Feedback gets treated as good manners: a survey at the end of a phase, an open invitation to share thoughts, a line in the communications plan. Treated that way it produces almost nothing useful, because the thoughts arrive after the decisions and the people with the sharpest observations have learned that sharing them changes nothing.
Section 2.4.2 of The Standard for Project Management describes soliciting and managing feedback as a function a project has to perform, and the word soliciting is doing real work there. Feedback that arrives unprompted is a bonus. Feedback that a project goes out and obtains, at a point where the answer can still change something, is a control, and it is considerably cheaper than the rework it prevents.
Timing decides the value. An observation about a screen layout is worth a great deal during design, something during build, and nothing at all a fortnight before go-live, when the honest response is to write it down for a later release. Every feedback activity should be placed against the decision it informs, which usually means earlier than feels comfortable and earlier than the thing looks presentable.
Know which decision it feeds. Asking people what they think of something produces opinions, which are hard to act on. Asking which of two arrangements would work better on a busy shift produces a decision. Where a project cannot name the decision a feedback exercise supports, the exercise is a consultation ritual and the participants will sense it.
Ask specific people a specific question. Twelve people who do the work daily, asked one focused question, produce more usable material than an open survey to four hundred. The broad survey has its own purpose, which is establishing how widely a view is held, and that is a different job from finding out what is wrong.
Show something concrete. People respond to a thing they can look at, walk through or hold: a prototype, a mock-up, a printed layout, a rehearsal. Descriptions produce agreement, because it is difficult to object to a description, and the objections appear later against the actual object.
Say what will happen to the answer. Who will read it, what decision it feeds, and when they will hear the outcome. This costs a sentence and it is the difference between a genuine response and a polite one, because most people have given feedback into a void at some point and adjusted accordingly.
A telecoms operator rolled out a new alarm-handling tool to its network operations centre: about forty staff across three shifts, replacing a system many of them had used for a decade.
The project had a feedback plan. A survey went out after the pilot, with a good response rate, and the results were positive. Training satisfaction was high. The tool went live on schedule.
Four months later the assurance review found that one shift was handling roughly a third of alarms outside the new tool, using a spreadsheet and the old console. Walking the floor, the reviewer found a single position where the monitor bezels were ringed with handwritten cards, layered two deep: the alarm types the tool grouped badly, the two filters that had to be reset every shift, the sequence that produced a false clear.
Every one of those observations had existed during the pilot. None had reached the project. The survey had asked whether staff felt prepared and whether the training had been useful, which are reasonable questions about a rollout and useless questions about a tool. Nobody had asked which alarms it handled badly, and the one operator who had raised something in week two had received no reply, which had ended the experiment for the whole shift.
The recovery took two half-days. The project's remaining analyst sat on the night shift twice, with the tool open, asking one question: show me the last three alarms where this got in your way. Eleven issues came out of it and seven were configuration changes made inside a fortnight. The grouping problem was real and took a supplier change, which was raised, funded and delivered in the following quarter, and the thing that made the difference on the floor was that the analyst came back and told them which of the eleven had been fixed, which had not, and why.
Report back, including the noes. People will accept that a suggestion was not taken far more readily than they will accept silence, and an explanation takes two lines. A project that never reports back gets one round of honest feedback and courtesy thereafter.
Keep the channel narrow enough to be answered. An open inbox nobody owns is worse than no channel, because it signals a welcome that does not exist. One route, with a named owner and a stated response time, will carry more useful material than five routes that lead nowhere.
For a PMP® candidate, what helps is seeing that feedback is a function to be performed deliberately, so a scenario where users are working around a delivered system is describing feedback that was never solicited at a point where it mattered. A response that surveys satisfaction after go-live has asked a question whose answer cannot change anything. A structured PMP exam preparation course keeps situations in view where the information existed in the organisation and never reached the project.
Try the operator's question this week. Find two people who will use what you are building and ask them to show you the last three times the current way got in their way. It takes an hour, it produces things no survey will, and it establishes with those two people that somebody is listening, which is what determines whether they tell you anything next time.
The most expensive feedback on any project is the observation somebody made early and stopped repeating. Omega's PMP® Exam Preparation works through soliciting information from the people who will live with the result.
Soliciting and managing feedback is one of the project functions in The Standard, published with the PMBOK® Guide Eighth Edition.
Ad · Amazon affiliate link.
A007: The Standard for Project Management vs the PMBOK Guide
A030: The Project Team Is Not the Project Manager’s Staff
A135: Stakeholder Power Is Not the Same as Engagement
A027: Project Manager, Sponsor, Product Owner and Team: Who Does What?
A024: Facilitation as a Core Project Management Skill
PMP and PMBOK are registered marks of the Project Management Institute, Inc.