The creative review starts. Someone says the blue feels flat. Someone else wants the logo bigger. A third person asks whether the headline can sound more exciting, which is how “Built for Wisconsin manufacturers” becomes “Powering Possibility Together.”
Forty-five minutes later, the work has more comments and less point.
The problem is the question. “Do you like it?” invites taste, politics, and whatever campaign somebody saw at the airport. A useful review asks whether the work does its job.
Creative feedback should diagnose a problem. It should not audition adjectives.
Set the room before you show the work
Send the brief with the creative. Keep it short enough that people will read it: audience, problem, single promise, proof, required action, channel, and hard constraints. If the brief is still changing, the creative is not ready for approval.
Name one decision owner. Invite subject-matter, legal, accessibility, and channel reviewers for the passes where their judgment matters. A dozen equal votes will produce a dozen small compromises and one very tired designer.
Then review in four passes. Do not mix them. A broken claim should not hide behind a typography argument, and a slightly awkward crop should not reopen the strategy.
Pass one: Does it fit the brief?
Ignore colors, crops, and commas for a moment. Check the big decisions.
- Can the intended audience recognize that this is for them?
- Is there one main promise, or did three departments each get a headline?
- Does the proof support that promise?
- Is the requested action appropriate for this stage of the buying decision?
- Would the work still make sense if the logo disappeared?
If the creative fails here, stop. A round of polish will only make the wrong idea more expensive. Go back to the brief or the concept.
This is where a clear brand strategy earns its keep. The reviewer can compare the work with an agreed position instead of a personal mood.
Pass two: Can someone get it quickly?
Now test the communication, not the explanation from the person who made it.
Show the creative at the size and for roughly the time the audience will encounter it. Ask a reviewer who has not sat through the project meetings to answer three questions:
- What is this about?
- Who is it for?
- What should happen next?
Record the answer before explaining anything. If a social ad needs a two-minute tour, the campaign does not have a comprehension problem. It has a launch problem.
Check the hierarchy too. Headline, subject, proof, and action should compete in that order. A decorative badge, stock photo, or giant logo should not win the first glance by accident.
Pass three: Can you defend every claim?
Read the words and pictures as a skeptical customer would. Mark every objective claim, comparison, testimonial, price, deadline, certification, and implied promise.
The FTC’s advertising guidance says ads must be truthful and non-deceptive, advertisers need evidence before a claim runs, and the full context matters, including words, pictures, and omissions. That last part catches a lot of trouble. A smiling patient, a rising chart, or an “average savings” badge can imply more than the body copy says.
Put the proof beside the claim in the review file. Link the study, approved product data, pricing terms, customer permission, or legal language. “The client said it was fine” is not a substantiation file. Agencies have their own duty to check the information used in a claim.
Disclosures need the same treatment. The FTC warns that fine print, buried qualifiers, and disclosures far from the claim may not work. If the qualifier destroys the headline when both are read together, fix the headline.
For synthetic or heavily edited assets, add the disclosure and rights checks from SigServe’s AI content provenance policy. Haunted hands are obvious. Invented proof is worse.
Pass four: Will it survive the actual channel?
Review the exported asset, not only the spacious design file. Open every real version: phone, desktop, email, presentation, print proof, paid-social crop, or landing page.
- Does the subject survive each crop and safe area?
- Is essential text readable on a small screen?
- Do the link, button, form, phone number, and tracking parameters work?
- Is the file size reasonable, and does it load without a layout jump?
- Does the destination continue the same promise?
Accessibility belongs in this pass, not in a someday backlog. W3C’s WCAG contrast guidance explains the 4.5:1 minimum for most normal text and 3:1 for large text. Its alt-text decision tree helps teams decide whether an image needs a description, an empty alt attribute, or a text equivalent elsewhere.
Do not approve a square master and assume the channel versions will behave. Inspect them. The tiny disclaimer and decapitated product usually appear five minutes before launch because nobody opened the final export.
Write feedback that can be used
Each comment should name the pass, the problem, and the consequence. Suggest a solution only when it is a real constraint.
- Weak: “Can we make it pop?”
- Useful: “Comprehension: at mobile size, the offer disappears behind the event name. The viewer may miss why this matters now.”
- Weak: “I do not love this photo.”
- Useful: “Brief fit: the image shows a large corporate team, but the audience is a two-person professional-services firm.”
- Weak: “Legal may have thoughts.”
- Useful: “Proof: we need the source and date for the 38 percent savings claim before approval.”
Keep a parking lot for preferences that do not affect the brief, comprehension, truth, accessibility, or channel. Some preferences can wait until the next brand-system discussion. Some can die peacefully in the comment log.
Use one approval line
The decision owner closes the review with one of three calls:
- Approve: all four passes clear.
- Revise: list the failed pass and the exact issue.
- Re-brief: the audience, promise, proof, or action changed enough that revisions would be guesswork.
This creates a clean record and limits late-stage drive-by edits. It also pairs well with the draft-to-done creative audit: one process measures the hidden work, while this one improves the decision that sends work forward or backward.
The next time someone asks whether the room likes the ad, pull up the brief and start pass one. Taste can have a chair. It does not get the gavel.
Is your creative review fixing the work or sanding off the point?
SigServe can sharpen the brief, approval roles, and review rules before the next campaign collects 47 contradictory comments.