"It doesn't feel modern": how to tell actionable client feedback from opinion

Why the distinction matters

"It doesn't feel modern." "Can we make it pop?" "I don't love it." Most designers, strategists, and developers have heard some version of these. They sound like direction. They look like something to fix. But if you actually tried to change the work based on the words alone, you wouldn't know where to start.

From our project work, the single most expensive mistake we see teams make with client feedback is treating every comment as actionable. A project manager or developer nods, takes notes, and an hour later the designer is pushing pixels trying to solve for "more modern" without knowing what that word is doing in the client's head. The team ships a revision, the client looks at it, says "still not quite right," and the cycle repeats. Two weeks burn. The problem was never the work. It was that nobody asked the right question first.

The fix is a basic skill: before acting on any client comment, decide what kind of comment it is.

 

The 60-second diagnostic

Run these two questions against any client comment:

- Does it point to something specific and observable (a page, a moment, a metric, a measurable thing)?

- Can I propose a next step without asking another question first?

Two yeses → it's Feedback. Act on it. The client has given you a useful input and you already have enough to move.

Any no → it's Opinion. That doesn't mean it's wrong. It means it isn't finished yet. Something needs to happen (usually a clarifying question or a data pull) before the comment can become action.

The test takes about 60 seconds and saves days of wrong-direction work.

 

Running the diagnostic on three real client comments

"It doesn't feel modern"

Is it specific and observable? No. "Modern" is a word whose meaning shifts by industry, decade, and personal taste. In design history, "modern" originally refers to the Mid-Century moment in the 1940s to 60s. A client using it today almost certainly doesn't mean that.

Can we propose a next step? Not without guessing. "Modern" might mean cleaner typography, more whitespace, more motion, warmer imagery, or a competitor the client saw last week and wants the work to echo.

Verdict: Opinion. Before making changes, we ask: "What specifically feels dated? Is it the typography, the color palette, the imagery, the interactions, or the overall layout?" Often the answer reveals the real request was about a single element, not the entire direction.

 

"We want the marketing to match our audience better"

Is it specific and observable? Partially. There is an audience and there is a marketing plan, but "match better" has no shape yet.

Can we propose a next step? No. We'd need to know which audience (existing or aspirational), what "match" means in this context (message, channel, creative tone, targeting), and what's failing now that prompted the comment.

Verdict: Opinion. The useful next step is to ask for data (most recent campaign performance, audience segment reports, any customer research that's been done) and to have a 30-minute working session on what "better" should look like in measurable terms before touching the plan.

 

"The app is slow"

Is it specific and observable? Yes. Load times are measurable, devices are measurable, specific pages are identifiable.

Can we propose a next step? Yes. Run performance diagnostics, identify the bottleneck (server response, image sizes, script execution, network), and prioritize the fix.

Verdict: Feedback. We still ask clarifying questions (which page, which device, which time of day), but those questions are narrowing the scope of the fix, not redefining whether there's a problem to fix in the first place.

 

Handling each type, with language that actually works

When it's Feedback

Confirm understanding before acting. This is the step most teams skip, and it's the one that catches the 20% of the time you misheard the comment.

"So if I'm hearing you right, the issue is that product pages take more than four seconds to load on mobile. Let me run diagnostics and come back with the top three contributors and a fix plan by end of week. Does that match what you're seeing?"

One sentence confirming, one sentence on the next step, one check-in question. Then do the work.

 

When it's Opinion

Acknowledge the intent, ask two or three clarifying questions, and offer data before making changes. The goal isn't to push back. It's to turn the comment into something you can both act on without regret.

"I want to make sure we nail this. When you say it doesn't feel modern, is there a reference (a site, a brand, a specific element) that represents what you're looking for? And is there anything on the current version that feels particularly dated? Before we start revising, I can also share the user testing data on the current layout so we're reshaping from the same starting point."

Notice what this does: it treats the client's perspective as real (it is), asks for a concrete anchor, and brings data into the conversation without making it feel like an argument.

 

A third category worth naming: Directives

Some client comments aren't observation or perspective. They're instructions. "Make the button blue." "Move the form to the top." "Cut the third section."

Directives feel like Feedback because they sound specific and actionable. They aren't quite. A directive answers the question "what to do," but skips the question of whether doing it will actually solve the underlying need. Sometimes it will. Sometimes the directive is a client's hypothesis about what they want, and acting on it produces a version they don't actually like either.

When we get a directive, we treat it as a fast-follow conversation: "Happy to move the form to the top. Before we make the change, are we trying to lift form completion, reduce bounce, or make something specific more visible? That'll help me check whether the move lands the way you're hoping." If the answer is clear, act. If it isn't, the directive is really an Opinion wearing instruction's clothes.

 

When acting on an Opinion is actually the right call

The framework isn't "always push back." Sometimes acting on an Opinion is the right move, and knowing when is part of working well with clients.

- When the client is the final decider on taste. Founder-brand alignment, category aesthetic choices, language tone: at a certain point the client owns the call. Our job is to bring craft and pattern recognition, not to keep converting every preference into a research question.

- When the client has domain knowledge we don't. If the client says "I don't think that message will land with enterprise security buyers" and they've sold to that buyer for ten years, their Opinion is worth a lot more than our data model. Act on it.

- When the stakes are low and the cost of iteration is cheaper than the cost of the meeting. If a revision takes an hour and clarifying takes a week of back-and-forth, sometimes the right move is to ship a version, let the client react to something concrete, and move forward.

The skill is telling these cases apart from the ones where acting on Opinion will send the team down a path that gets unwound two weeks later.

 

Upstream setup: how to get more Feedback and fewer Opinions

Opinion-to-Feedback conversion isn't just reactive. It's how we set up the engagement.

A few habits that materially shift the ratio:

- Ask for feedback against specific artifacts, not the overall work. "What's your reaction to the homepage hero?" gets a more useful answer than "What do you think of the site?"

- Share success metrics at kickoff. If the client has seen the campaign metrics we're measuring against, their comments tend to reference those metrics rather than personal taste.

- Use a "what we'd change" agenda at every review. Asking the client to name specific things they'd change forces the conversation into observable territory.

- Give reviewers structured questions ahead of the review. "By Thursday, please flag: one thing you'd keep, one thing that feels off, and one area where you'd like to see alternatives." Prepared reviewers produce more Feedback and less "I don't know. It just feels off."

- When Opinions do surface, park them visibly. A running "to investigate" list signals that the comment was heard and is being addressed with care, rather than silently ignored or prematurely actioned.