Anton Reed

I’m building a SaaS product to measure product-market fit and customer expectations, and documenting the process in public. Real decisions, real mistakes, real trade-offs. From zero to PMF to growth. No theory. No hype. Just building and learning in the open.

Your overall PMF score is lying to you. 35% overall could mean 60% among power users and 12% among casual signups. One number hides two completely different products. Segment or stay confused.

Rahul Vohra open-sourced the PMF engine Superhuman used to go from 33% to 58%. Survey. Segment. Analyze. Iterate. Track. The method is free. Setting it up isn't. I spent weeks stitching together surveys, spreadsheets, and custom filters before it actually worked. That's why I built FitSignal so you can skip the setup and just run the method.

Growth before PMF is the most expensive mistake in SaaS. Every dollar spent on acquisition before 40% "very disappointed" is a dollar lighting the churn fire. Fix the product first. Then step on the gas.

Good framing. The next layer: it's not just what would happen at the product level — it's what would happen for specific user types. Casual users barely notice. Power users scramble to replace you. Ask the "disappeared" question. Then cut it by segment. The gap between those two groups is where the real PMF signal lives.

A simple way to think about product-market fit: If your product disappeared tomorrow, what would actually happen? Not what people say would happen, but what would materially change. Would anything slow down? Would anything break? Would someone go out of their way to replace it?

Retention is PMF measured over time. If 40% of users say "very disappointed" today but only 20% are still active in 90 days, you don't have PMF. You have a first impression. Track both. The gap between them is your real problem.

@goreste Portfolio approach only works if you kill the losers fast. Most solo builders hold product #1 too long because "some users" feels like progress. Run the Sean Ellis survey on each product. Below 40%? Pivot or move to the next one.

39% of indie saas founders are solo builders hitting $5k-50k mrr the portfolio approach works: solve your own problem, charge others for the solution, repeat with different niches most people quit after product 1 because they confuse users with revenue

@bowale4you One question cuts through it faster than any audit: "How would you feel if this product disappeared?" Below 40% "very disappointed" and you have a PMF problem, not an ads problem.

Stop running ads if your ads stop converting and check your metrics.. It could be a product-market fit issue, creatives or other things. Take time out to do a survey amongst your existing customers and discover the cause of your issue. Consumer wants and mannerisms change

Agree that surveys beat spreadsheets for PMF. But "listening at scale" only works if you ask the right question. Most survey tools optimize for NPS. The PMF question is different: "How would you feel if this disappeared?" And the real unlock is segmenting responses by user cohort. Your power users and casual signups live in different worlds. One number hides that.

Product-market fit isn't found in a spreadsheet. It's found in what your customers tell you. Surveys are how you listen at scale.

Half agree. Behavior shows what people do. It does not show why they stay. I've seen products with solid usage and terrible PMF scores. Switching costs held users, not love. One question fills the gap: "How disappointed would you be without this?" Behavior plus that answer gives the full picture.

@reddit_lies Buyers are liars Observed behavior is the only feedback that matters You think users want X like AI to cure cancer You ship They show you they want Y Stop trying to predict product market fit Ship a usable product Watch what people do The product is the survey

This is straight out of a horror movie.

Retention is the lagging indicator. The leading one is the Sean Ellis survey - "how would you feel if you could no longer use this product?" If 40%+ say "very disappointed," retention usually follows. The trick is running it early enough to course-correct before churn shows up in your MRR.

Want a real measure of product-market fit? Look at retention. If customers who bought a year ago are still paying, you’re close. If they’re churning, you’ve got work to do. #SaaS #Startups #Metrics

Your activation rate is the leading indicator of your PMF score. If users can't reach the "aha moment" in their first session, they'll never say "very disappointed" on a survey. Fix activation first. Then measure PMF.

Pattern I keep seeing in indie SaaS: Build for 6 months. Ship. Wait. Nobody comes. The missing step: validating demand before building. A PMF survey on an MVP with 20 users would have saved 4 of those months.

The hardest part of measuring PMF is not the survey. It's the segmentation. Your overall score hides everything. Power users vs casual users. Paid vs free. Organic vs paid acquisition. That's why I built FitSignal to segment automatically. One less excuse not to measure.

Good list. The pattern I keep seeing across solo SaaS winners: they don't just validate demand, they validate who has the pain acutely enough to pay immediately. Same idea, different segment = wildly different PMF score. "Customer feedback tools" as a category has weak PMF, but "PMF survey tool for bootstrapped founders who can't afford Survicate" is a specific enough segment to actually measure.

🚀 Just wrapped up my deep survey on solo-founder SaaS! Researched real, validated categories/ideas that one person can actually build — and they’re already printing serious MRR: 1. Creator stores for selling digital products, courses & bookings 2. No-code tools for digital product ecommerce & sales 3. AI-powered all-in-one social media management tools 4. AI ecommerce/dropshipping store builders 5. Health & product safety analysis apps 6. AI tools for viral faceless short videos 7. AI image & generative art platforms 8. Advanced link-in-bio pages 9. X/Twitter-specific content & growth platforms 10. Mobile AI art generators

Exit surveys are underrated for the data they give back upfront, not just at cancel. I've found the same 2-3 objections that cause cancellations usually show up in PMF survey open-ends months earlier — people who say "somewhat disappointed" and give a reason are telling you exactly what would make them "very disappointed" instead. Treat cancel reason research and PMF surveys as two ends of the same signal.

📊 Exit surveys to capture cancel reasons (Growth) No code. No webhooks to configure. Just OAuth and go.

I kept watching MRR leak in slow motion. Failed payments. Cancel clicks. Surprise chargebacks. So I built something to stop it automatically. 🧵

That 95% distrust stat is basically the Sean Ellis "very disappointed" question in proxy form — 95% of devs saying they'd be genuinely upset if unsupervised AI code shipped. Qodo didn't invent the pain, they just measured it and built for it. That's the job. Strong need signal, right move to raise on it.

Software safety is now product-market fit - Qodo raised $70M to verify AI-generated code; survey: 95% of devs distrust AI code, only 48% review it. So verification and governance (plus bounties) are the pragmatic response — someone still has to grade the robot’s homework.

Europe just bought 13,800 Nvidia GPUs and turned them into a political strategy - Mistral’s $830M debt deal for a Paris data center is the single biggest signal yet that compute sovereignty is now a battlefield. 🧵 Here's what happened...

The super-customer survey group framing is underrated. Most founders survey their whole user base and get a 35% PMF score that hides everything. Break it down — power users vs. casual signups vs. trial abandonments — and you often find 65%+ among your best users and 15% among the rest. Two completely different products in their minds. That breakdown is where the actual roadmap lives.

How do you find strong hero products? > A strong PMF process > Focus heavily on Product / Solution > Speak to your customers consistently > Create a strong group of super-customers to survey and chat with > Be really aware of your niche I have another threads about how to do PMF properly

It's not an ad problem, you need more hero products If you want to do 10M in revenue, you need 7-10 hero products It’s rare that brands can scale a product infinitely 🧵

That's what PMF looks like after you have it. The harder question is how you know you're trending toward it before those signals show up. The Sean Ellis "very disappointed" survey gives you a leading number. Not perfect, but better than waiting for word-of-mouth to kick in and realizing you've been wrong for six months.

Product-market fit (PMF) is the moment when your product is so good for a specific group of people that it basically sells itself. Users sign up faster than you can keep up. Word of mouth spreads without you pushing it. People would be genuinely upset if your app disappeared.

Procurement docs are a great lagging PMF signal — you've already got it by then. The leading question is harder: what tells you it's coming six months out? For us that's the Sean Ellis score by segment. When power users hit 60%+ "very disappointed," the procurement docs follow. Eventually.

You can tell who has product market fit by watching for suddenly fully fleshed out procurement documents On that note, announcing paper.design/legal Let those procurement teams go wild ✨

Hot take: most founders measure PMF once, get a number they like, and never measure again. PMF is not a checkbox. It's a trend line. You can have it in Q1 and lose it by Q3.