The Opportunity Score Formula, Explained for PMs Who Hate Spreadsheets

The opportunity score formula turns two customer ratings — how important an outcome is, and how satisfied customers are with it today — into one ranking number. Anthony Ulwick's version, Opportunity Score = Importance + max(Importance − Satisfaction, 0), is deceptively simple: it rewards importance twice, and only adds points for a gap when satisfaction genuinely lags behind it.

Quick answer: Opportunity Score = Importance + max(Importance − Satisfaction, 0). Importance sets a hard floor under every score, and the gap term only fires when satisfaction trails importance. That combination rewards a real importance-satisfaction gap, hands a "free" respectable score to outcomes that are already well-served, and almost never lets low-importance outcomes rank as opportunities — no matter how badly they're neglected.

What the Formula Is Actually Measuring

The opportunity score formula scores exactly two inputs per customer-desired outcome: an importance rating (usually 1–10, how much this matters) and a satisfaction rating (usually 1–10, how well it's handled today). There's no revenue estimate, no effort input, no strategic-fit judgment — it's a pure customer-stated-need ranking tool, nothing more.

This is the core mechanism behind Anthony Ulwick's Outcome-Driven Innovation (ODI) methodology, first published in What Customers Want (2005) and refined across two decades of work through his consultancy, Strategyn. It's often shorthanded as the Ulwick opportunity algorithm, and it sits inside the broader jobs-to-be-done tradition — worth a full read in our jobs-to-be-done complete guide if you're building an outcome list from scratch.

Clayton Christensen's parallel jobs-to-be-done research, popularized through Harvard Business School case studies and Competing Against Luck (2016), makes a related point about why outcome-level measurement matters. Christensen pointed to research suggesting that a large majority of the tens of thousands of new products launched each year fail to gain traction, largely because teams never pinned down the metric customers use to judge whether a job went well. The opportunity score formula is one attempt to make that metric explicit, instead of leaving it as a gut feeling.

The two inputs come from customer interviews or surveys, scored against desired outcome statements — sentences like "minimize the time it takes to reconcile a failed payment" or "increase the likelihood that a new teammate finishes setup without help." If those statements are vague or solution-flavored, the scores built on top of them are noise. That's exactly why writing outcome statements in the correct JTBD job-statement format matters more than the arithmetic that follows.

Once you have both ratings for an outcome, the formula is:

Opportunity Score = Importance + max(Importance − Satisfaction, 0)

In plain English: start with importance, then add the shortfall — but only if there is one. If satisfaction already meets or beats importance, the shortfall is zero and you add nothing. If satisfaction lags, you add the full difference. That asymmetry is the entire personality of the formula, and it's worth understanding before you trust a single ranked list it produces.

Why Importance Gets Counted Twice and Satisfaction Only Once

Look closely at the formula and importance appears in it twice — once standing alone, once again inside the gap calculation — while satisfaction appears exactly once, and only as a subtraction. That structural asymmetry means the score can never fall below the outcome's own importance rating, regardless of how satisfied customers already are.

Here's the proof, and it takes one line: since the gap term is max(Importance − Satisfaction, 0), it can never be negative. The smallest it can be is zero, which happens whenever satisfaction is equal to or greater than importance. So the smallest possible Opportunity Score for any outcome is just its Importance rating, plus zero. Importance is a floor, not just an input.

That floor has a very practical consequence: an outcome rated highly important will always land somewhere respectable on the leaderboard, even when the company is already nailing it. Compare two fully-satisfied outcomes — same zero gap, wildly different importance:

Both outcomes are equally, fully satisfied — zero unmet need in either case. But one scores 9 and the other scores 2, purely because of how important customers say each one is. That's the mechanism working as intended: importance is doing real information work even when there's no gap to close. But it also means:

• A high score is not proof of an open opportunity. It might just be proof that customers care a lot about something the product already does well.
• Ranking by raw score alone will always push high-importance outcomes toward the top, even the ones that are already solved.
• Two outcomes with identical scores can represent completely different situations — one urgent and unmet, one already excellent and simply important.

A Numeric Walkthrough: Three Outcomes, Three Verdicts

Running three outcomes through the formula side by side shows exactly how the same inputs produce three different verdicts — a big win, a borderline watch-item, and a non-issue — using nothing more than subtraction and a max() clamp. The verdict changes entirely based on where the gap falls relative to importance, not on the gap's absolute size.

Take a B2B project-management product doing onboarding research. Three desired outcomes come back from customer interviews, each scored 1–10 on importance and satisfaction:

Row one is the textbook opportunity: customers care a lot (importance 9) and the product is failing them (satisfaction 3), so the full 6-point gap gets added to importance for a score of 15. That's a genuine, well-evidenced candidate for the next roadmap cycle — high stakes, clearly unmet.

Row two sits in the middle deliberately. Importance is still high, but satisfaction has improved to 6, shrinking the gap to 3 and the score to 12. It's not screaming for attention, but it's not settled either — worth tracking across the next research cycle rather than shipping against immediately or ignoring outright.

Row three shows a fully satisfied, moderate-importance outcome. There's no gap, so the score equals importance itself — 6, which is low enough to fall well outside opportunity territory. Nothing here needs roadmap attention right now.

These three labels — Underserved, Target, and Served — aren't arbitrary; they're a standard way of bucketing opportunity scores into action tiers, and we break down exactly where those lines fall and why in our guide to the underserved, served, and target opportunity labels. The short version: scores in the mid-teens and above are your best evidence-backed bets, scores in the low teens deserve a second look next cycle, and everything below that is customers telling you, in effect, "we're fine."

The Quiet Lie: When a Well-Served Outcome Still Looks Like an Opportunity

A highly important outcome that's already well satisfied still lands respectably on a sorted list, purely because of the importance floor described above — and that respectable number can trick a PM into thinking there's still unmet demand to chase. The score doesn't distinguish a real open gap from "customers care a lot, and we've already handled it."

Go back to the "keep customer payment data secure" example: importance 9, satisfaction 9, score 9. Sorted next to a handful of moderate-importance-with-moderate-gap outcomes scoring 7 or 8, that fully-solved security outcome can outrank genuinely open opportunities on a raw-score leaderboard. Nothing in the number itself flags "this one's already done."

A high opportunity score only ever tells you the gap is big relative to how much customers say they care. It never tells you whether the gap is actually still open, or whether you already closed it and the score just hasn't caught up to that fact.

The fix isn't a different formula — it's reading the score alongside its inputs, not instead of them. Before acting on a ranked opportunity list:

1. Check the satisfaction number, not just the final score. A score of 9 built from importance 9 and satisfaction 9 needs zero engineering effort. A score of 9 built from importance 5 and satisfaction 1 needs real work.
2. Re-run the survey periodically. Satisfaction shifts as you ship; a score that was earned honestly six months ago can be stale today.
3. Treat "Served" as informative, not empty. An outcome sitting at Served with high importance is a signal to protect what's working, not to deprioritize the conversation entirely.

The Failure Mode: Why Low-Importance Outcomes Never Surface

Because importance appears twice in the formula and satisfaction only subtracts once, an outcome's maximum possible score is capped at roughly double its importance rating — which means low-importance outcomes can be almost completely unsatisfied and still never cross into "opportunity" territory. The formula structurally can't surface them, no matter how loud the complaints get.

The math, spelled out

Take the worst-case scenario for any outcome: satisfaction bottoms out at 1 (on a 1–10 scale). The gap becomes Importance − 1, and the score becomes Importance + (Importance − 1), or 2 × Importance − 1. That's the absolute ceiling for any outcome at a given importance level — it cannot score higher than this, ever, regardless of how dissatisfied customers become.

Read that table plainly: an outcome rated 6 or below on importance can be rated dead-last on satisfaction (a 1 out of 10) and it will still never reach even the "Target" threshold, let alone "Underserved." The customer could be describing genuine, sustained pain — and the number will report back a shrug.

What this misses in practice

This is precisely the importance-satisfaction gap trap that makes scoring feel untrustworthy to experienced PMs: an outcome doesn't need a small gap to get buried, just a low importance rating relative to everything else on the list. That's often the stuff that shows up as recurring, quiet friction rather than dramatic complaints — the kind of moment that only becomes visible when you trace a full journey rather than a single rated statement.

That's exactly where opportunity scoring needs company, not replacement. Mapping the emotional highs and lows in a complete customer journey surfaces low-grade friction that never gets its own outcome statement, let alone a high importance rating. And a systems-thinking view of causal loops and downstream effects catches structurally important issues that customers themselves may rate as merely "minor" because they don't see the second-order consequences.

Ulwick's own framework anticipates part of this gap with Forces of Progress analysis, which looks at the push, pull, anxiety, and habit forces around a switching decision rather than a single importance number. It's a reminder that the opportunity score was never meant to be the only lens — Ulwick built ODI as a system of methods, and the arithmetic is just the sorting step at the end of it.

Using the Opportunity Score Without Getting Burned

The opportunity score formula works best as a fast, defensible first pass at ranking dozens of outcomes — not as the final word on what to build. Treat it as a triage tool that narrows a long list to a short one, then bring in judgment, journey context, and engineering reality before anything hits a roadmap.

That discipline matters more than it sounds. Bain & Company's long-running customer-experience research found that a large majority of executives rated their own company's experience as superior, while only a small single-digit share of their customers agreed. Guessing importance and satisfaction ratings from inside the building, instead of actually asking customers, risks baking that exact same gap into your opportunity rankings.

A few habits keep the formula honest instead of misleading:

1. Write outcome statements before you score anything. Scores attached to vague statements ("better onboarding") are unusable; scores attached to a proper job statement format ("minimize the time to complete first setup") are comparable and specific.
2. Score in batches from the same interview set. Importance and satisfaction are relative judgments; mixing scores from different research rounds, different segments, or different survey wording quietly corrupts the ranking.
3. Re-score on a cadence, not once. A score is a snapshot. Ship against the top "Underserved" outcome, then re-survey — satisfaction should move, and the ranking should reshuffle accordingly.
4. Carry the winning outcome forward carefully. An outcome statement that scored well in research still has to survive contact with engineering — see our notes on how job statements survive the handoff to engineering for what tends to get lost in translation.
5. Don't let "Served" mean "ignore." A well-served, high-importance outcome is worth monitoring for regression, even if it never needs a sprint.

This is also the exact math Prodinja's Customer Jobs tool runs under the hood. Log your desired outcomes with importance and satisfaction ratings, and it computes Importance + max(Importance − Satisfaction, 0) for every one of them automatically, then labels each outcome Underserved, Target, or Served — so you can see the ranking, and check the inputs behind it, without opening a spreadsheet or rebuilding the formula yourself. It's the same arithmetic covered above, just applied consistently across a whole outcome list as part of walking through the JTBD workflow.

Key Takeaways

• The opportunity score formula is Importance + max(Importance − Satisfaction, 0) — two customer-rated inputs, no revenue or effort variables involved.
• Importance is a floor, not just an input. Because the gap term can't go negative, no outcome can ever score below its own importance rating.
• A well-served, high-importance outcome will still look respectable on a ranked list — check the underlying satisfaction number before assuming there's real work left to do.
• Low-importance outcomes are structurally capped: an outcome can be almost entirely unsatisfied and still never reach "opportunity" status if its importance rating is modest.
• The standard Underserved / Target / Served labels turn the raw number into an action tier — useful for triage, not a substitute for judgment.
• Pair the formula with a properly written job statement, a mapped customer journey, and a systems view of downstream effects — each catches something the arithmetic alone misses.
• Treat the score as a fast first pass across many outcomes, then apply engineering and strategic judgment before committing anything to a roadmap.