In B2B, revenue is the last signal of product-market fit to arrive and the easiest one to fake. Real fit shows up earlier, in how buyers behave: they name the problem before you pitch, multi-thread the deal themselves, and expand before you ask.
If you have ever sat in a board update or an investor call, you know the question:
"So, do we have product-market fit?"
And you know the answer everyone reaches for.
Revenue. Logos. A chart that goes up.
Here's the uncomfortable reality: in B2B, revenue is the slowest signal you have, and it is the easiest one to manufacture.
You can close a quarter on founder charisma. You can close it on a 40% discount. You can close it on one unusually motivated champion who will be at a different company in eleven months.
None of that is fit. All of it looks like fit on a slide.
In my last organisation I was the only product marketer, selling HR technology into enterprise People and HR teams. The quarter looked fine. Revenue was in. We had logos we were happy to put on a slide.
Then I did something that changed how I think about this entirely: instead of reading the CRM stages, I went back and read the actual call notes on every won deal from the previous two quarters.
The pattern was not comfortable.
Closed deals? Healthy. Deals that closed because the market wanted this? Far fewer.
Seven of our eleven wins traced back to a single, unusually invested champion. Remove that one person from the account and the deal had no independent reason to exist. We weren't selling to an organisation. We were selling to an individual who happened to work there.
That is not product-market fit. That is relationship-market fit. And it does not survive a reorg.
What product-market fit actually means in B2B
Product-market fit is the point at which a defined segment of the market pulls your product toward them faster than you push it toward them.
The word doing the work in that sentence is pull.
In B2C, pull is easy to see. Usage spikes. People come back daily. The product is the whole relationship.
B2B is structurally different, and this is why so much generic PMF advice fails founders:
The buyer is not the user. The person who signs is rarely the person who logs in.
The decision is collective. Five to eleven people touch an enterprise deal, each with different risk thresholds.
Budget cycles hide the truth. A team can buy because money was allocated in Q4, not because the problem was urgent.
Contracts delay the verdict. An annual contract means a customer who stopped caring in month two still shows as revenue in month eleven.
So in B2B, revenue arrives late, arrives contaminated, and stays on the books long after the fit has gone.
Which means if revenue is your primary read on fit, you are steering with a mirror.
You are not looking for proof that someone bought. You are looking for proof that the market wants this — and those are different questions.
Why revenue is a false positive more often than founders admit
Before the signals, it's worth naming the three patterns that most reliably fake fit. If you recognise one of these in your own numbers, treat your revenue as unproven until the behavioural signals confirm it.
Founder-led sales hides the product's weakness
Early deals closed by a founder are a genuine signal — of the founder.
Founders sell with context, conviction, credibility and the implicit promise that they will personally fix anything that breaks. That is an enormous, non-transferable advantage.
The honest test is not "are we closing?" It's this: when someone who is not the founder runs the same deal, does it still close at a similar rate?
If win rates collapse the moment a founder steps out of the room, you have founder fit. You don't yet have product fit.
Discounting buys the logo, not the conviction
A deal closed at a steep discount tells you the buyer wanted a bargain, not that they needed the outcome.
Watch what those accounts do next. Heavily discounted customers churn harder, expand less and refer almost never — because the thing that motivated the purchase was price, and price is the one thing a competitor can always beat.
Discounted revenue is evidence of price sensitivity, not of fit.
Pilot budget is not conviction
A $15k pilot out of an innovation budget and a $15k line item out of an operating budget look identical in your bank account. They mean opposite things.
Innovation budget says: this is interesting, let's see. Operating budget says: we cannot run next year without this.
Only the second is a fit signal.
The 9 signals that tell you before revenue does
These are behavioural. Every one of them is observable weeks or months before it shows up in the P&L — and every one of them can be captured with a field in your CRM or a tag on a call recording. That is the point. A signal you cannot record is just a feeling.
Signals in how the deal gets bought
1. Buyers describe the problem in your language before you pitch it.
This is the single most under-rated signal in B2B, and the earliest one available to you.
When a prospect opens a discovery call by articulating the problem using the framing your website uses — unprompted, before you have said anything — the market has adopted your worldview. You are no longer educating. You are confirming.
How to capture it: tag every discovery call as "buyer named the problem first" or "we had to name it." Track the ratio monthly. If it climbs, your narrative is doing the selling before your reps arrive.
2. Sales cycles compress without discounting.
Shorter cycles are only a fit signal when nothing was given away to shorten them.
A cycle that shortens while pricing holds means the buying committee reached internal agreement faster — which means the problem was well understood and the decision was easy to justify.
How to capture it: track median days from qualified opportunity to close, segmented by discount band. If cycles shorten in the zero-discount band, that's fit.
3. Champions multi-thread the deal for you.
I wrote elsewhere that multi-threading is the PMM's job — influencing several personas inside one account instead of nurturing the form-filler.
Here is the version of that which signals fit: you stop having to do it.
When your champion pulls in Finance, Security and their own Head of Department without being asked, they are spending internal political capital on your behalf. Nobody does that for a product they find mildly useful. They do it for a problem that is costing them personally.
How to capture it: count stakeholders added to the deal by the customer versus added by your rep.
4. You win deals where you are not the cheapest option.
Winning on price proves your pricing is attractive. Winning despite price proves your positioning is.
When a buyer chooses you knowing a cheaper alternative exists, they have concluded that the cost of getting this wrong exceeds the saving. That is the clearest commercial evidence that the problem is real and urgent.
How to capture it: a single win/loss field — "were we the lowest-priced option? Y/N" — and the win rate against it.
Signals in what happens after the money lands
5. Onboarding questions change shape.
Listen to the questions in week two.
"How does this work?" → they are still evaluating. "Can it also handle X?" → they have already integrated it into how they plan to operate.
The second question is an adoption signal disguised as a support ticket. It means the customer has moved from assessing your product to assuming it.
How to capture it: have CS tag first-30-day questions as clarifying or extending.
6. Expansion happens before you ask for it.
Net revenue retention above 100% means your existing customers are growing faster than your churn is shrinking them. Across B2B SaaS, 110–120% is considered strong and above 120% is best in class, with the median sitting nearer 100–105%.
But the number matters less than the direction of the conversation. If expansion is something you have to sell, it's an upsell. If expansion is something the customer initiates, it's a fit signal.
How to capture it: tag every expansion as customer-initiated or vendor-initiated. The ratio is your real NRR story.
7. Churn conversations are about budget, not about value.
You will lose customers at any stage. What they say on the way out is the signal.
"We never really got value from it" → a fit problem. "We love it, our budget was cut" → a market or timing problem.
Those require completely different responses, and founders routinely treat the first as if it were the second because it's less painful.
How to capture it: a mandatory single-select churn reason on every closed-lost renewal, owned by someone who isn't the account manager.
Signals in how the market treats you
8. Inbound arrives pre-qualified.
Volume of inbound is a marketing metric. Quality of inbound is a fit metric.
When inbound leads arrive already knowing what you do, already matching your ICP and already having heard about you from a peer, word of mouth is doing work you didn't pay for. Peer recommendation is the most expensive kind of credibility to buy and the cheapest signal to read.
How to capture it: ask one question on the form — "how did you hear about us?" — and track the share answering a colleague or peer.
9. You can say no, and you actually do.
This is the signal founders reach last and the one I'd argue is most definitive.
When you know precisely which accounts you turn away — and you turn them away even in a slow quarter — it means you have a describable ideal customer profile that is backed by evidence rather than hope.
A company with genuine fit has a shape. If you cannot name who you are not for, you have not found the market yet. You have found some customers.
How to capture it: keep a disqualification log. Review it quarterly. If it's empty, that's your answer.
How to turn these into something you actually run
Nine signals is a list. A list is not an operating system.
Here's how to make it one, in the order I'd do it:
Pick the three signals closest to your current bottleneck.
If deals stall, take signals 1, 2 and 3. If churn is the issue, take 5, 6 and 7. Do not try to instrument all nine.
Add the field before you need it.
Every signal above has a capture method that is one CRM field or one call tag. Add them this week. You cannot retrofit a signal you never recorded.
Set a baseline, not a target.
For the first 60 days you are only establishing what normal looks like. Judging yourself against a benchmark you haven't measured yet is how teams talk themselves into false confidence.
Review the direction, not the number.
Fit is a trend, not a threshold. Three signals moving the right way over two quarters is far more meaningful than one good month.
Pair every number with a quote.
The number tells leadership what moved. A verbatim line from a buyer tells them why. You need both for the story to be undeniable.
If you want a quantitative counterweight to all this, the Sean Ellis test is still the cleanest single question available: ask your active users how disappointed they would be if they could no longer use your product. If 40% or more say very disappointed, that's a strong signal. Survey people who finished onboarding and came back at least twice, and get 40–100 responses before you trust the number.
Use it as one input. Not as the verdict.
What to do when the signals say you don't have it yet
This is the part most articles skip, because it isn't the fun answer.
If you run through the nine and recognise almost none of them, the correct response is not to hire more SDRs, buy more intent data or run a bigger campaign. Scaling distribution on top of a problem you haven't validated just buys you more expensive proof that it wasn't working.
The correct response is to go back to demand.
Narrow the segment until you can describe it in a sentence. Find the buying moments that already exist in that segment. Then rebuild the motion around what the signals tell you — not around what the pipeline target requires.
That sequence — demand, then pattern, then motion — is deliberately slow at the start and considerably faster afterwards, because you stop paying for outreach into accounts that were never going to convert.
Product-market fit is not something you declare once and move on from. It is a reading you take repeatedly, in the same way you'd check a compass — because markets shift, buyers change, and the fit you had in 2024 is not automatically the fit you have now.
Final takeaway
Don't wait for revenue to tell you. It arrives late, arrives contaminated and stays on the books after the fit has gone.
Measure behaviour, not just outcomes. Language adoption, unprompted multi-threading and customer-initiated expansion are all observable before the P&L moves.
Instrument before you need it. Every signal here is one CRM field or one call tag. A signal you don't record is just a feeling.
Read the direction, not the threshold. Fit is a trend across quarters, not a number you cross once.
Be able to name who you're not for. If you can't, you haven't found the market. You've found some customers.
Revenue tells you someone bought.
The signals tell you whether the market wants this.
Only one of those is worth building a go-to-market motion on.