The meeting was fine. The founder was sharp, the market was real, and the deck had a customer retention chart that looked healthy at first pass. I asked one question on the way out, standing up, coats already on. She didn't hesitate exactly. But she restated the question before answering it. That's the tell.
The Product Market Fit (PMF) conversation in early-stage investing has become a ritual without meaning. Founders have learned what angels want to see so they produce it. Net promoter scores, cohort charts, Letter of Intent from enterprise logos, testimonials from customers who sound delighted. These are not lies, exactly. They are the outputs of effort. Founder effort, specifically. The problem is that angels have started treating evidence of effort as evidence of gravity and they are not the same thing.
Gravity means the product pulls customers back without being pushed.
Everything else is a founder in a hamster wheel, moving fast enough that it looks like momentum.
The question
What percentage of your paying customers came back and paid again without any further involvement from you?
Not net revenue retention or renewal rate. Not “we have strong relationships with our customers.” but unprompted return. Paying customers who came back because the product created a reason to, not because a founder or a CSM or an automated drip sequence created a reason to.
If the founder hesitates then you have your answer.
If they qualify the question then you also have your answer.
If they say “that’s a really interesting framing” and then reframe it, again you have your answer.
The discomfort in the room when you ask this is itself a data point. Founders who have real pull know this number because they are proud of it. They know it because it surprised them. They noticed customers coming back before they built the infrastructure to manage it and so the question lands easily for them just as it lands hard for everyone else.
Three buckets
Real pull looks like this: a high unprompted return rate, and a founder who can name the trigger. Not “they like the product” but specific mechanism. What happens in the product, or in the customer’s world, that creates the next purchase without a push. Founders with this answer are usually slightly bored by the question as they expected you to already know.
Founder-dependent retention is different. The numbers are there. Renewal rates are acceptable. Churn is low. But when you trace it, every retained customer has a founder touchpoint in the history. Every renewal had a check-in. Every at-risk account got a personal call. This is a real business, not a venture business. The economics of scaling a founder-dependent retention model are brutal and angels systematically misprice the transition risk.
PMF theater is the third bucket and the most common one at seed. Retention numbers exist because churn is being hidden, delayed, or manually prevented. Cohorts look healthy because the unhealthy ones have been quietly excluded, the time window is short, or the definition of “active” has been adjusted. I have seen this enough times that I now ask founders to define “active customer” before I look at any retention chart. The pause before the answer is instructive.
The investor mistake worth naming
Angels confuse founder quality with product quality and often fund the wrong thing.
A strong founder who is personally closing customers, personally managing relationships, and personally rescuing churn is demonstrating something real. Their ability to sell, their commitment to the company and their capacity to hold things together in early chaos. These are not small things as they matter for survival.
But survival is not the same as scalability. And angel returns do not come from survival.
The mistake I made, a while back, on a SaaS business I backed at seed: the founder was extraordinary. He was relentless and every customer loved him. The NPS was genuine but what I hadn’t seen was him that the customers loved. Yes the product was fine but, on it’s own, it was not pulling anyone back, it was manual effort of the founder that was responsible. At Series A, the investors wanted to see the product work without him in every account. It couldn’t and the round didn’t close anywhere near the terms we needed.
I knew something was off in the early numbers. I funded the founder instead of interrogating the model. That is a defensible call but it is not a repeatable one.
The modified question that makes the filter more credible
Some businesses have genuinely long or infrequent purchase cycles. Enterprise software with annual contracts, high-consideration consumer categories, certain B2B services etc. Asking for unprompted return rate in these cases produces a number that is structurally low regardless of product strength.
The modification is simple. For long-cycle businesses, the question becomes: what is the unprompted engagement rate between purchase events? What does the customer do with the product when no one is watching, when nothing is due and when there is no renewal conversation on the calendar?
If the answer is nothing, then the product is not part of the customer’s life between purchases. That is a retention risk that will not show up until the renewal. By then you are already inside the problem.
Acknowledging this modification matters. It makes the question more accurate and it also signals to founders that you have thought about this carefully enough to ask the right version.
What angels avoid concluding
The uncomfortable conclusion most angels do not say out loud: a large percentage of seed-funded PMF is not PMF but FMF (founder-market fit) combined with enough early customer goodwill to produce metrics that clear the bar. The product may be fine, the founder is exceptional but the pull is not there.
This could still be fundable at seed. The bet is that the product catches up to the founder before the founder burns out carrying it.
Sometimes that bet pays but more often it doesn’t and the potential Series A investors are going to find out either way.
The question is whether you price for that risk at seed, or pretend it isn’t there because the founder is compelling and the deck is clean.
Most angels pretend - it’s easier.
Angels should ask the question in your next meeting. Ask it plainly, without softening it. Watch what happens in the two seconds before the answer starts.
Founders, don’t even ask for a meeting until you can instantly answer that question with pride without a gap.
The hesitation is where the real information lives. Not in the answer as much as in the gap. The answer can be prepared. The gap usually cannot.
PMF is not a moment but a behaviour pattern that exists without the founder in the room. If the founder has to be in every room for the pattern to hold, you are not investing in a product, you are investing in a person who is one bad quarter away from exhaustion.
Some of those investments work out but you should know which kind you are making.
If you found this article useful, consider joining our growing community. There you will find more detailed advice, whether you’re an angel or a founder keen to get an advantage by looking over angels’ shoulders.
My DMs are open to paid subscribers only and you get access to my private contact book of over 150 family office and HNWs that I deal with on a regular basis. You’ll also get a free copy of my new book series Angels & Unicorns (RRP $29.99) and access to our private chat of Angel Investors.
