Insurance apps: US customers trust one to take a premium but not to run a claim, and 57% would refuse an app-only policy at the same price
57% of US insurance customers (114 of 200) would refuse an app-only version of their policy at the same price, and 68% name disputing a claim decision as the moment they most want a person (n=200, October 2026).
- Who was asked
- US insurance customers, 200 clones, US
- When
- October 2026 (fielded 6 October 2026)
- How
- Predicted responses from Brox clones, each built from a long-form interview with a real person.
- What changed from the obvious question
- Instead of asking whether people trust insurance apps, which invites a general yes, the study rated four named tasks from a premium payment to a $5,000 claim, then offered a fully written app-only version of the clone's own policy at the same price, then asked the single point they would want a person and what would most raise trust.
Section 01: What the panel said
- #
57%
57% of US insurance customers (114 of 200) would refuse an app-only version of their current policy at the same price, keeping or finding one with a person to call; 21% would take it happily and 21.5% only if it were cheaper.
base 200
- #
68%
Claims are where a person is wanted: 68% (136 of 200) name disputing a claim decision or settlement as the one point they most want a person, 27.5% (55) filing a claim, and buying, changing or canceling a policy together just 4.5% (9).
base 200
- #
74%
Comfort falls with the stakes: 91.5% (183 of 200) are comfortable paying a premium entirely in the app, 46.5% canceling, 40.5% changing coverage, and 19% filing and tracking a $5,000 claim, which 74% (148) are uncomfortable doing without a human.
base 200 · contrast
- #
83.9%
The refusal is concentrated in older customers: 83.9% of those aged 65 and over (52 of 62) would refuse an app-only policy, against 55.4% of 45-64 (46 of 83) and 30.2% of 30-44 (16 of 53).
base 62 · Aged 65 and over · caveat
57% of US insurance customers (114 of 200) would refuse an app-only version of their current policy at the same price, keeping or finding one with a person to call; 21% would take it happily and 21.5% only if it were cheaper.
02Do customers trust an insurer's app to run their whole policy?
Not on this panel. Offered their own policy at the same price but run only through the app, with staff reachable only by written message, 57% (114 of 200) said they would keep or find a policy with a person to call. Another 21.5% (43) would take it only if it were cheaper, and 21% (42) would take it happily. Almost nobody sat in between: one clone chose "yes, but reluctantly".
The app itself is not the problem. 91.5% (183 of 200) are comfortable paying a premium or updating payment details in the app with no human involved. The trust runs out when money is in dispute.
03Where exactly does the app stop being enough?
At the claim. 74% (148 of 200) would be uncomfortable filing and tracking a $5,000 claim in the app without a human, and asked for the single point they would most want a person, 68% (136) chose disputing a claim decision or settlement and 27.5% (55) filing a claim. Buying, changing and canceling a policy took 9 picks between them.
Canceling is the split case: rated, it divides the panel evenly, 46.5% (93) comfortable and 46.5% (93) uncomfortable. Forced to pick one moment, only 3% (6) chose it, because the claim crowds everything else out.
04Who refuses, and is it only older customers?
Mostly older customers, but not only them. 83.9% of those aged 65 and over (52 of 62) would refuse the app-only policy, against 55.4% of 45 to 64 year olds (46 of 83) and 30.2% of 30 to 44 year olds (16 of 53). Women and men refuse at the same rate, and no income band falls below 50%.
This panel skews older, with 145 of 200 aged 45 and over and only 2 under 30, so the 57% headline would likely be lower on a younger sample. The claim finding does not depend on age in the same way: the dispute is the top pick among 30 to 44s (44 of 53), 45 to 64s (58 of 83) and those 65 and over (33 of 62).
05What would make an app-first policy acceptable?
Asked what would most raise their trust in running the whole policy through the app, 48% (96 of 200) chose a guaranteed call back from a person within one hour, 24% (48) written confirmation of every change and claim decision, and 9.5% (19) a lower price. 13.5% (27) said nothing would.
- The defensible reading: an app-only policy with no phone line is refused by most of this panel, and the refusal sits on the claim.
- The interesting reading, which this study did not test: customers are refusing the loss of the phone, not the app. 71 of the 114 who refused name the one-hour callback as their biggest trust lever, which suggests an app-first policy with a guaranteed person at the claim and the dispute is a different offer from an app-only one.
Section 06: What they said, and the interview line it rests on
“No, I wouldn't do that at all. I would keep or find a policy where I have a person to call. I really hate dealing with insurance companies online or through an app, just sending messages back and forth. When it comes to my insurance and my claims, I want to pick up the phone and talk to a real person. I want to clarify things, get their name, and get a confirmation number from an actual human being, not deal with messaging or robots.”
Predicted from this line in the interview
“I hate dealing with insurance companies online. So I always call because I want to clarify and then like I'll take down their name and like I want confirmation numbers. Like I want to be able to speak to you. I don't want to do an IA robot or like an email back and forth. I want to talk to a real person on the phone.”
“The one point where I would most want to speak to an actual person rather than use an app is disputing a claim decision or the settlement amount. If something happens to my car and they tell me they aren't covering something or there are surprise fees, that gets very confusing and stressful. An app can't properly explain things to you or work through a disagreement. I value good communication and customer care, and when it comes to money, car repairs, and settlement decisions, you need to talk to a real person who can explain the policy clearly and treat you like a human being rather than dealing with an automated system.”
Predicted from this line in the interview
“they don't really tell you the whole policy that you have. And you will really find out something happens to your car and they will tell you that, oh, this and that fee and this fee and we can't do this and we can't do that. It'd be confusing to me. I feel like they're more confusing than anything because they're not really explaining to you the policy.”
“Yes, happily. That is honestly my ideal way of doing things anyway. I prefer to handle everything online and through apps because it's just so much easier. You don't have to deal with driving through traffic or sitting around dealing with people, and to be honest, I really don't like talking on the phone. Messaging through an app and having everything digital is completely fine with me as long as the coverage and price stay where they are.”
Predicted from this line in the interview
“I prefer to do everything online. It's just easier. You're not dealing with a bunch of traffic and people. I'd much rather do it online. It's just easier.”
Section 07: How much should you trust this?
What we would defend
- A majority of this panel, 57% (114 of 200), would refuse an app-only policy at the same price, and the scenario was read as written (167 of 200 reasoning traces cite the same price).
- Claims, and above all a disputed claim, are where customers want a person: 95.5% (191 of 200) pick a claim step over buying, changing or canceling.
- Comfort with app-only handling falls steeply with the stakes, from 91.5% for paying a premium to 19% for a $5,000 claim.
What we would not
- The 57% as a national figure: the panel skews older (145 of 200 aged 45 and over, 2 aged 18-29), and refusal runs from 30.2% at 30-44 to 83.9% at 65 and over.
- That a one-hour callback would convert refusers: 71 of 114 refusers name it as the biggest trust lever, but nobody was offered it and asked again.
- 'Only if it were cheaper' (21.5%) as price elasticity: no discount was named, so it says nothing about how much cheaper.
- Any split by product line: auto, home and renters customers answered together.
The main limitation
A stated hypothetical on an older-skewing panel: no one was actually moved to an app-only policy, and with almost no under-30s the headline share would likely fall on a younger sample.
Section 08: How was this run?
| Segment | Clones |
|---|---|
| Aged 30-44 | 53 |
| Aged 45-64 | 83 |
| Aged 65 and over | 62 |
| Aged 18-29 | 2 |
Question wording as fielded
- Do you currently hold an auto, home or renters insurance policy in your own name or jointly with someone else?
- How comfortable would you be doing each of these with your insurer entirely in its app, with no human involved, on a scale of 1 (very uncomfortable) to 5 (very comfortable)?
- Imagine your insurer offered your same policy at the same price, but run only through its app: you change it in the app, you file any claim in the app, and you can reach the insurer's staff only by written message in the app, never by phone or in person. Would you take that app-only policy?
- At which one point in managing your insurance would you most want to speak to a person rather than use your insurer's app?
- Which one of these would most increase your trust in handling your whole insurance policy through your insurer's app?
What was not asked, and why
No insurer or brand was named, no discount for going app-only was priced, no product line was separated, and the callback was not tested as an offer, to keep the study within six questions.
What changed from the obvious question
Instead of asking whether people trust insurance apps, which invites a general yes, the study rated four named tasks from a premium payment to a $5,000 claim, then offered a fully written app-only version of the clone's own policy at the same price, then asked the single point they would want a person and what would most raise trust.
Fielded through Ask Brox on 6 October 2026.
More on Claims and the moment of truth · Insurance · Insurance (US) panel
Ask this of your own customers: humans@brox.ai
Related insights
All in this topic-
After a car accident: 45% would shop around at renewal if no person calls for 24 hours, against 3.5% after a call within the hour
45% 45.0% would shop around for another insurer at renewal if no person contacted them until the next day after they reported a car accident, against 25.0% after a four-hour wait and 3.5% when a named claims handler calls within the hour. (base 200)
Read -
Nearly half of US drivers would quit an AI-led car claim intake at the opening question, and none would quit at the photos
48% The exit is at the first step, where the app asks them to describe the crash in their own words. (base 200)
Read