How do you measure customer satisfaction for consulting services?

Measure customer satisfaction for consulting services by asking every client the same short set of scored questions at the same two points in every engagement — about a third of the way in, and again at close — then tracking the average across engagements. Fixed questions are the whole trick. They are what turn a pile of opinions into a number you can compare with last year, with a colleague, and with the client you are pitching next week.

Most consultancies do the opposite. They write a fresh feedback form per project, send it once, at the end, to the one person who signed the contract — and get back a polite 9 that tells them nothing. This guide covers the version that works: when to ask, who to ask, what to score, how to keep the answers honest, and what to do with the result.

What counts as customer satisfaction in consulting?

Customer satisfaction in consulting is the client's rated judgment of both what you delivered and how you delivered it, collected per engagement from the people who actually worked with you. It is not the same measurement problem as product CSAT, and treating it as one is why so many consulting feedback programmes produce useless data.

Three things make consulting different:

• Tiny sample sizes. A product team gets 4,000 responses a month. You might get eight in a year. Statistical smoothing is not available to you, so every response has to be worth something on its own.
• Relationship bias. Your respondents are people you had lunch with. They will round up, especially if they have to say it to your face.
• More than one judge. The person who signed the contract often saw the least of your work. The team you sat with every day saw the most. Surveying only the buyer measures the relationship, not the delivery.

How do you measure it? A five-step method

This is the whole loop. It takes about an hour to set up and a couple of minutes per client after that.

1. Fix your questions before the first survey. Decide once what you will ask, then never change it. Comparability is the entire value of the exercise, and you destroy it the moment you tailor the form to a client.
2. Decide who answers. List everyone who experienced the work — the sponsor, the product owner, the developers you sat beside — not just the person who pays the invoice.
3. Ask at least twice inside the engagement. Once about a third of the way in, while there is still time to act. Once at close. At least once a year if the work runs longer than that.
4. Score on a 0–10 scale with guidance at the point of scoring. Tell responders what each score signals, so a 10 means world class rather than “you were nice.”
5. Track the average across engagements, not per project. One project's score is an anecdote. Twelve responses across four clients is a track record.

When should you ask a consulting client for feedback?

Ask at three points: informally as you go, formally at one-third, and formally at the close. Consulting-practice advisor David A. Fields recommends exactly these three intervals: an informal conversation with the project sponsor roughly monthly (weekly on short, intense projects), a formal survey about one-third of the way through the engagement, and a second formal survey at the project's conclusion — see his guidance on how and when a consulting firm should solicit feedback (opens in a new tab).

The mid-project survey is the one almost nobody runs, and it is the valuable one. Feedback collected at the end tells you what you should have done. Feedback collected at one-third tells you what to do — while the client is still your client, and while a fix still counts as service rather than an apology.

Long engagements need one more rule. If the work runs past a year, add a formal survey at least once a year, and every six months if you can. Twelve months is long enough for a small irritation to harden into a reputation, and a two-minute survey is a much cheaper way to find it than a non-renewal.

Who should ask the client for feedback?

You should, as long as the client answers alone and in writing. Fields is right about the underlying problem: a client will not tell the consultant who ran the work, to their face, that the work was mediocre. His fix is an executive assistant, a business coach or an independent interviewer. For most consultants that is a fix of the wrong shape, because what softens the criticism is the conversation, not the person asking.

A short survey link removes the conversation without removing you. The client opens it in their own time, scores four fixed questions with the scale explained beside each one, and writes what they want to say with nobody watching. The request comes from you, in your own words, saying you want honest, constructive feedback. The meeting comes afterwards, if at all, to go through the results together.

What should you actually measure?

Score what the client was in a position to observe. A workshop attendee cannot rate how you handled a six-month delivery, and a client you met once a fortnight cannot rate your collaboration inside their team. The question set should follow the shape of the engagement, not the individual client:

Four scored categories is enough. Fields' mid-project protocol runs to 15–20 rating questions, which suits a firm sending an interviewer into a one-on-one conversation; if you are sending a link, four questions and a free-text box will be answered and twenty will not. These are the three fixed question sets used in client satisfaction surveys for consulting work, and they are deliberately not configurable. Everyone on the engagement gets the same four. A responder who did not see a category answers “I don't know”, and that answer is left out of the average.

Add one open question, and count your responses

Scores tell you where to look; free text tells you what happened. Keep one open field, and record two numbers alongside every result: how many people responded, and when the last response arrived. Volume and recency are what make a satisfaction figure credible to somebody who was not there — an 8.4 from 12 responders last month is evidence, and an 8.4 with no denominator is a claim.

Is NPS enough for a consulting firm?

No — not on its own. Net Promoter Score comes from Frederick Reichheld's 2003 Harvard Business Review article “The One Number You Need to Grow” (opens in a new tab), which argued that a single “would you recommend us?” question predicted growth better than long satisfaction questionnaires. It was designed for companies measuring thousands of customers.

Run it across six engagements a year and the arithmetic turns against you: one detractor can move the score by 17 points, and the result still cannot tell you whether the problem was your expertise or your status reporting. Use category scores as your working instrument and treat any single headline number as the summary, not the measurement.

How do you stop clients from just being polite?

Show responders what each score means at the moment they choose it. Left alone, a satisfied client picks 9 or 10 out of courtesy, and your data flattens into noise where every engagement looks identical. Guidance at the point of scoring fixes this more effectively than any wording in the invitation email.

The nudge that works is simply telling the truth about the scale:

• 10 — world class. Worth saving for a special occasion. Are you sure there is nothing to improve?
• 9 — excellent, and genuinely unusual.
• 7–8 — a good job. This is what a successful engagement normally looks like.
• 5–6 — average. Fine, with room to improve — a good moment to leave a note.
• 0 — “I don't know.” Make it a dignified answer, so people stop guessing on things they never saw.

Do this and your scores spread out. In practice most honest results land between 6 and 10, which is a range you can actually manage against — and a 7 you can trust is worth more to you than a 10 you cannot.

How do you turn satisfaction scores into improvement?

Pick the lowest category and work on that one thing until the next survey. A satisfaction programme that only reports is an expensive way to feel judged; the measurement earns its keep when a weak score becomes a named goal with an end date and a plan. If communication came back at 6.4, that is your quarter.

Because the questions never change, the next survey tells you whether the work landed. That is the loop: turn client feedback into an improvement plan, run it, measure again on the same scale, and keep the ones that moved.

How do you prove client satisfaction to the next buyer?

Publish results a buyer can verify without asking you. Selected quotes on a proposal are worth almost nothing, because everyone knows you chose them. What survives scrutiny is a dated result covering a fixed range of engagements, showing the number of responders and the date of the most recent response, that the reader can check at source.

This is what a verifiable proof of client satisfaction is for, and it is the reason the collection method matters so much: you cannot make results checkable if you rewrote the questions each time. The same discipline is what makes client feedback for consultants travel between employers — the track record is attached to the person who earned it.

If your firm is ISO 9001 certified

Documented satisfaction evidence stops being optional. ISO 9001:2015 (opens in a new tab) clause 9.1.2 requires an organisation to monitor customers' perception of the degree to which their needs and expectations have been fulfilled, and to retain documented information as evidence of the results. Dated, standardised survey results with a recorded responder count clear that bar; a folder of thank-you emails does not.

What tools measure customer satisfaction for consulting services?

They fall into three groups, and they answer different questions:

• Generic survey builders (SurveyMonkey, Google Forms, Typeform) give you a blank canvas. Fine if you have the discipline to write one question set and never touch it — but nothing stops you, and nothing about the result is verifiable by a buyer. See the comparison with SurveyMonkey.
• Public review platforms (Trustpilot, G2) rate the company's public reputation, not whether a named consultant delivered. Useful for a firm's marketing, useless in an individual's CV. See the comparison with Trustpilot.
• Purpose-built client feedback tools ship the fixed question sets, the scoring guidance and the verifiable output as one thing, which is the combination this guide has been describing. Feedback Journey costs €67 a year for a single user, with a free plan for the first three surveys — see pricing.

Frequently asked questions

Sources

• David A. Fields, “How and When Your Consulting Firm Should Solicit Feedback” (opens in a new tab) — feedback intervals, why clients soften criticism face to face, and the mid-project rating protocol.
• Frederick F. Reichheld, “The One Number You Need to Grow” (opens in a new tab), Harvard Business Review, December 2003 — the origin of Net Promoter Score.
• ISO 9001:2015, Quality management systems — Requirements (opens in a new tab), clause 9.1.2 — monitoring customer perception and retaining documented evidence.

Start measuring

Set up your first client satisfaction survey in a few minutes — free, no credit card required.

Start free — no credit cardMore guides