sustain. See where to launch.

Niche directories. Small ongoing referral traffic that keeps arriving long after launch day. Submit to DezignHunt and the others in one week of work.

Building in public. Sharing the process attracts people who become users because they watched it get made. Slow, compounding, and it works.

What does not work at this stage

Browse 500+ design tools

Heads up

Every founder tries paid ads at 20 users, spends a few hundred pounds, gets three signups who never return, and concludes marketing does not work. Paid acquisition needs a converting funnel to amplify. You do not have one yet.

SEO. It works and it takes twelve months. Start it now, expect nothing from it now.

Cold email at volume. Low response, high annoyance, potential domain damage.

Influencer outreach. They want an audience or a fee. You have neither yet.

More features. The instinct when nobody signs up is to build more. It is almost always wrong. Nobody is refusing to use your tool because it lacks a sixth feature.

Talk to all of them

This is the real point of the first hundred, not the number itself.

Email every signup personally. Not automated. Ask what they were trying to do and whether it worked.

Watch people use it. {T('hotjar')} Hotjar session recordings show where people stall. Ten recordings usually surface something no survey would.

Ask the churned ones. The people who signed up and never returned know exactly what's wrong, and most will tell you if asked directly and briefly.

Record what you learn. {T('dovetail')} Dovetail keeps findings searchable so the same problem doesn't get rediscovered in six months.

Narrow, then narrower

The most common failure at this stage is being useful to many people slightly rather than a few people completely.

A tool that perfectly solves one annoying problem for freelance designers who build client sites will get its first hundred faster than a general design platform. The narrow thing is describable in one sentence, shareable in one message, and memorable — which is how word of mouth actually functions.

You can broaden later. Nobody has ever succeeded by starting broad.

What 100 users actually tells you

They came and stayed → you found something. Do more of what got them here.
They came and left → the product is wrong, not the marketing. More traffic makes this worse.
They didn't come → nobody understood the description. Rewrite it before building anything.

Frequently asked questions

Through manual, unscalable effort: people you already know, communities where you are an existing participant, individual direct outreach, one good launch, and niche directory submissions.