Loss Aversion, Applied to Declined Service Work

Numa works with 1,500+ car dealerships and helps service teams follow up on work customers decline during a service visit. Approving a repair means accepting a sure cost today to avoid a cost that may or may not come later. Prospect theory suggests that many people will take that gamble. Loss aversion research also suggests a follow-up ask has a better chance when the cost of waiting is concrete and yes is easy.

Kahneman and Tversky’s 1979 prospect theory paper in Econometrica shows the pattern. Students and university faculty answered hypothetical choice problems. In one problem, 95 respondents chose between a sure loss of 3,000 and an 80% chance of losing 4,000. At the time, 3,000 pounds was about a month’s median net income for a family. Of the 95, 92% chose the gamble, even though its average cost was 3,200.

Flipped to gains, the choice reversed: 80% took a sure 3,000 over an 80% chance at 4,000. The authors called this mirror image the reflection effect.

Key takeaway: In one 1979 problem, 92% of respondents facing a sure loss chose a gamble that cost more on average. Kahneman and Tversky concluded that certainty makes a loss feel worse, just as it makes a gain more attractive.

Why a Sure Loss Feels Worse Than a Risky One

The paper put the core idea plainly: “losses loom larger than gains.” The aggravation of losing a sum of money, the authors wrote, appears greater than the pleasure of gaining the same amount.

The choice above adds a second finding. Certainty, they concluded, increases the aversiveness of losses. In that problem, a loss that might happen looked better than a smaller loss that would happen.

The same paper argued from its data that people judge outcomes as changes from a starting point, the reference point, rather than as final totals.

The 1979 paper gives no ratio for how much larger losses loom. A later estimate comes from Brown, Imai, Vieider and Camerer’s 2024 meta-analysis in the Journal of Economic Literature, which pooled 607 estimates from 150 articles. The mean loss-aversion coefficient was about 1.96, with a 95% interval of 1.82 to 2.10. In plain terms, a loss weighed roughly twice as much as a gain of the same size.

Why Declined Service Work Looks Like a Loss Decision

A customer arrives expecting a certain visit and a certain bill. That expectation is one plausible reference point. Measured against it, a recommended repair can register as a loss. When the advisor goes over the multipoint inspection (MPI) findings, each approved item adds a sure, immediate cost to today’s repair order (RO).

Declining trades that sure cost for an uncertain, later one. The part may last another season, or it may fail sooner. That is the shape of the 1979 choice: a sure loss against one that may or may not come. Prospect theory suggests many customers will take the gamble and wait.

One passage comes close to the service lane. It describes protective actions where “one pays a certain cost to reduce the probability of an undesirable event.” One of its examples is “the replacement of old tires.” In a related insurance problem, 80% of 95 Stanford University students turned down protection that lowered the odds of a loss without removing it.

That mapping is an inference. The 1979 problems were hypothetical choices, and no one in them was standing at a service counter. The 1979 gamble was also the worse bet by design, and a real decline can be the right call for a customer’s budget or timing.

One published dealership figure comes from J.D. Power’s 2015 U.S. Customer Service Index (CSI) Study, which covered owners of 1- to 5-year-old vehicles serviced at franchised dealerships. In that 2015 study, 29% said the advisor recommended additional work. Of those, 47% agreed to have it done. The revenue side of declined work is the subject of an earlier post on the declined-service follow-up gap.

Cox Automotive’s 2025 Fixed Ops and Ownership Study surveyed 2,502 consumers in September and October 2025, all with a service visit in the past year. Among dealership service customers, 69% said reminders about previously declined maintenance are helpful.

What Raises Loss Aversion, and Where the Research Has Limits

In one large study, loss aversion ran higher among people who knew less about what they were weighing. Mrkva, Johnson, Gächter and Herrmann’s 2020 study in the Journal of Consumer Psychology measured loss aversion in five samples totaling 17,720 people. More domain knowledge and experience went with lower loss aversion, though people at every level were still loss averse.

One sample was 360 European car buyers. Buyers who knew more about one attribute, such as fuel economy, were much less loss averse for it, with little or no change for others. The study valued car features, so applying it to repairs is an inference. It suggests that explaining a finding well can help a customer weigh it more evenly.

There is also a competing account. Gal and Rucker’s 2018 review argued that the evidence then available did not show losses are, on balance, more impactful than gains. In status quo studies, they argued, losses and gains are tangled up with doing nothing and acting. Declining a repair is the do-nothing option, so some declines may come from inertia. Mrkva and colleagues, in turn, cast doubt on the claim that status quo bias fully explains loss aversion.

Research on message framing gives little reason to lean on loss wording. O’Keefe and Jensen’s 2007 meta-analysis in the Journal of Health Communication reviewed 93 studies of disease prevention messages. Messages stressing the benefits of acting were slightly more persuasive than messages stressing the costs of inaction. The edge was very small and appeared to come mostly from dental hygiene messages.

Gallagher and Updegraff’s 2012 meta-analysis in Annals of Behavioral Medicine, covering 94 studies, found the same direction for actual prevention behavior. These are health studies, so the lesson for a service department is an inference. Stressing losses over benefits looks like a weak lever. The advice below leans on specifics instead: the item, the measurement and the likely cost of waiting, with no fear figure or discount deadline. Deadline pressure carries its own risk, covered in an earlier post on how manufactured urgency can push buyers away.

Key takeaway: In a 2020 study of 17,720 people, loss aversion ran lower among those who knew more about what they were weighing. In prevention messaging, benefit-framed messages have matched or slightly beaten loss-framed ones.

What a Service Department Should Do Differently

Show the finding clearly when the advisor presents it. Give the customer the item, the measurement, what it affects, and a photo or video when the technician has one. J.D. Power’s 2026 U.S. CSI Study surveyed 51,228 owners of 1- to 3-year-old vehicles. Of those, 64% said they would like photo or video evidence with MPI results. Only 26% of mass-market customers and 44% of premium customers said they get it.

In the Cox Automotive study, 63% of dealership service customers who received photos or videos said they are more likely to approve recommended services. Both figures are survey self-reports.

Treat the decline as a decision with a date. When a customer declines, tell them when the dealership will check back and which item it will ask about. The follow-up then arrives as something the customer expected.

Make the second ask about the same item. Name it, and say what waiting is likely to cost this car when the dealership knows. Leave out discounts and fear statistics. The framing research gives little reason to expect loss wording on its own to do better. Numa’s how-to on following up on declined service work covers cadence and wording.

Make yes easy. Put a link that starts booking in the message, or the dealership’s number. If part of a decline is inertia, which Gal and Rucker’s account would suggest, a shorter path to yes may help with that too.

Numa perspective: Someone at the dealership sets up a declined-services campaign in Numa, with a customer list the dealership supplies. By default, Numa texts each customer when they’re enrolled and again seven days later, with a link that starts booking if the dealership has configured one. The dealership can edit the message before the campaign goes live, and when its list includes each customer’s declined service, the edited message can name it.

The Bottom Line: A Second Ask Should Name What Waiting Costs This Car

When an advisor presents a finding, the customer weighs a sure cost today against a cost that may never come. In one 1979 problem, most respondents facing a sure loss took the gamble instead. Applying that to the service lane is an inference, and a decline can be a sound call.

A follow-up may have more to work with when it names the same item, shows the likely cost of waiting, and makes yes easy. Numa sends the follow-up texts, in wording the dealership can edit, to the list the dealership supplies. Give the customer something concrete to say yes to.

Frequently Asked Questions

What is loss aversion?

Loss aversion is the tendency for a loss to feel bigger than a gain of the same size. Kahneman and Tversky’s 1979 prospect theory paper put it as “losses loom larger than gains.” A 2024 meta-analysis of 607 estimates put the average loss-aversion coefficient at about 1.96, so on average, losses weighed roughly twice as much as equal gains.

Why do customers decline recommended service work?

A decline can be a reasonable call for a customer’s budget or timing. Prospect theory suggests another factor, since approving a repair means a sure cost today. In one problem in Kahneman and Tversky’s 1979 study, 92% of 95 respondents chose a gamble over a sure loss. Some researchers argue that part of what looks like loss aversion is a preference for doing nothing, and declining a repair is the do-nothing option.

Is loss-framed messaging more persuasive?

For prevention behaviors, the evidence leans slightly the other way. O’Keefe and Jensen’s 2007 meta-analysis of 93 studies found gain-framed disease prevention messages slightly more persuasive than loss-framed ones, a very small difference. Gallagher and Updegraff’s 2012 meta-analysis of 94 studies found the same direction for actual prevention behavior.

How should a service advisor follow up on declined work?

Name the same item the customer declined and what waiting is likely to cost that car, when the dealership knows it. Skip discounts and fear statistics, and include a link to start booking. In Cox Automotive’s 2025 Fixed Ops and Ownership Study, 69% of dealership service customers said reminders about previously declined maintenance are helpful.

Do photos and videos help customers approve recommended repairs?

Customers say they do. In J.D. Power’s 2026 U.S. CSI Study, 64% of owners of 1- to 3-year-old vehicles said they would like photo or video evidence with inspection results. Only 26% of mass-market and 44% of premium customers said they get it. In Cox Automotive’s 2025 study, 63% of dealership service customers who received photos or videos said they are more likely to approve recommended services.

See how Numa helps your service team follow up on declined work.