25%
Of 3,796 US hospitals that reported national patient-survey (HCAHPS) scores, 962 had ratings on Yelp at all.
BMJ Quality & Safety 2013
People are the experience.
48 verified figures on how institutions are seen, and how their people learn, stay and grow.
57.0%
Of 244 narrative online hospital reviews, only this share mentioned any domain the national HCAHPS survey asks about.
BMJ Quality & Safety, 20168%
Organisations that consistently capture knowledge from departing retirees, in a global survey of 1,000 organisations.
APQC, 2025> 1 in 3
Feedback interventions that decreased performance, in a meta-analysis of 607 effect sizes. The average effect was positive (d = 0.41).
Psychological Bulletin, 1996Why these numbers
An image is built from inside.
People are the heart of every experience. A visitor remembers the person at the desk, the wait, the answer. So the evidence on reputation and the evidence on staff belong on one page. Software can organise this evidence. People decide what to do with it.
01 · Experience and online reviews · 14
What people say in public.
Reviews are not a survey. They are written by the people who chose to write. Read together, they show what surveys miss.
0.49
Correlation between the share of high Yelp ratings and high HCAHPS ratings, among hospitals with more than five Yelp ratings.
BMJ Quality & Safety 2013
0.31–0.49
Range of correlations between 9,997 online ratings of 146 English NHS hospital trusts (2009/2010) and the national paper survey of patients.
BMJ Quality & Safety 2012
12
Domains of experience found in 16,862 Yelp reviews of 1,352 US hospitals that the national HCAHPS survey does not reflect, such as cost, billing and staff compassion.
Health Affairs 2016
31% · 33%
Share of those hospital reviews giving one star, and five stars. Reviews cluster at the extremes.
Health Affairs 2016
57.0%
Of 244 narrative online hospital reviews, only this share mentioned any domain the national HCAHPS survey asks about.
BMJ Quality & Safety 2016
41.4%
Of the same reviews, the share in which safety or loss of trust was a theme.
11.5%
Share of those reviews raising financing: unexpected costs or difficult dealings with billing.
38.1%
Share of those reviews written by friends or family members, not by patients. Patient surveys do not hear this voice.
BMJ Quality & Safety 2016
16 of 17
US academic medical centres on the 2002 honour roll that advertised to attract patients. Emotional appeal was the most common approach in 122 ads.
82%
US adults who at least sometimes read online ratings or reviews before buying something for the first time; 40% always or almost always do.
Pew Research Center 2016
51%
Review readers who say reviews generally give an accurate picture of true quality. 48% say it is often hard to tell if they are truthful.
Pew Research Center 2016
79% → 42%
US consumers who trust online reviews as much as personal recommendations: 79% in 2020, 42% in 2025 (survey of 1,026 consumers).
BrightLocal 2025
5–9%
Effect on revenue of each additional Yelp star for Seattle restaurants, mostly independents; chains were largely unaffected.
02 · Service waiting · 4
The wait people feel.
Minutes on a clock and minutes in a chair are not the same. Information and courtesy change how long a wait feels.
1,631
Emergency patients surveyed. Actual waiting time did not predict overall satisfaction; waits perceived as shorter than expected, information and staff courtesy did.
5,030
Primary care patients rating their physicians. Longer waits went with lower satisfaction, but time spent with the physician was the stronger predictor.
15.5 vs 27.03 min
Mean actual emergency waiting time with a queue management system, against a control group, in a trial of 236 patients. Satisfaction was significantly higher with the system.
11.9 vs 32.8 min
Mean perceived waiting time in the same trial. With the system, people felt the wait was shorter than it was; without it, longer.
03 · Turnover and onboarding · 8
The first year decides a lot.
New colleagues decide early whether to stay. Structured programmes help, but not every programme works.
17.5%
About this share of newly licensed US nurses leave their first job within one year of starting.
12% vs ~25%
First-year turnover of new nurses in hospitals with established transition programmes, against nearly 25% in hospitals with limited ones.
15% vs 16.7%
Voluntary first-year turnover in hospitals given a new standardised programme, against control hospitals. The difference was not significant.
83%
New nurses still employed at their hospital after one year, across 1,464 nurses in 97 hospitals. Voluntary turnover was 16%, involuntary 1%.
About 20%
Organisations that formally address all four building blocks of onboarding: compliance, clarification, culture and connection.
SHRM Foundation 2010
0.5×–2×
Cost of replacing one employee, as a multiple of annual salary. Gallup calls it a conservative estimate.
Gallup 2019
52%
Employees who left voluntarily and say their manager or organisation could have done something to keep them.
Gallup 2019
11 million
Projected global shortfall of health workers by 2030.
04 · Coaching and feedback · 9
Feedback helps. Not always.
Coaching changes practice. Feedback can lift or lower performance. Leaders on the floor help when they act on what they hear.
> 1 in 3
Feedback interventions that decreased performance, in a meta-analysis of 607 effect sizes. The average effect was positive (d = 0.41).
0.79
Average effect size of feedback on learning across 12 meta-analyses, 196 studies and 6,972 effect sizes; about twice the average educational effect.
0.49 SD
Pooled effect of teacher coaching on instructional practice across 60 studies with causal designs.
0.18 SD
Pooled effect of the same coaching on student achievement. Larger programmes showed only a fraction of the effects of small ones.
26% vs 48%
Share of the variance in skill group explained by practice in a replication with 39 violinists, against 48% in the original 1993 study.
15–27%
Higher safety culture scores in work settings with the most leadership WalkRounds with feedback, against the fewest; 829 settings, 16,797 responses.
BMJ Quality & Safety 2018
24.3%
Respondents in that study who had taken part in a WalkRound and heard back about the actions taken.
BMJ Quality & Safety 2018
23 units
Clinical units randomised to executive walk rounds or not. Unit safety climate did not differ significantly; nurses who attended a round scored higher.
56 areas
Work areas in 20 hospitals in an 18-month management-by-walking-around programme. Taking part went with lower nurse ratings of performance when problems were not fixed.
FierceHealthcare 2015
05 · Knowledge loss and lessons learned · 7
What leaves with people.
Few organisations capture what experienced people know. Even well-funded lessons systems are rarely used when the next decision comes.
8%
Organisations that consistently capture knowledge from departing retirees, in a global survey of 1,000 organisations.
APQC 2025
16%
Organisations in the same survey that do not try to capture that knowledge at all.
APQC 2025
58%
NASA programme and project managers who said current systems did not let them retrieve the right lessons at the right time (115 respondents).
17.5%
Managers in the same survey who had applied a lesson they found through NASA’s lessons learned system.
53%
Managers who found lessons learned useful less than 25% of the time.
16 of 28
NASA project managers, surveyed a decade later, who said they used the lessons system during their projects. Twelve did not use it at all.
2 of 10
NASA centres that fully complied with the agency’s own lessons learned policy.
06 · Employee monitoring · 6
Watching is not growing.
Monitoring is common. The evidence links it to stress, not to better work.
51%
US workers aware that their employer uses technology to monitor them at work (2,515 employed adults, April 2023).
39% vs 22%
Emotional exhaustion in the past month among monitored workers, against workers who are not monitored.
42% vs 23%
Monitored workers who intend to look for a new job within a year, against those not monitored.
70%
UK adults who would find it intrusive to be monitored by an employer (1,012 adults, August 2023).
57%
UK adults who would feel uncomfortable taking a new job knowing they would be monitored; 19% would feel comfortable.
94 samples
Independent samples (23,461 people) in a meta-analysis that found no evidence electronic monitoring improves performance, and found it linked to higher stress.
Personnel Psychology 2023
How we verify
Every figure, traced.
- 1
Open the original
We read the study, report or survey page itself. When only a copy is open, we use the author’s manuscript or a reputable page that names the original.
- 2
Copy, never round
The figure appears as published. Where the source says “about”, so do we.
- 3
Keep the context
Each figure carries its sample, place and year. A 2001 survey of NASA managers is not a 2025 survey of hospitals.
- 4
Drop what cannot be traced
Popular figures that circulate without a findable original are left out, however often they are quoted.
- 5
Date the page
We re-check the sources each time we update the page, and change the date above.
Read the figures with care. Most samples come from the United States and the United Kingdom; published figures for Lebanon and the Arab region are still scarce. A correlation is not a cause. Hover a source to see the full citation.
Cite this page
Use them. Cite them.
You are welcome to use any figure here. Please cite the original study, and link to this page so others can find the context.
Last updated
Citation
Roadyom. (3 November 2026). Institutional experience statistics. https://roadyom.com/en/statistics
Link
<a href="https://roadyom.com/en/statistics">Institutional experience statistics — Roadyom</a>One figure
Each card has its own link. Use the # in its corner.
Go further
From numbers to practice.
RESOURCES Glossary The terms behind these figures, defined plainly.JOURNAL Roadyom Journal Field methods, with their sources.ROADYOM Mirror See what your institution has not seen.ROADYOM Reflection Practice that holds, shift after shift.ROADYOM Refraction People operations, in order.ROADYOM Prism What your team knows, ranked by evidence.
See your own numbers.
Mirror reads what people say about your institution in public, with every source.