5 Red Flags to Watch for Before You Advertise

Every one of these takes about a minute to check, and any single one is enough to walk away. They are ordered by how much money they typically cost the people who ignore them.

1. Guaranteed Results

Nobody can guarantee outcomes on an open network. A guarantee is either a bot farm or a refund policy dressed up as a promise.

Those are the only two possibilities, and it is worth understanding why. On a genuine network, outcomes depend on real people deciding to do something — which nobody can promise. The only way to guarantee a number is to control both sides of the transaction, which means the “users” are not users.

The second case is less harmful but still misleading: a “guarantee” that turns out to be a refund clause. That is not a guarantee of results, it is a return policy, and the distinction matters because you have still spent the time.

What legitimate platforms offer instead: transparent pricing per verified action, and the ability to stop. Not a promise about the outcome.

How to check: ask what happens if the guaranteed number is not reached. If the answer is a refund, it is a refund policy. If the answer is “it always is”, walk away.

2. No Verification Layer

If a platform cannot explain how it checks that an action was real, you are buying noise.

This is the most consequential flag on the list, because it determines whether anything you buy has value. A reward attached to an action creates an obvious incentive to produce the minimum that collects it. A platform without active verification does not just tolerate that — it profits from it, because it bills you either way.

A platform with real verification should be able to tell you, without hedging:

• What signals it checks (time on page, interaction depth, content accuracy)
• What happens to a submission that fails
• Whether you are billed for failed work — this is the question that matters
• Whether there is a dispute route for both sides
• What gets an account removed rather than warned

The billing question is the one that reveals the incentive structure. If failed work still costs you money, the platform has no financial reason to catch it. On a well-designed system, spend is only drawn against verified activity, which means the platform only earns when verification works.

How to check: ask “if a review fails verification, am I billed?” Any answer other than a clear no is a no.

3. Opaque Reporting

You should be able to see what you paid for, action by action. Aggregate-only dashboards hide bad news.

Aggregate reporting is not a UX simplification. It is a way of making bad results indistinguishable from good ones. A dashboard that shows “1,240 actions delivered” without letting you inspect them could be describing anything.

What you should be able to see:

• Individual actions, with timestamps
• The actual review text, in full — not a sentiment score
• Which segment each action came from, so you can cut by performance
• What failed verification, and why
• Cost per verified action, broken down by segment

That last one is what lets you manage a campaign at all. Without segment-level cost per verified action you cannot reallocate budget, which means you cannot improve.

How to check: ask to see a sample report before committing. Reluctance is the answer.

4. Prepay Pressure

Large upfront commitments before you have any performance data are a way to transfer risk onto you.

The structure to watch for: a substantial minimum spend, an annual commitment, or a “package” that must be bought whole — offered before you have run anything. The pitch is usually a discount for volume.

The problem is not prepayment itself; it is prepayment before evidence. You are being asked to commit budget against a performance assumption that has not been tested, and the discount is compensation for taking that risk on the platform’s behalf.

Legitimate platforms let you start small. Daily budgets you can set and pause, a modest first campaign, and volume pricing available after you have data — not before.

How to check: ask if you can run a $100 test first. If the structure does not allow it, the structure is the flag.

5. No Pause Button

If you cannot stop spend instantly, you do not control the campaign.

This sounds minor and is not. Things go wrong: a landing page breaks, a brief turns out to be misworded, a segment starts burning budget on nothing. The correct first response is to stop, then diagnose. A platform where stopping requires a support ticket, a notice period, or waiting for a billing cycle has taken that response away from you.

It is also a reliable proxy for the platform’s general posture. Instant pause is trivial to build. Its absence is a choice.

How to check: find the pause control before you fund anything. If you cannot find it, ask. If the answer involves contacting anyone, that is your answer.

The One-Minute Checklist

Before spending anywhere:

1. Are results “guaranteed”? — walk away
2. Can they explain verification, and am I billed for failed work? — must be a clear no
3. Can I see individual actions and segment-level cost? — must be yes
4. Can I start with a small test? — must be yes
5. Can I pause instantly, myself? — must be yes

Two More Worth Knowing

They discourage disclosure. If a platform suggests publishing incentivised reviews without labelling the incentive, it is asking you to break consumer protection law on its behalf. Liability increasingly reaches the brand, not just the publisher.

They let you specify the rating. A platform that lets advertisers buy a conclusion is selling review fraud. The output is worthless as information and actively dangerous to hold — detection has become good, and penalties scale with how well it worked.

Related Reading

Websites that pay for reviews: how to spot the trustworthy ones is the reviewer-side version of this checklist. How verification keeps reviews honest covers what real verification looks like, and what is a website review platform? covers how to evaluate the model itself.