Opportunity Cost
Opportunity cost is the value of the next-best alternative you didn't pick — the hidden price tag on every test slot, dev sprint, and agency hour you spend.
Opportunity Cost
The value of the next-best alternative you gave up when you chose one option over another.
Opportunity cost is the value of the option you didn't pick. Every decision — which test to run this sprint, which page to redesign, which agency retainer to renew — silently consumes the upside of the alternatives you set aside. The cost isn't what you spent; it's what you could have earned instead.
In e-commerce and CRO, opportunity cost becomes load-bearing the moment your test velocity is constrained. With only four test slots a month and a backlog of forty hypotheses, the real price of running a low-confidence button-colour test isn't the developer time — it's the checkout-flow test you didn't run in that slot.
Accountants count what you paid. Strategists count what you missed. Opportunity cost is the strategist's metric — it shows up nowhere on the invoice, but it determines whether the next quarter beats the last one.
The concept applies anywhere resources are finite: marketing budget, engineering sprints, content production, test slots, agency hours. If picking option A means you cannot also pick option B, then B's expected value is the true cost of A — on top of A's cash cost.
Opportunity Cost = Expected Value (Next-Best Option) − Expected Value (Chosen Option)
EV(next-best)
Expected value of the next-best alternative
The probability-weighted return of the option you ranked second.
EV(chosen)
Expected value of the chosen option
The probability-weighted return of the option you actually picked.
A Shopify apparel store has one test slot left in the sprint. Option A: test a new product-page hero image (estimated +1.2% conversion lift, 60% confidence, on €180k of monthly traffic). Option B: test a simplified checkout step (estimated +2.8% lift, 50% confidence, on the same €180k).
EV of Option A (hero image): 0.012 × 0.60 × €180,000 = €1,296
EV of Option B (checkout): 0.028 × 0.50 × €180,000 = €2,520
→ Picking Option A carries an opportunity cost of €2,520 − €1,296 = €1,224 per month.
The hero-image test isn't free even if dev time is zero. Running it costs you €1,224 of expected lift that the checkout test would have produced. Over a year, that's roughly €15k of foregone revenue — for one bad slot allocation.
This is why Experiment Prioritization frameworks (ICE, PXL, PIE) exist: they're opportunity-cost calculators in disguise. They force you to rank hypotheses so the slot always goes to the highest-EV option, which mechanically minimises what you're leaving on the table.
Typical opportunity cost of common test-slot misallocations (monthly, illustrative)
| Slot used on… | Instead of… | Typical EV gap | Annualised cost |
|---|---|---|---|
| Button colour / micro-copy | Checkout-step reduction | €800 – €2,500 | €10k – €30k |
| Homepage hero swap | Product-page social proof | €500 – €1,800 | €6k – €22k |
| Pricing-page font test | Cart-abandonment email flow | €1,200 – €3,500 | €14k – €42k |
| Generic CTA wording | Mobile PDP layout overhaul | €900 – €2,800 | €11k – €34k |
| Footer link reshuffle | Search-results page UX | €600 – €2,000 | €7k – €24k |
Opportunity cost is also why Mental Models like this one stay useful long after the spreadsheet is closed. You won't run the EV calculation for every micro-decision, but internalising the question — "what am I giving up by saying yes to this?" — changes how you triage the backlog, defend roadmap cuts, and push back on stakeholder requests.
Frequently asked questions
It's the value of the best thing you didn't choose. If you spend a test slot on a homepage tweak instead of a checkout fix, the opportunity cost is the lift the checkout fix would have produced.
Sunk cost is money or time you've already spent and can't get back. Opportunity cost is the value of an alternative you're still able to pick. One is backward-looking, the other forward-looking.
Test slots, dev sprints, and analyst hours are finite. Every test you run blocks another from running this cycle. Ignoring opportunity cost is how teams end up with a heavy testing calendar and a flat conversion rate.
Estimate the expected value of each candidate test (lift × confidence × revenue base), rank them, and subtract the second-place EV from the first-place EV. That gap is what you'd lose by picking the wrong one.
No. ROI measures return on what you spent; opportunity cost measures return on what you didn't choose. A test can have positive ROI and still carry a brutal opportunity cost if a better test was on the table.
Prioritization frameworks (ICE, PXL, PIE) exist to minimise opportunity cost. They rank hypotheses by expected impact so your scarce test slots always go to the highest-EV option in the backlog.
Only if all alternatives have identical expected value, which almost never happens in practice. If you've done the work to estimate impact, one option will dominate — and ignoring that gap is the cost.
Frame every "yes" as a "no" to something else. "If we run the hero test this sprint, we delay the checkout test by four weeks — that's roughly €5k of foregone lift." Concrete numbers stop the politics.
Yes. Any finite resource has an opportunity cost. An agency hour spent on a low-impact report is an hour not spent on funnel diagnostics. The discipline is the same: rank, pick, and acknowledge what you're skipping.
Treating it as invisible. Because no invoice ever lists it, teams default to picking whatever's easiest or loudest, not what's highest-EV. The fix is making the trade-off explicit in every prioritisation meeting.
Test ideas before you ship them
Run unlimited A/B tests, attach hypotheses to outcomes, and build a searchable archive of what works — and what doesn't.