How to Write Reviews That Actually Get Noticed
Name something concrete: a feature that surprised you, a step that confused you, a page that loaded slowly. Generic praise reads as filler and scores poorly. Structure it as impression, one thing that worked, one that did not, then who it is for.
The difference between a review that scores well and one that gets flagged is almost never writing ability. It is whether the review contains evidence that you were actually there.
Be Specific, Not Nice
The best reviews name something concrete: a feature that surprised you, a step that confused you, a page that loaded slowly. Generic praise reads as filler and scores poorly.
This is the whole principle, and it is worth seeing side by side:
Flagged: “Really clean, professional site. Easy to navigate and the design is modern. I would recommend it to anyone looking for a tool like this.”
Scores well: “Signed up with Google in about forty seconds. Creating the first invoice was obvious — the button is right where you expect it. Setting a default currency was not: it is under Settings > Regional, three levels down, and I looked in the invoice editor twice first. The /pricing page also does not say whether Starter includes recurring invoices, which was the one thing I needed before paying.”
The second is not better writing. It is a record of contact with an actual product. Every sentence contains something checkable, which is exactly what verification is looking for.
A useful test before submitting: could this have been written by someone who never visited? If yes, rewrite it.
Structure That Works
Open with your honest first impression, describe one thing that worked and one that did not, then close with who you think the product is genuinely for.
Expanded, that is five moves:
- Context in one line. Who you are relative to this product, and what you were trying to do. “I came at this as a freelancer looking for invoicing” tells the advertiser how to weight everything that follows.
- What you actually did, in order. The path you took.
- One thing that worked, named specifically.
- One thing that did not — the exact step, what you expected, what happened instead.
- Who it is for, and who it is not.
That last line is the single most useful sentence in any review, and most reviewers skip it. “Good for solo freelancers, probably frustrating for a team that needs approval workflows” is worth more to an advertiser than three paragraphs of praise, because it tells them something about positioning they cannot get from analytics.
The one-thing-that-did-not-work rule
Reviews that mention a real drawback are trusted more than uniformly positive ones. Naming a genuine weakness is the fastest way to earn a reader’s belief in everything else you said — and it is the part advertisers actually act on.
You are not being asked to manufacture criticism. If genuinely nothing went wrong, say what nearly did, or what you had to think about. There is almost always a moment of friction worth reporting.
What Gets Flagged
Reviews that could have been written without visiting the page get flagged. So do reviews that contradict what is actually on the site. Verification is looking for evidence you were really there.
Concretely, the common causes:
- Generic praise that would fit any product in any category. The most frequent single reason.
- Describing features that do not exist, or a flow that works differently than described. The clearest possible evidence the visit was not real.
- Reviewing the wrong page. If the brief asked about pricing and you wrote about the homepage, it fails regardless of quality.
- Reused text across campaigns. Caught by cross-account pattern detection, and the risk is to your whole account rather than one submission.
- Impossible timing — a review describing a signup flow completed in eleven seconds.
What is not flagged: negative sentiment. Critical reviews pass verification at the same rate as positive ones. Advertisers specify what to evaluate, never what to conclude. If a brief tries to tell you the conclusion, that breaks platform rules and is worth reporting.
Before You Write
Most review quality is determined before you open the text box.
Read the brief first. If you cannot tell what a good submission would look like, skip the campaign rather than guessing.
Actually use the thing. Complete the flow the brief asks about. Fifteen focused minutes produces material that thirty minutes of writing-around-the-edges cannot fake.
Notice where you get confused. That moment is the most valuable and most checkable thing you will write about, and it is easy to forget by the time you start typing. Note it as it happens.
Pick campaigns you can genuinely evaluate. You will write a far better review of something in a category you know. Domain familiarity reads as credible because it is.
Earning More Over Time
Consistency compounds. Publishers with a track record of high-scoring reviews get matched with higher-reward campaigns automatically.
That mechanism is worth understanding because it changes the optimal strategy. Since access to better campaigns is gated on approval history, a high approval rate on fewer reviews beats a low one on many — rejected work costs the same time and earns nothing.
The pattern among top earners is consistent: they review fewer campaigns than average and write far more about each. Depth pays better than volume, and once you see the reward tiers that stops being surprising.
Two more habits:
Read your rejections. A rejection arrives with a reason attached. Most reviewers who plateau are repeating one fixable mistake, and the reason names it.
Expect a slow month one. Reward tiers and campaign access build with history. Almost every experienced publisher describes the first month as slow and the temptation to quit as real — the curve is genuinely back-loaded.
Frequently Asked Questions
How long should a review be?
Long enough to contain specifics, no longer. A concise review naming three checkable details beats a long one that names none — length alone is not what verification measures.
Will a negative review get rejected?
No. Sentiment is not a verification criterion. A well-argued criticism is more useful to a serious advertiser than vague praise.
Can I review the same product for two campaigns?
You should not reuse text. Pattern detection catches templated submissions across campaigns, and the consequence is account-level rather than submission-level.
What if I genuinely liked everything?
Say what nearly went wrong, or what you had to stop and think about. Uniformly positive reviews are trusted less by readers and are less useful to advertisers.
Related Reading
What makes a trustworthy review, really? covers the reader’s side of this. How verification keeps reviews honest covers exactly what is checked, and how to earn rewards for reviews covers the earning strategy in full.
Further Reading
