Stop Calling Product-Market Fit a Feeling

On My Mind

Hi — Asad here. Before you dive into the brilliance that is Mark Roberge, a quick update: Today is our first day on Substack. Since many of you have been along for the ride, you deserve to know why we’re here.

This started as a scrappy weekly newsletter to 10,000 people, sent via HubSpot. It wasn’t winning any design awards, but it got the job done, and we didn’t want to gold-plate anything until we knew that people cared. And they did. So, last Q4 we leveled up to Beehiiv — better cadence, sharper contributors, and design that finally stopped embarrassing us.

Why leave now? Because the long-form writing we were actually reading ourselves — and more importantly, engaging with — wasn’t on Beehiiv. It was on Substack. The people who love long-form were congregating here, and not being here felt silly.

And so, with this inaugural edition, we’re here with you, and excited to talk about everything GTM.

Now — on to Mark. Enjoy!

The Second-Act Stumble

Steve had a way of making growth sound inevitable. He’d done it once already — taken revenue from $27M to $54M in 12 months — and the board wanted another lap.

Another 2x.

Inside sales could squeeze out $30M more from their current motion. Not $54M. There was a $24M hole in the spreadsheet, and Steve hated holes.

The fix seemed obvious: build a new product. Sell it to current customers, dangle it in front of prospects, slap “new platform” on the website. When he unveiled the plan, heads nodded. The math worked.

The company mobilized. Engineers abandoned their roadmap. Marketing rebuilt the website. Sales held boot camps until every rep could pitch the new product cold.

The customer conference was supposed to be their coronation. It wasn’t.

A year later, the product sold about 10% of forecast. Early buyers found bugs, friction, and broken promises. Meanwhile, the original engine — the one that powered every prior growth milestone — had been left to idle. An entire quarter burned chasing pipeline that couldn’t carry the weight.

Steve didn’t just miss his number. He missed the point.

It’s a cautionary tale, but never more relevant than today. In a company’s early years, executives know not to commit to revenue goals too soon — they need design partners, time to work out product kinks, and room to learn how to sell the thing. But when those same executives reach scale and launch a new product or enter a new market, overconfidence kicks in. They assume everything automatically transfers.

Some institutional knowledge will, though much less than you think. Which is why you need to restart the product-market fit (PMF) and go-to-market fit process from scratch.

PMF Is a Number, Not a Feeling

Every year, whether I’m on stage at a tech conference or in front of a new cohort of Harvard MBAs, I ask the same question: “What is product-market fit?”

The answers are all over the place: “A revenue milestone.” “Customer count.” “Inbound leads.” “A gut feeling.”

I disagree with all of them.

Qualitatively, PMF occurs when customers continuously realize the value they were promised. Quantitatively, it shows up in long-term customer retention — when a customer renews, they’ve essentially bought the product again. That’s a far more reliable signal than any instinct.

Most industries target >90% annual retention. That’s your benchmark. Hit it, and you’ve got PMF.

Or, do you?

Customer retention may be the best statistical representation of PMF, but it takes quarters, sometimes a full year, to know whether the customers you’re acquiring today will stick. You don’t have that time.

That’s where the leading indicator of retention (LIR) comes in — what Silicon Valley calls the “aha moment.” Objective, measurable, and tied to your core value proposition, LIR is defined by three variables:

LIR = P% of customers achieve E event every T time → PMF

• P (Percentage): The minimum share of customers that must hit the leading indicator for you to claim PMF. Calibrate based on competitive intensity and your moat.

• E (Event): The specific behavior that signals value, correlated with your value proposition. It must be binary and measurable — it happened or it didn’t. ”Processed their first transaction” qualifies. “Customer sees value” doesn’t.

• T (Time): The frequency at which you evaluate. Shorter cycles accelerate learning — unless your product’s usage pattern doesn’t support it.

Two things to keep in mind as you build this out:

Track it by cohort. Recent cohort performance matters more than overall averages — your earliest customers aren’t necessarily representative of your current GTM reality.

Validate it over time. Compare LIR achievers against non-achievers using actual retention data. If customers who hit your LIR retain at 93% and those who don’t retain at 39%, you’ve got a signal worth betting on.

Prove You Can Acquire Customers Profitably

Proven PMF tells you that 100 new customers next quarter will likely realize value. But you’re not ready to scale yet. Because you haven’t proven that you can acquire and serve those customers profitably.

That’s go-to-market fit (GTMF): the ability to acquire and retain customers consistently and at a profit. Pure unit economics. The stuff that actually scales.

The software industry has settled on three GTMF benchmarks:

• LTV/CAC > 3

• Payback period < 12 months

• Magic Number > 1.0

But like retention, these are lagging — you won’t see the data for quarters.

Instead, work backwards to leading indicators of unit economics (LIUEs): short-term GTM targets that, if hit, confirm you’re on track for healthy, long-term economics. Plot actual LIUE performance against your targets. If actuals stay above the goal line, you’ve got GTMF. Now you can scale.

Scale Is a Cadence, Not a Switch

Most executive teams get this wrong. They treat scaling like a single event:

PMF achieved. GTMF proven. Go hire 40 salespeople. Then wait.

But that’s not scaling. That’s a bet.

The optimal pace of scale is the fastest pace your organization can sustain without losing PMF and GTMF. Think of it as a cadence, and your LIR and LIUE dashboards as your speedometer: add 10 salespeople per quarter over four quarters. If something breaks, pause, diagnose, fix, then resume. If everything holds, accelerate to 20.

Done right, this gives you six to 12 months of forward visibility your competitors don’t have.

The “Stay-or-Go-or-Slow” Framework

To operationalize this cadence, we use a quarterly operating motion at Stage 2 Capital: “Stay-or-Go-or-Slow.” Pre-agree with your board on three questions, then evaluate at each quarter-end:

• Stay: What would we need to see to hold course?

• Go: What would justify accelerating — increasing the hiring pace?

• Slow: What would tell us to slow down until a scale issue is resolved?

All green → Go. Green and yellow → Stay. Any red → Slow. No boardroom theater required.

A word to boards skeptical of this model: Successful exits take five to 10 years. Acquirers don’t care about your Q2-to-Q3 growth rate from eight years ago; they care about the last eight quarters. Adjusting pace isn’t lowering accountability but rather, managing with the best information you have.

Don’t Follow Steve’s Example

No one in GTM sets their mission as “grow slowly.” But half of the executives I work with scale too early and too fast, and the other half scale too late and too slow. Both are equally dangerous.

The goal isn’t a short-term “triple, triple, double, double.” If you’re shooting for long-term home runs, your LIR and LIUE dashboards are your scoreboard. Watch them. Act on them. And adjust in real time.

Subscribe now

Mark Roberge is the Co-Founder of Stage 2 Capital, the Founding CRO at HubSpot, and a Senior Lecturer at Harvard Business School. He has been featured in The Wall Street Journal, Forbes magazine, Inc. magazine, The Boston Globe, and Harvard Business Review. He’s also the best-selling author of The Sales Acceleration Formula and The Science of Scaling.