Problem/Solution Fit — How You Know You Have It
P/S fit is not a feeling — it is a measurable state defined by specific behavioral evidence from real customers using the real solution.
🎯 By the end you'll understand
The eight behavioral signals that operationally define P/S fit — and why only behavioral evidence qualifies. The critical distinction between P/S fit (solution working for known customers) and product-market fit (solution working for a market). How to score pilot customers against each criterion and identify specific evidence gaps before pursuing paid conversion. And the most dangerous failure mode: the pilot that looks like fit because the founder is still doing all the onboarding — and collapses the moment the founder leaves.
KEY INSIGHTP/S fit is a state of evidence, not a state of mind
Founders who say "I can feel that customers love it" have not yet found P/S fit — they have found a feeling, which is precisely what fit is designed to replace.
Three pilot customers are actively using Forge's edge device and dashboard. They attend weekly calls. They send you Slack messages about features they want. They said "this is exactly what we need" in the first month. That evidence says almost nothing about whether you have P/S fit — because none of it is behavioral. A customer who attends your calls, praises your product, and requests features can still churn the moment you stop managing the relationship. The question is not whether they like it. The question is whether they have changed how they work because of it — and whether the change is sticky enough to persist without you in the room.
P/S fit exists when real customers using the real solution produce behavioral evidence — actions that reveal genuine value capture, not stated preference or politeness. There are eight specific behavioral signals that qualify. When a pilot customer shows at least six of the eight at a meaningful intensity, you have P/S fit for that customer. When you have it for three or more customers who did not require extraordinary founder involvement to achieve it, you have the evidentiary basis for pursuing paid conversion and eventual product-market fit.
DEFINITIONWhat behavioral evidence actually means
The distinction between behavioral and attitudinal evidence is not pedantic — it is the difference between a signal you can act on and a signal that will mislead you.
| Evidence type | Example | Weight for P/S fit | Why |
|---|---|---|---|
| Attitudinal | "This is really useful — we love the dashboard" | Does not count | Stated preference; can coexist with zero intention to continue, pay, or integrate |
| Attitudinal + specificity | "The alert on Monday caught a defect our inspector missed" | Weak signal only | One-time event recall; still not evidence of durable behavioral change |
| Behavioral — usage | Logs in 3x per week without a prompt from you for 6+ consecutive weeks | Qualifies | Revealed preference: time is a scarcer resource than praise |
| Behavioral — economic | Renews or pays above minimum; extends pilot scope unprompted | Qualifies strongly | Money has the highest cost of any signal; spending it is harder to rationalize than words |
| Behavioral — social | Introduces you to a peer at another plant without being asked | Qualifies | Puts their own credibility at stake; only done when confident in the value |
Notice that the behavioral category is not about the magnitude of the action — a three-line Slack message to a colleague at another plant is stronger P/S fit evidence than a glowing 30-minute testimonial call. The behavioral signal has cost attached to it. The testimonial does not.
CORE CONCEPTThe eight behavioral signals of P/S fit
These eight signals were not derived from theory — they describe what customers do when a solution is genuinely working, as distinct from what customers do when they are being polite to a founder they like. Not every fit scenario requires all eight; treating six as a meaningful threshold is defensible. But the absence of several of these signals in a pilot that "feels successful" should give you pause.
Repeat usage without prompting
The customer logs in, checks the dashboard, or runs inspections on a schedule that was not set up by you. This is the floor of behavioral fit. If the product is only used when you send a reminder, check-in email, or weekly call invitation, the usage is being driven by your relationship management, not by the product's value. Threshold: independent, self-initiated usage at least 3x per week for a rolling 4-week period.
Unprompted referrals
The customer introduces you to someone else — a peer at another facility, a colleague at a supplier, an industry contact — without being asked or incentivized. The referral does not need to close. The act of making it is the signal. A referral costs the customer their credibility. It is only made when confidence in the product's value is high enough to stake that credibility on. One spontaneous introduction beats a hundred positive survey responses.
Workflow integration
The customer has changed how their team works because of your product. This can be formal (a new QA sign-off step that includes dashboard review before a shipment is cleared) or informal (the shift supervisor checking the alert feed at the start of each shift as part of their opening routine). Workflow integration is evidence of the most durable form of value: the product has been embedded into behavior that would require deliberate effort to remove. Products that sit "alongside" a workflow are much more fragile than products that have become part of it.
Specific and quantified value articulation
The customer can articulate the value they have received in specific, quantified terms without being coached. Not "it's been really helpful" but "we caught three defects last month that would have been quality escapes — at $18,000 each, that's $54,000 we avoided." The specificity and quantification matter: they indicate that someone on the customer side did the work of connecting product events to business outcomes. That calculation does not happen unless the value is real and visible.
Resistance when usage is threatened
When you mention the possibility of ending the pilot, pausing access, or changing the pricing significantly, the customer reacts with genuine resistance — not politeness, but pushback that references specific business impact. "We can't go back to the old way" and "I'd have to go to my VP to fight for budget if you raised the price" are resistance signals. "That would be a shame" is not. Test this deliberately: mention a scenario that would end or significantly constrain their access and observe the quality of the response.
Willingness to pay above competitor price
The customer, when given a realistic pricing conversation, expresses willingness to pay at a level that reflects the value they have received — and that exceeds what the obvious alternatives cost. This is not "they said they'd pay"; it is "they engaged seriously with a pricing conversation and named a number or a budget range that exceeds what they currently spend on alternatives." Willingness to pay above competitor price is the economic validation that the value premium is real.
Proactive feature requests
The customer requests features that would make the product more deeply embedded in their specific workflow — not features that would be generally nice, but features that reveal they are thinking about how to get more value from a product they are already committed to. "Could you add an export that feeds into our MES system?" is a deeper commitment signal than "could you add dark mode?" The former requires integration work they would have to manage on their side; requesting it means they are planning to be a customer long enough for it to matter.
Complaints when unavailable
When the product goes down, the customer notices and complains — quickly, and with specific reference to what they could not do. Complaints are a fit signal that counterintuitively most founders miss: they reveal that the product is part of an active workflow, not a passive experiment. The character of the complaint matters: "the alert feed was down for two hours and we had to call the inspector back to the line" is a genuine disruption. "I noticed the dashboard was slow yesterday" is politeness. Real disruption-based complaints are strong evidence of workflow integration (Signal 3) and confirm that the product is genuinely embedded.
COMPARISONP/S fit vs. product-market fit — not the same thing
This distinction gets collapsed constantly, and the collapse produces decisions that look reasonable but lead to premature scaling, underbuilt sales teams, and failed transitions from founder-led to repeatable growth.
| Dimension | P/S fit | Product-market fit |
|---|---|---|
| What it proves | The solution works for a specific, known set of customers who have the specific problem you solved | The solution works for an identifiable market at the pace and cost of acquisition required for the business model |
| Sample required | 3–10 pilot customers with deep behavioral evidence; founders personally involved in activation | Enough cohorts to distinguish genuine retention from sampling artifact; acquisition without founder selling |
| Acquisition method | Any — including founder network, warm intros, white-glove onboarding | Must be repeatable at CAC the unit economics can support; not founder-dependent |
| Primary signal | Behavioral evidence from existing customers that the solution is working | Retention curves flattening; organic referral as a measurable acquisition channel; CAC declining as playbook matures |
| Common confusion | Treating founder-managed pilots as representative of independent customer behavior | Treating fast early growth from founder network as market pull |
| Where you are in this course | End of Stage 2 | Mid Stage 5 (after you have built the GTM motion and collected retention cohorts) |
P/S fit is not a weaker version of PMF. It is a different question. You can have genuine P/S fit — the product is working, customers are showing behavioral evidence, and the value is real — without having PMF, because you have not yet proven you can acquire the next 50 customers the same way you acquired the first three. Conversely, you cannot have PMF without P/S fit: a market that is adopting a product that does not actually work for them is a distribution illusion, not a business.
Closing your first three paid pilots does not mean you skip P/S fit and go straight to scaling. The paid contract gives you the right to try to achieve P/S fit with that customer. P/S fit is achieved only when the behavioral signals appear in the absence of founder handholding. This is the gate that separates Stage 2 from Stage 3: you do not build a revenue architecture until you know the solution works without you in the room.
SEE ITP/S Fit Scorecard — Forge's three pilot sites
Two things to notice from the scorecard. First, the aggregate score can look healthy while a specific site has a critical gap — a site with strong usage signals but zero WTP evidence is not fit-ready for conversion regardless of the aggregate. Second, the falsify field forces you to articulate what you have not yet tested. The most common dangerous gap for early-stage founders: a site with excellent usage scores and zero resistance-to-loss data, because the founder has never once tested what happens when they are not actively managing the relationship.
FORGEScoring Forge's pilots toward the Stage 2 milestone
Forge's Stage 2 milestone requires three paid pilot contracts at $8K each, an edge device plus dashboard MVP deployed at one manufacturing site, first activation data collected, and evidence that customers who complete onboarding see a measurable defect detection improvement within 30 days. P/S fit is the evidentiary basis for calling Stage 2 complete and moving to Stage 3 (Revenue Architecture).
All three pilot sites must reach a score of at least 10 out of 16 on the eight-criterion scorecard as measured without the founder on a call in the preceding two weeks. At least two sites must show the workflow integration signal (Signal 3) — confirmed by a change to an internal QA procedure that references the Forge dashboard or alert output. At least one site must have given a specific, quantified value statement unprompted. And at least one site must have expressed unprompted interest in converting to a paid contract before the pilot end date — not in response to a renewal conversation you initiated.
The most critical factor here is the parenthetical: "as measured without the founder on a call in the preceding two weeks." This is the specific operationalization of the founder-effect failure mode described in the next section. Forge should design a 2-week withdrawal period into the Stage 2 conclusion — deliberately stopping all proactive outreach to pilot sites and observing whether usage continues, whether questions arise without prompting, and whether any of the commitment signals appear spontaneously. What you see in that 2-week window is the closest approximation to what the product will look like when a salesperson (not you) onboards the next customer.
WEAKNESSThe founder effect — the failure mode that looks like success
This is the most dangerous failure mode in P/S fit assessment because it passes every surface-level check. The pilot is running. The customers are engaged. Usage is happening. Calls are going well. The NPS is high. The product roadmap is being co-developed with the customer. And none of it is evidence that the product works independently of you.
The failure mode is this: you are doing all the onboarding. You are explaining the alerts, contextualizing the data, answering questions, nudging usage when it slips. The customer's engagement with the product is mediated entirely through their relationship with you. When you leave — when a salesperson takes over the account, or when the company scales beyond founder-managed relationships — usage collapses, not because the product is bad, but because the product never actually replaced the value you were providing in person.
You are the primary user-support channel for all pilot customers. You have never missed a weekly call and neither have they. Customers ask you operational questions about their own manufacturing process that the product should be answering. When you were unavailable for a week, usage dropped. Your customers describe the value as "working with you" rather than "using the product." Every one of these is evidence that the pilot's success is founder-dependent and will not transfer to a repeatable customer acquisition motion.
The test is simple to design but psychologically difficult to run: stop. Stop the weekly calls. Stop the proactive check-ins. Stop answering the quick questions that come in on Slack. Wait two weeks and observe. Independent usage continuation is the signal that the product has genuine value. Usage collapse is the signal that the pilot was measuring your relationship management capability, not the product's value delivery.
Note: running the withdrawal test does not mean abandoning customers who are genuinely stuck on a real product issue. It means distinguishing between "stuck on a product issue" (requires a fix) and "not using the product unless the founder is present" (requires an honest conversation about whether P/S fit has been achieved). Most founders, if they are honest with themselves, know which one they are dealing with before they run the test.
WEAKNESSWhere this bites
A customer who praises your product in every call, introduces you to their VP, and asks you to present to their team is exhibiting relationship behaviors, not product-value behaviors. Enthusiasm is correlated with fit but is not itself evidence of it. The specific failure mode: you score a pilot customer as "fit-ready" because they are positive, engaged, and vocal — and then they churn three months after you transition the account to a customer success manager who does not have the same personal relationship. Every behavioral signal in the scorecard has a corresponding attitudinal decoy that looks similar from a distance. The discipline is to require the behavioral version before counting it.
The withdrawal test is uncomfortable to run because it risks disrupting relationships during a critical pilot period. Most founders find a reason not to do it: "we're still in the onboarding phase," "we don't want to signal that support is ending," "we can test independence after we sign the paid contract." These reasons are defensible on their own terms and collectively catastrophic. Signing a paid contract with a customer whose fit is founder-dependent does not reveal the dependency — it buries it under a contractual obligation. The dependency surfaces later, at the worst possible time: during a renewal conversation with a customer who stopped getting value six months ago.
A scorecard output of "strong fit" does not mean the customer is ready for zero-touch renewals. It means the behavioral evidence supports moving to paid conversion and continuing to deepen the integration. P/S fit is an entry gate, not a finish line. The specific error: treating a strong scorecard score as permission to deprioritize customer success and redirect attention to new customer acquisition. The customers who showed P/S fit are the ones who will generate the case studies, references, and retention data that make PMF achievable. Neglecting them after a strong scorecard score is one of the most common Stage 3 mistakes in B2B startups.
KEY INSIGHTHow this connects
- 1.9 · The Validation Verdict — Go, Pivot, or Kill — the behavioral evidence framework for P/S fit assessment in this lesson is the downstream application of the falsification criteria introduced at Stage 1's close. The "kill signal" and "pivot signal" distinction from Lesson 1.9 is directly relevant: a pilot that fails the behavioral scorecard is a pivot signal (the solution needs redesign), not necessarily a kill (the problem is still real).
- 2.6 · Testing the MVP — What Good Feedback Looks Like — the feedback quality rubric from Lesson 2.6 provides the foundation for the behavioral vs. attitudinal distinction made here. The MVP testing framework identified what good feedback looks like during development; this lesson defines what good evidence looks like after deployment.
- 2.7 · The Pivot Decision — When and How to Change Course — a failed P/S fit scorecard is one of the clearest pivot triggers available. Lesson 2.7's framework for distinguishing product pivots from market pivots from business model pivots applies directly when specific behavioral criteria are absent in multiple pilot sites.
- 2.8 · From MVP to Product — Designing the First Shippable Version — the first shippable version was designed to create verifiable value; this lesson defines what verification actually looks like at the behavioral level. The activation data collected in the deployed MVP (the Forge edge device + dashboard milestone) is the raw material for the scorecard criteria in this lesson.
- Next: 3.1 · Business Model Archetypes → — Stage 3 opens by defining what kind of business Forge is building. The P/S fit evidence from Stage 2 directly informs the pricing mechanism and the value metric that Stage 3's revenue architecture will be built on. You do not design a revenue architecture until you know what customers are willing to pay to keep, which is precisely what the behavioral scorecard measures.
P/S fit is the inflection point between building and selling. Before it, you are still discovering what the product needs to be. After it, you have a working solution that creates verifiable value for specific customers — and the question shifts from "does this work?" to "can we make it work for the next 500 customers at a cost the economics can support?" The behavioral scorecard is the instrument that tells you which side of that inflection point you are on.
GLOSSARYKey terms introduced
- Problem/solution fit (P/S fit)
- The state in which a specific solution creates verifiable, durable value for a specific set of customers who have the specific problem being solved — evidenced by behavioral signals, not stated preference. Achieved before product-market fit; a necessary precondition for it.
- Behavioral evidence
- Signals derived from what customers do rather than what they say. Usage patterns, payment decisions, workflow changes, referrals, and complaints when unavailable are all behavioral. Praise, positive survey scores, and enthusiastic calls are attitudinal and carry significantly less epistemic weight for fit assessment.
- Founder effect
- The distortion of pilot success metrics caused by active founder involvement in usage, support, and relationship management. A pilot that passes fit criteria only when the founder is actively managing the relationship is measuring founder capability, not product value. The withdrawal test is the diagnostic.
- Withdrawal test
- A deliberate pause in founder-initiated outreach and support — typically 2 weeks — designed to measure whether customer usage and engagement continue independently. Independent continuation is evidence of genuine P/S fit; usage collapse is evidence of founder-dependent engagement.
- Workflow integration
- The behavioral signal in which a customer has modified their internal operating procedures, checklists, or team routines to incorporate a product's output. Products that are integrated into workflows are substantially harder to churn than products that sit alongside workflows. The highest-durability form of P/S fit evidence.
- Resistance to loss
- A behavioral P/S fit signal revealed when the customer responds to a threatened loss of access — a pricing increase, a pilot end date, a feature removal — with specific, substantive pushback referencing real business impact. Distinguishable from politeness: genuine resistance includes reference to what specific operations would be disrupted.
- Willingness to pay above competitor price
- The customer's demonstrated readiness (in a pricing conversation, not a hypothetical) to pay a premium over alternatives that nominally address the same problem. Evidence that the value premium your solution creates is recognized and real — not just relative to doing nothing, but relative to the next-best option.
- P/S fit vs. PMF
- P/S fit proves the solution works for known customers who were acquired through any means, including founder relationships. Product-market fit proves the solution works for an identifiable market segment at acquisition economics the business model can sustain — without founder-dependent sales. P/S fit is achieved at the end of Stage 2; PMF is the target of Stage 5.
ACTIVE RECALLCheck your understanding
Answer in your own words first, then reveal.
1A pilot customer praises your product in every weekly call, introduces you to their VP, and has expanded the pilot to a second production line. Why might this not constitute P/S fit evidence — and what specific evidence would you need to see instead?›
All three behaviors described are relationship signals, not product-value behavioral signals. Praise on calls is attitudinal. Introducing you to their VP is a social behavior mediated by the personal relationship; it would happen whether or not the product was working. Expanding the pilot could be the founder's suggestion rather than an unprompted customer initiative. None of these tells you what happens when you stop managing the relationship.
The behavioral evidence you'd need instead: self-initiated login data showing usage without a weekly call prompt (Signal 1); a documented change to their QA procedure that references Forge alerts (Signal 3); a specific, quantified value statement the customer articulated unprompted — e.g. "three quality escapes caught last month, $54K avoided" (Signal 4); genuine pushback when you raise the prospect of pilot ending, with reference to specific operational disruption (Signal 5); and at least one unprompted referral to an external contact they made on their own initiative (Signal 2). The distinction between enthusiastic customers and fit-ready customers is whether they are doing things that cost them something — time, credibility, or money.
2What is the founder effect, and why is it specifically dangerous at the Stage 2 to Stage 3 transition?›
The founder effect is the distortion where pilot success metrics reflect the founder's personal relationship management rather than independent product value. The founder is running weekly calls, answering questions, contextualizing alerts, and providing de facto consulting services that make the product seem more useful than it would be if used independently. Customers who appear "highly engaged" are engaging with the founder, not the product.
It is specifically dangerous at the Stage 2 to Stage 3 transition because Stage 3 involves building a revenue architecture — pricing, packaging, and a repeatable commercial model — predicated on the assumption that the product creates value for customers in a way that can be delivered by a sales and customer success team, not just by the founders. If the pilots passed the Stage 2 milestone under founder-dependent conditions, the revenue architecture will be built on false premises. The contracts will be signed, the pricing will be set, and the CSM team will be hired — all before anyone discovers that usage drops when the founders step back. By then, the business model assumptions are baked in and painful to revise.
3Why is the "complaints when unavailable" signal (Signal 8) evidence of P/S fit rather than evidence of a bad product?›
Because the signal is not "the product went down" — it is "the customer noticed the product went down and experienced a specific, disruption-based reaction." A complaint about downtime requires that the product was in use at the time of the outage, that the outage caused a workflow disruption, and that the customer cared enough to contact you. All three of those conditions require that the product is embedded in an active operational workflow — which is Signal 3, the workflow integration signal.
Products that are not being used cannot produce downtime complaints. Products that are used but not workflow-embedded produce muted reactions ("I noticed the dashboard was slow"). The specific, disruption-referencing complaint — "we had to call the inspector back to the line because the feed was down" — reveals operational dependency that is only possible if the product has replaced some part of the previous workflow. That replacement is the thing P/S fit is trying to measure. The complaint is evidence that the replacement happened and is real.
4A startup has three pilot customers who all show strong scores on signals 1 (repeat usage) and 4 (quantified value), but zero scores on signal 2 (unprompted referral), 5 (resistance to loss), and 6 (WTP above competitor). What does this scorecard tell you, and what should you do before pursuing paid conversion?›
The scorecard tells you the product is being used and the customers can articulate the value — which is meaningful — but that none of the commitment signals have appeared. Customers are getting value but have not yet made the behavioral moves that reveal genuine commitment: they have not put their credibility at stake with a referral, they have not been tested on what they would do if they lost access, and they have not engaged in a real pricing conversation. This pattern is common in early pilots and usually means one of three things: the value is real but diffuse enough that it has not yet produced commitment; the founder has not yet tested the commitment signals deliberately; or the customers are genuinely engaged but the product has not yet crossed the threshold of importance in their operational context.
Before pursuing paid conversion: deliberately test Signal 5 by having a conversation that mentions the pilot end date or a pricing increase, and observing the quality of the response. Test Signal 6 by engaging in a specific pricing discussion with a dollar figure attached. For Signal 2, ask each customer if they know any peer at another facility who might benefit — but make this a favor request with real stakes, not a referral-program pitch. If none of these tests produces the behavioral signals, you have high-usage, low-commitment pilots — they are paying you in time but not yet willing to pay in money or credibility, which is a meaningful gap before commercial conversion.