What is feedback management in ITIL 4? Owners, metrics and the loop
Definition
Feedback management in ITIL 4 is the collection, analysis and action on input from users, customers and stakeholders, and it is not one of the framework’s 34 named management practices. ITIL 4 distributes the work instead: service level management and the service desk collect feedback, relationship management owns stakeholder satisfaction, and continual improvement acts on what the feedback says through the continual improvement register.
Why feedback management matters: the business cost of feedback nobody acts on
Feedback that nobody acts on costs twice. Users lose productivity to problems that are never fixed, and then they stop reporting problems at all.
The Service Desk Institute found in its 2017 benchmarking data that 97% of IT service desks collect customer feedback, so collection is rarely the constraint. Only about half use it for improvement activity, and one organization in twelve does nothing with what it gathers. Its later benchmarking put the share of service desks running regular employee satisfaction surveys at 67% in 2023.
The 2026 Global IT Benchmark from HappySignals and ISG puts perceived lost time at 3 hours 18 minutes per IT incident. HappySignals puts the user-side productivity loss at roughly EUR 160 per incident. The loss is not spread evenly: HappySignals’ 2025 benchmark analysis finds 13% of tickets account for 80% of total productivity loss, so an average satisfaction score describes the wrong population.
97%
of IT service desks collect customer feedback, but only about half use it for improvement
3h 18m
of perceived lost time per IT incident
13%
of tickets account for 80% of total productivity loss
Source: Service Desk Institute, 2017 benchmarking data; HappySignals and ISG, Global IT Benchmark 2026; HappySignals, 2025 benchmark analysis
Organizations that use experience data consistently report 24% higher end-user happiness and 26% more productivity, on HappySignals’ own figures. Those organizations act on what the data says.
Which ITIL 4 practices own feedback management?
Five ITIL 4 practices own a piece of feedback management: service level management, service desk, relationship management, service request management and continual improvement. Nobody owns it end to end.
Who owns feedback management in ITIL 4: the ownership map
Two rows above are routinely misread. Relationship management is one of ITIL 4’s 14 general management practices, not a service management practice, which matters because satisfaction sits a tier above the service desk. ITIL 4 groups its 34 practices as 14 general management, 17 service management and 3 technical management. Business analysis appears in the last row only to be ruled out.
Who is responsible for collecting customer feedback in ITIL?
The service desk collects feedback from users, and service level management is responsible for collating and analyzing customer feedback alongside operational and business metrics. ITIL 4 asks service level management to identify measures that truthfully reflect the customer’s actual experience, and notes those measures can only be learned directly from customers.
How do you collect user feedback in IT service management?
ITIL 4 names five feedback channels: post-incident and post-request surveys, direct customer engagement, periodic service reviews, feedback logged as a service request, and prototypes tested during service design.
Those five sit in four different practice sections, which is why most organizations operate exactly one of them and call it feedback management.
Each channel misses a different population. The post-ticket survey reaches only the people who opened a ticket, and a service review only the customer who signs the agreement. Neither reaches the user who found a workaround and never called. ITIL 4 separates immediate event-based follow-up surveys from reflective periodic surveys of the overall experience. Its service design guidance adds the one channel that runs before the money is spent: a prototype gathers feedback and improves an idea before anything is built.
Ticket data is the channel most teams already own and never read. Reassignment is the clearest example. In the 2026 benchmark from HappySignals and ISG, happiness runs at +83 on a first-time fix with 2 hours 7 minutes of lost time. It then falls through +78, +69, +59 and +49 to +42 by the fifth handoff, where lost time reaches 10 hours 35 minutes. Tickets with two or more reassignments are 13.3% of volume.
+83
happiness on a first-time fix, with 2 hours 7 minutes of lost time
+42
happiness by the fifth handoff, where lost time reaches 10 hours 35 minutes
13.3%
of tickets carry two or more reassignments
Source: HappySignals and ISG, Global IT Benchmark 2026
How does the feedback loop work in ITIL 4?
The feedback loop runs through two Service Value Chain activities rather than through any practice. Engage produces improvement opportunities and stakeholders’ feedback, and improve takes that feedback as an input.
That output and input pair is ITIL 4’s closed loop. The framework states it plainly; no organization chart shows it, because activities do not appear on one.
ITIL 4’s guiding principle progress iteratively with feedback says that in a well-functioning organization feedback is actively collected and processed along the value chain. But nobody is appraised on whether the loop closes.
PeopleCert’s ITIL (Version 5) confirms the pattern rather than breaking it. It adds no feedback practice. It does add a dedicated ITIL Experience certification module built around experience as anticipation, perception and evaluation over time. Transition guidance names experience level management and Experience Level Agreements (XLAs) alongside service level management. Feedback still has no practice of its own in ITIL (Version 5), which develops it as a discipline with a dedicated certification and its own vocabulary.
What is a continual improvement register (CIR)?
A continual improvement register is the database or structured document ITIL 4 organizations use to track improvement ideas from identification through to final action. It is the artifact that turns a comment into something with an owner, a status and a review date. Without one, an organization collects measurements that never reach a decision.
Nowhere in the continual improvement practice section do the words feedback or satisfaction appear. It is the destination for feedback, not its collector, so the handover into the register has to be designed, or it will not happen.
What is the difference between CSAT, NPS and CES in IT service management?
Customer Satisfaction (CSAT) measures how satisfied someone was with one interaction, Net Promoter Score (NPS) measures whether they would recommend the provider, and Customer Effort Score (CES) measures how hard they had to work to get the outcome.
Giva puts a good CSAT in the 75% to 85% range. All three scores describe a single transaction, which is narrower than the service a user actually receives.
HappySignals’ benchmark shows what that costs you. Its happiness measure is an NPS-style score from minus 100 to plus 100. For 2025 the IT support ticket touchpoint scored about +81, while the service portal scored +33, laptops and computers minus 6, and mobile devices minus 24.
+81
happiness at the IT support ticket touchpoint in 2025
+33
happiness with the service portal
-24
happiness with mobile devices, against minus 6 for laptops and computers
Source: HappySignals, Global IT Experience Benchmark 2025
ITIL 4 says service level agreements should relate to defined outcomes rather than simply operational metrics, achieved through balanced bundles of metrics such as customer satisfaction and key business outcomes. The framework also warns that pushing service desk staff to keep calls short can damage both customer satisfaction and resolution times: a feedback metric turned into a target destroys the thing it measures. Service level management assembles that bundle; the XLA comparison belongs there.
What is the watermelon effect in IT service management?
The watermelon effect is a service that looks green on the outside, because every service level target was met, and is red inside, because the customers receiving it are dissatisfied.
ITIL 4 names it in its own guidance rather than leaving it to vendors, and puts the point more sharply than most commentary does: the provider frustrates customers twice, first with the service and then by not noticing.
The 2026 benchmark from HappySignals and ISG supplies the mechanism. Among positive incident indicators, only 6.6% relate to solving the ticket. Speed of support accounts for 29.2%, service personnel’s attitude for 20.5% and their skills for 19.0%, so those three together carry 68.7%.
6.6%
of positive incident indicators relate to solving the ticket
29.2%
of them relate to speed of support, the largest single factor
68.7%
is carried by speed, attitude and skills together
Source: HappySignals and ISG, Global IT Benchmark 2026
What is a good response rate for IT service desk satisfaction surveys?
Response rates run around 22% for post-incident surveys and around 17% for post-request surveys, according to the 2026 Global IT Benchmark from HappySignals and ISG.
Four in five requests therefore produce no answer at all. Of those who do respond, roughly 30% leave an open-text comment on an incident survey and roughly 21% on a request survey. The part that tells you what to change comes from a small fraction of a minority.
22%
respond to a post-incident survey
17%
respond to a post-request survey
30% / 21%
of those respondents leave an open-text comment, on incident and request surveys
Source: HappySignals and ISG, Global IT Benchmark 2026
A satisfaction percentage calculated on that base is precise and nearly uninformative. Response rates, sampling and non-response bias are a measurement and reporting problem.
The usual intervention, surveying more often, raises volume rather than representativeness; at these levels a cadence and sampling policy does more than another reminder email. The one signal that needs no response is the ticket text, which Matrix42 Intelligence scores for sentiment on incoming tickets rather than on the minority who answer a survey.
Key takeaways
- Feedback management is an activity, not an ITIL 4 practice ITIL 4 names 34 management practices and none of them is feedback management. The phrase appears nowhere in the Foundation publication, so there is no single practice guide to follow.
- Five practices own pieces, none owns the loop Service level management and the service desk collect, relationship management holds satisfaction, service request management logs complaints, and continual improvement acts. No practice is accountable for the whole.
- The loop is wired into the value chain Engage produces stakeholders’ feedback for improve, and improve takes it as an input. That pair is ITIL 4’s closed loop, and because it is an activity flow rather than a practice, nobody is appraised on whether it closes.
- ITIL 4 names the watermelon effect itself The Foundation publication warns that reports can read green while customers are dissatisfied and frustrated that the provider has not noticed. Meeting every target is not evidence of a good service.
- Satisfaction data is thinner than the score suggests About 22% of users answer a post-incident survey and 17% a post-request survey, and only around 30% and 21% of those leave a comment. The number is precise; the signal behind it is small.
Hear the problem before it becomes a ticket trend
The feedback that would have changed your service is rarely in the survey. It is in the ticket somebody reopened, the request nobody finished, and the colleague who stopped asking.
Matrix42 Service Management rates closed incidents in the Service Store Portal, captures feedback on knowledge articles and uses it to improve article quality, and reads sentiment on incoming tickets through AI Assist for Service Desk in Matrix42 Intelligence. Configurable dashboards track service level and performance data next to those signals, so a theme in the comments can be checked against what the tickets actually did.
The boundary is worth stating plainly: these are platform features rather than a packaged experience-management suite, and a continual improvement register is a workflow you configure, not a module you switch on. Matrix42 Enterprise holds PinkVERIFY certification on 11 IT service management processes, and feedback management is not one of them, for the same reason it has no practice guide. See how the practices connect in IT service management.
Closing the loop starts with naming an owner for each feedback channel and recording what each one last changed.
FAQs
Related articles
Sources
- 1 AXELOS, “ITIL Foundation: ITIL 4 Edition (AXELOS / PeopleCert),” 2019.
- 2 ITSM.tools, “The 34 ITIL 4 Management Practices Explained: Full List and Purposes,” 2026.itsm.tools/34-itil-4-management-practices
- 3 PeopleCert, “New ITIL Explained for Certified Professionals: ITIL (Version 5),” 2026.itil.com/Itil-News-and-Announcements/itil-version-5-explained
- 4 PeopleCert, “Evolving from ITIL 4 to ITIL (Version 5),” 2026.itil.com/Itil-News-and-Announcements/Evolving-from-ITIL-4-to-ITIL-Version-5
- 5 PeopleCert, “ITIL Experience (Version 5) certification,” 2026.itil.com/professionals/certifications/itil-experience-version-5
- 6 HappySignals and ISG, “The Global IT Benchmark 2026,” 2026.happysignals.com/global-it-benchmark-2026
- 7 HappySignals, “Global IT Experience Benchmark,” 2025.happysignals.com/global-it-experience-benchmark
- 8 ITSM.tools, “Global IT Experience Benchmark findings,” 2025.itsm.tools/global-it-experience-benchmark
- 9 Joe the IT Guy, citing SDI Service Desk Benchmarking Report 2017, “Are IT Support’s Customer Satisfaction Surveys Their Own Worst Enemy?,” 2018.joetheitguy.com/are-it-supports-customer-satisfaction-surveys-their-own-worst-enemy
- 10 Service Desk Institute, “Service Desk Benchmarking Report V.10,” 2024.servicedeskinstitute.com
- 11 Giva, “CSAT vs. NPS vs. CES: Key Differences and When to Use Each,” 2025.givainc.com/blog/csat-vs-nps-vs-ces