Ask for decisions, not opinions. “What do you think?” is a request for an opinion, and an opinion cannot be acted on, which is why it comes back as three paragraphs that leave you exactly where you started. A usable feedback request points at one specific element, states the criterion the brief already set for it, and asks whether that criterion is met. The change is small and it changes everything, because it moves the client from critic to decision-maker and gives them something they are actually qualified to answer.
Why “What Do You Think” Fails
It fails for a reason that has nothing to do with the client being unhelpful.
An open question about a piece of design gives the respondent no standard to measure against, so they measure against the only thing available, which is their own taste and their memory of designs they have liked. That produces commentary. Commentary is not a defect in the client, it is the correct response to the question asked.
It also invites the client into your job. Asked for a general reaction, many people will try to be useful by suggesting solutions, because suggesting a solution feels more helpful than describing a problem. The result is a note like “maybe try a different font” when the underlying observation was “the headline does not feel serious enough for our industry.” The second sentence is genuinely useful information. The first is an instruction to do work that may not fix anything.
There is a second, quieter failure. A vague request produces a vague answer, and a vague answer cannot be closed. You cannot tell whether it has been addressed, which means you cannot tell whether the round is finished, which means the round does not end. That is a scope problem wearing a communication problem’s clothes.
The Question Shape That Works
Three parts, in this order.
- Point at one thing. Not the design. The headline, the color of the primary button, the mark at small size, the third slide.
- State the criterion. Preferably in the client’s own words from the brief. “We agreed this needed to read as established rather than trendy.”
- Ask whether it is met. A yes-or-no question with room for a reason.
Put together: “Looking at the headline treatment only, and thinking about the goal we set of reading established rather than trendy, does it land? If not, what specifically makes it feel wrong?”
The follow-up matters as much as the question. When a client answers, ask why once. Not twice, which reads as interrogation, but once, because the first answer is usually the symptom and the second is usually the cause.
| Instead of | Ask |
|---|---|
| What do you think? | Does the headline read as established rather than trendy? |
| Any feedback on this? | Which of these two directions gets us closer to what we agreed? |
| Do you like the colors? | Will this palette sit alongside your existing packaging without clashing? |
| Let me know your thoughts | Is there anything here that would stop you signing this off this week? |
That last one is worth keeping. It is the single most efficient closing question available, because it surfaces blockers instead of preferences.
Three Questions, Not Thirty
A great deal of published advice on this subject takes the form of a long list, twenty or thirty questions to ask a client about a design. Those lists are useful as reference material for choosing questions. They are not useful as a thing to send.
A client who receives thirty questions does one of two things: skims them and answers generically, or does not reply at all. Either way you are back to the original problem, and now you have also spent the client’s goodwill. Long questionnaires produce exactly the unusable feedback they were meant to prevent.
Three questions per round is a workable ceiling for most jobs. Choosing which three is the actual skill, and the rule is to ask about whatever is most expensive to change later. Structural decisions before surface ones. A layout question outranks a color question, because the color can be adjusted after approval and the layout cannot, not cheaply.
Taste Feedback Versus Brief Feedback
Not all feedback carries the same weight, and separating the two kinds is what stops a round becoming an argument.
Brief feedback says the work does not meet something that was agreed. “We said this had to work in one color and it does not.” That is binding. It is a defect against a stated requirement and fixing it is your job.
Taste feedback says the client would have made a different choice. “I would have gone with something warmer.” That is information, not an instruction. Sometimes it is worth acting on, because the client knows their audience and their own reaction is data. Sometimes it is worth a conversation.
The distinction is not a device for dismissing clients. It is how you decide what a round of changes consists of, and it needs to be visible to both sides. A note logged as “client prefers warmer, discussed, keeping current palette because of the contrast requirement we agreed” is a closed item. The same note left unresolved reappears in round three.
Some taste notes also arrive in language that does not describe anything actionable at all, and the most famous of those has its own diagnosis path in decoding “make it pop” and other vague client feedback.
Giving the Feedback a Deadline
An unbounded feedback window produces feedback in fragments. One note on Tuesday, three more on Friday, a forwarded email from a colleague the following week, each one arriving after you have already started acting on the last. That is not one round. It is an open channel, and it is the most common way a two-round job becomes a five-round job.
Ask inside a stated window, and say what happens when it closes. “If you can get me consolidated notes by Thursday, I will have the revision back to you Monday. If notes arrive after that, they roll into the next round.” That sentence is not aggressive. It is a schedule, and clients generally respect schedules that were stated in advance.
The word doing the work there is “consolidated.” One message containing everyone’s notes, rather than several people replying separately, is worth asking for explicitly. Where multiple stakeholders are involved, that ask is the difference between a round and a committee.
When the Answer Is Still Vague
It will be, sometimes, even with a good question. When it happens, do not guess and do not start editing. Convert the vague answer into a choice.
“You mentioned it feels a bit corporate. Is that the type, the color, or the amount of space around things? If it helps, here are two versions and I would like to know which is closer.”
Two options are usually better than an open question at this stage, because recognizing a preference is much easier than articulating one. This is also the point where asking for a reference is genuinely useful: something the client thinks does the job well, so you can look at it and identify what they are actually responding to.
If vague feedback keeps arriving round after round, the problem is usually upstream. Either the criteria were never specific enough to judge against, in which case the design brief needs revisiting, or the work is being shown without the frame that tells the client what to react to, which is a presentation issue covered in how to present design work without losing the room.
Writing the Request
A feedback request that works is short and has four parts: what is attached, what you want them to look at, the criteria, and the deadline.
Something close to this: here are the two directions from Tuesday’s call. Before anything else, I would like to know which one we develop. Looking at them against the goal we set of reading established rather than trendy, and remembering the mark has to hold up small and in one color, which is closer? If you can come back by Thursday with consolidated notes from your side, the revision will be with you Monday.
That is one paragraph, it names a decision, it repeats the criteria, and it sets a date. It also tells the client what to ignore, which is most of what makes it work.
One thing to keep in mind about feedback on contextual renders: a client reacting to a mockup is reacting to a composite, and they often cannot tell whether it is the artwork or the scene they are responding to. Asking about the artwork specifically is worth doing deliberately, for the reasons set out in presenting in context with mockups. The wider sequence this stage sits inside is in the solo design project process.
FAQ
How do you ask a client for design feedback?
Point at one specific element, restate the criterion the brief set for it, and ask whether that criterion is met. Then ask why once. An open question like “what do you think” produces commentary rather than a decision, because it gives the client no standard to judge against.
What questions should you ask clients about a design?
Three per round, chosen so that the most expensive-to-change decisions are asked about first. Structural questions before surface ones. A useful closing question is whether anything here would stop them signing it off this week.
Why do clients give vague design feedback?
Usually because they were asked a vague question, or because no criteria were stated for them to judge against. When it persists across rounds, the cause is normally an under-specified brief rather than an unhelpful client.
How long should you give a client to give feedback?
Long enough to gather everyone’s notes, short enough that they arrive as one batch. State the window and what happens when it closes, so that late notes roll into the next round rather than extending the current one.
What do you do when a client’s feedback contradicts the brief?
Treat it as information rather than a defect and say so plainly, naming the requirement it conflicts with. Feedback that the work fails an agreed requirement is binding. Feedback that the client would have chosen differently is a conversation.