Sunk cost fallacy: stop chasing deals for hours already spent

The sunk cost fallacy is carrying on with something because of what you've already put into it, rather than what it will give you from here. In sales it's the lead you've messaged 14 times, the proposal you rewrote over a weekend, the trade show you paid for last year and book again. The fix is one habit: judge every deal on what's still ahead, and leave what it already cost you out of the sum.

A sunk cost is anything you've spent that can't come back. Those weekend hours on the proposal are gone whether the buyer signs next week or never replies again, which is why they can't tell you what to do next. Only the costs you can still avoid, and the results you can still get, belong in the decision.

Cambridge's business dictionary frames it as a company pressing on with a project because so much money, time or effort has gone in, when stopping is the better move. It gives the everyday name too: throwing good money after bad. You'll also hear it called the Concorde fallacy, after the British and French governments kept funding the Concorde aircraft once it stopped making economic sense.

The trap isn't foolishness. The past is simply the part you remember best. You remember every unanswered message. You don't feel the five new leads you didn't call while you were sending them.

Three different costs sit inside every "should I keep going?" question, and the fallacy comes from mixing them up. Here's how they look on one stalled deal.

The opposite of a sunk cost is a future cost, one you haven't paid yet and can still avoid, and it's the only kind worth arguing about. When a deal feels "worth it", check that you're weighing the second and third rows rather than the first.

Opportunity cost is the row people forget, because nothing reminds you of it. A lead who never got your call doesn't send you a message to complain.

The fallacy hides inside normal work. You won't think "I'm honouring a sunk cost". You'll think "one more try". Here's where it turns up from Monday to Friday:

• Monday's first call goes to the lead you've chased longest, since that's the name on your mind, even when a newer lead is likelier to say yes.
• Your prospecting block shrinks when forty minutes meant for new names go on rewording a follow-up to someone who stopped replying in August.
• A follow-up sequence runs on past any sense. Touch 12 goes out because touches 1 to 11 felt like an investment.
• The proposal gets a fourth version nobody asked for, because version one took your Saturday and you want it to pay off.
• The lead list you bought gets another month, because the first month cost so much.
• A client who eats your week and pays the least stays on the books, because you've given them two years.
• The Friday pipeline review keeps a deal at "likely" after six silent weeks, because writing "lost" feels like writing off the hours.
• In a manager's 1:1, a rep defends a deal they found themselves, and the manager defends the hire or the lead source they picked.

Outside work the pattern is the same. You sit through a film you dislike because you paid for the ticket, finish a course you've stopped believing in, or hold a share until it "gets back to what I paid". In each one, the reason to continue points backwards.

You aren't being careless when this happens. Your mind is built to hold on to what it has paid for.

Losing feels bigger than winning

Daniel Kahneman and Amos Tversky's prospect theory, published in Econometrica in 1979, describes a value curve that is steeper for losses than for gains. Put simply, a loss hurts more than a win of the same size feels good. Dropping a deal turns your effort into a loss you can see, so keeping it open feels safer, even when it isn't.

Not wanting to look wasteful

In 1985 Hal Arkes and Catherine Blumer published The psychology of sunk cost. They followed theatre season-ticket holders, and the ones who'd paid more for the subscription turned up to more plays over the next six months. Arkes and Blumer put it down to people not wanting to look as if they wasted the money. Sellers do the same with deals: the more one has cost you, the more you keep turning up for it.

Inflated odds

The same paper found that people who had already sunk money into a project rated its chance of success higher than people judging the same project fresh. That's the deal sitting at 70% in your pipeline review. The number rose with your effort. The buyer didn't move it. Set each deal's odds from the buyer's last move, and check your sales forecast against what actually closed.

It was your call

Barry Staw's 1976 study, Knee-deep in the big muddy, put 240 business school students through a role-played investment decision. They put the most money back into a losing course of action when they had chosen it themselves and were responsible for the bad result. Researchers call this escalation of commitment, a close cousin of the sunk cost fallacy. On a team, it's why the rep who sourced a lead fights hardest to keep it, and why the manager who picked a lead source keeps paying for it.

The frame you put on it

"Give up on six weeks of work" and "spend that time on leads who reply" describe the same choice. The first sounds like failure and the second sounds like a plan. Nothing about the deal changes. The wording changes your answer.

Knowing about it doesn't protect you

Arkes and Blumer also found the effect was no smaller in people who had taken economics courses. Knowing the definition won't save you. A rule you set in advance and a test you run on a schedule do more than good intentions.

Run this on one deal at a time. Once you've done it twice, it takes a minute.

1. What have I already spent? Put it on one line, such as "14 touches, 6 hours on the proposal, one lunch", because once it's written down you can set it aside on purpose.
2. What did the buyer do last? Look for their last reply, their last question and the last date they gave you. Deals move on what the buyer does, so check whether this one stopped on your move or on theirs.
3. Would I take this lead today? Picture it arriving this morning with only what you know now, and if you wouldn't start a proposal for it, the history is the only thing keeping it alive.
4. What's the next step, and by when? Keep going with a dated next step, pause with a date to check back, or close it as lost. A deal with no next date is a deal you've stopped deciding about.

Question 2 does most of the work. If the honest answer is "they said they'd check with their partner, five weeks ago", you already know. Next time, ask the four qualifying questions before you write a proposal, so you know what the buyer has committed to before you've spent a weekend.

Not every decision to keep going is the fallacy. Persistence wins deals, and a buyer who's busy isn't a buyer who's gone. The difference is which way the reason points.

Keep going when:

• The buyer has told you a budget, a date or how they'll decide, and none of it has changed.
• There's a next step on their side: a call they booked, a question they asked, a document they want.
• You've promised something, such as the work a paying client is owed, even if you've gone off the account.
• Stopping has its own future cost. Switching tools in your busiest month costs hours you haven't spent yet, so those hours belong in the sum.

The test is simple. If you'd pick this deal with a blank slate, you're persisting. If your best reason is "I've already put so much in", you're paying for the past. A written sales follow-up process decides how many tries a quiet lead gets before you're attached to it, so the line is drawn in advance.

You can't switch the bias off, but you can put the facts in front of you at the moment you'd otherwise go on feel, and two habits do most of that work.

Set the exit rule before you start. When a deal enters your pipeline, decide what would close it: "three follow-ups after the proposal with no reply", or "no meeting booked within 30 days". A rule made on day one, before you're invested, is hard to argue with on day 40.

Then hold a 20-minute stale-deal review every Friday. For each open deal that hasn't moved, check three facts: the date of the buyer's last reply, how many times you've followed up since, and how long it has sat in its stage. Run the four questions on anything that breaks your exit rule. If you already run a weekly pipeline review, fold this into its stuck-deals slot.

Write each week's closes down in one line too: what made you close, and how long you held on after you knew. After a month you'll see your own pattern, such as holding on three weeks too long to any deal with a site visit in it. A weekly reflection template gives that line a home.

Managers meet the fallacy twice: in their reps' pipelines and in their own decisions.

In a rep's pipeline it looks like a forecast full of deals that are only history, where the same names sit at "likely" every week, follow-up dates keep sliding and the proposal stage never empties. Skip "why hasn't this closed?" in the 1:1. It makes the rep defend the hours. Ask "If this came in today, would you take it?" instead, and then "What did the buyer do last?"

Make closing a deal as lost a normal thing to do, and say so out loud. A rep who's praised only for wins learns to keep dead deals open, because a lost deal looks worse on the board than a stalled one. A deal closed with a reason teaches you something. A stalled one teaches you nothing.

Get a reason on every lost deal. Your rep drags the card into Lost, ActivityTracker asks why, and one tap on "No reply" or "Not interested" answers it. After a month the reasons show whether your team loses deals to silence or to a clear no, and those are different problems with different fixes.

Then run the same test on your own choices: the hire who still isn't working out after a fair run, the lead source you championed, the event you book every year. Staw's students put the most money back into choices they'd made themselves, and a manager's choices are no different.

You can run the same four questions at home, more gently. A course you've paid for and stopped believing in, a car you keep repairing because of the last repair bill, a share you hold until it "gets back to what I paid": in each one the money is gone whether you continue or not, and only the next step is yours to choose.

People also stay in friendships, partnerships and relationships because of the years already given. Those years were real, and they mattered. On their own, they're still not a reason to spend the next ones. Ask what the next year looks like, and whether you'd choose it today knowing what you know. If the stakes are high, talk it through with someone you trust.

• Cambridge Business English Dictionary, sunk cost fallacy, read 28 October 2026
• Cambridge Business English Dictionary, Concorde fallacy, read 28 October 2026
• Hal R. Arkes and Catherine Blumer, The psychology of sunk cost (Organizational Behavior and Human Decision Processes, 1985, abstract), read 28 October 2026
• Barry M. Staw, Knee-deep in the big muddy: a study of escalating commitment to a chosen course of action (Organizational Behavior and Human Performance, 1976, abstract), read 28 October 2026
• Daniel Kahneman and Amos Tversky, Prospect theory: an analysis of decision under risk (Econometrica, 1979, abstract), read 28 October 2026