← BlogService recoveryGuest research series · 16 October 2026 · 6 min read

The response can hurt more than the problem itself.

The same broken shower can cost you a little or a lot. What decides it is what happens after the guest tells you — and whether they're left wondering if anyone is coming.

11%

of satisfaction was still lost when a reported problem was resolved.

J.D. Power, 2010
20%

was lost when a reported problem was never resolved.

J.D. Power, 2010
26%

was lost when the problem was handled badly — worse than not trying at all.

J.D. Power, 2010
3 in 4

resolution attempts left guests disappointed or indifferent (50% and 23%).

J.D. Power, 2010
RoomCheck device on a balcony table overlooking hills
Close the loop

Every request needs an owner

A guest left unsure whether anyone is coming experiences a second failure on top of the first.

J.D. Power, 201014,400+ guests in Europe
Same problem, three outcomes

The response decides the cost

J.D. Power's European study of 14,400+ guests gives one of the clearest pictures of what happens after a guest reports a problem. Measured against a problem-free stay:

Satisfaction lost vs a problem-free score of 760

1Reported and resolved

≈11% lost

86 points below a problem-free stay. Fixing it doesn't erase it — but it limits the damage.

2Reported, never resolved

≈20% lost

150 points below. The guest spoke up, nothing happened, and they noticed.

3Handled badly

≈26% lost

Nearly 200 points below — a score of 561. A poor attempt did more harm than no attempt at all.

Fewer problems, bigger stakes2025

Only 12% of stays had a problem in 2025, but those guests rated their stay about a third lower than problem-free guests (J.D. Power, 2025).

Reports that went nowhere2011

Only 35% of reported noise problems and 39% of reported Wi-Fi problems were fixed (J.D. Power, 2011).

The second failure

When the fix becomes the problem

"Someone will be up shortly"Betrayal

Nobody comes. The guest calls again and gets someone who knows nothing about it. The broken shower is now the smaller complaint.

One owner, loop closedRecovered

The request has a name against it. It's fixed, and someone checks back: "Is the shower working now?" The guest feels heard, even though something went wrong.

Failed recoveries fuel reviewsTran et al., 2021
RoomCheck device on a hotel room desk
Sorry first

A free drink without an apology does little

Compensation didn't make up for a slow recovery unless it came with a genuine apology (Wirtz & Mattila, 2004).

What "handled badly" means

How guests are treated matters as much as the fix

J.D. Power describes good resolution as listening, empathy, accepting the concern as valid, committing to a fix and following through, a genuine apology and prompt action. Miss those, and the guest experiences a second failure on top of the first.

The double failure
Problem→Broken promise→Betrayal

A second failure pushed hotel guests toward betrayal and even a desire for retaliation (Lee et al., 2021) — and failed recoveries produce more negative word of mouth than the original problem (Tran et al., 2021).

Fair treatment
Problem→Listened to→Loyalty kept

Loyalty after a complaint depended slightly more on how customers were treated than on what they received (Gelbrich & Roschk, 2011).

Across 60 studies, satisfaction with how a complaint was handled drove what customers told others (Orsingher et al., 2010).

Putting it into practice

Close every loop

Most of what makes a resolution go badly is avoidable: nobody owning the request, nobody following up, and promises nobody checked could be kept.

What creates a second failure
  • • Requests passed on verbally between shifts
  • • "Someone will be up shortly" with no follow-through
  • • Compensation instead of an apology
  • • Promising times you can't keep
What closes the loop
  • • Acknowledge immediately and say when
  • • Give every request a named owner
  • • Check with the guest that it's fixed
  • • Apologise genuinely before offering anything else
⅓

lower satisfaction for guests whose stay had a problem in 2025 — 460 against 677 for problem-free stays (J.D. Power, 2025).

RoomCheck device on a coffee bar in a hotel room
Between shifts
Where requests get lost

The handover is where a promise made at 9pm becomes nobody's job at 11pm.

Reading the evidence fairly

What the research doesn't say

Where the evidence is weaker than a catchy headline would suggest.

The breakdown is from 2010

The detailed resolution figures come from a 2010 European study. J.D. Power hasn't published a resolution breakdown since.

Compare percentages, not points

J.D. Power changed its scale around 2022, so the 2010 and 2025 point figures aren't directly comparable.

"A quarter" means handled badly

About a quarter of satisfaction is lost when a problem is handled poorly. Problems simply left unresolved cost about a fifth.

RoomCheck device on a bedside table
Application

How RoomCheck helps

The research is clear that the second failure — the forgotten request, the broken promise — does more damage than most original problems. Preventing it is mostly about ownership and follow-through.

RoomCheck gives every guest request an owner and a status, so nothing depends on a note at the front desk or a message passed between shifts. The guest sees their request has been received, and requests that run long are escalated — which removes the uncertainty that makes a problem feel unresolved.

  • Every request has an owner and a status
  • Nothing lost between shifts
  • Late requests escalate automatically
Bibliography

Industry research sources

Vendor and commissioned research is noted in the text

J.D. Power studies

• J.D. Power (2010). European hotel guest satisfaction study: problems and resolution.

• J.D. Power (2011). North America hotel guest satisfaction study: noise and problem reporting.

• J.D. Power (2025). North America Hotel Guest Satisfaction Index Study.

Academic research

• Tran et al. (2021). Failed service recovery and online word of mouth.

• Lee et al. (2021). Does love become hate or forgiveness after a double deviation?.

• Orsingher et al. (2010). Complaint handling satisfaction: a meta-analysis of 60 studies.

Complaint handling

• Gelbrich & Roschk (2011). Complaint handling and post-complaint loyalty: a meta-analysis.

• Wirtz & Mattila (2004). Recovery speed, apology and compensation.

Keep reading

More from the research series

All posts
Protect your direct rate

Ready to catch guest friction before it hits Booking.com?

Put a discreet RoomCheck device in every room. Keep communication warm, housekeeping accountable, and reviews immaculate.

  • No long-term lock-ins
  • 1-minute room setup
  • One device per room
  • Trusted by 500+ rooms live