Lesson 3 of 8 · 8 min read · intermediate
Cloaking and misleading creatives
Real people, tricked into clicking. Learn what cloaking and misleading ads are, why feeds treat the resulting clicks as invalid, and how to recognise both.
Not all bad traffic is fake. A large share of what gets search feeds closed is made by real people who were misled. The visitor exists, the phone is real, the click is real. But the person clicked because an ad promised one thing and the page delivered another, so their click carries no real interest in the advertiser at the end of the chain.
A shop puts a sign in the window: "Free phones today". People walk in and find a room of brochures for phone contracts. Some leave. Some, confused, pick up a brochure on the way out. The brochure company paid for each brochure taken, believing those people wanted a contract. Nobody was a robot. The count is still wrong.
Misleading creatives
The creative is the ad the operator runs on a traffic source such as Meta, TikTok or a native network. A misleading ad is one whose promise does not match the destination. In search arbitrage the destination is a page of related searches or keywords followed by sponsored results, so any creative promising a specific product, price, giveaway, job or government benefit is likely to mislead.
- False offers: "free", "$0", "everyone qualifies" or "claim yours" when the destination only shows search ads.
- Implied functionality: fake play buttons, download buttons, chat bubbles or system alerts that look like part of the phone, not like an ad.
- Borrowed authority: brand names, logos or government imagery the operator has no right to use, which also raises trademark policy problems.
- Sensational clickbait: shock headlines with no connection to the page's content.
- Missing ad disclosure: a native ad dressed up as a news story so the reader does not realise it is paid.
Google's rules for Related Search on Content address this directly. Its help page says traffic sources used to bring people to related-search pages must be relevant to, and accurately describe, what the user will see on the destination page. Since 1 November 2025 publishers have also had to pass the exact text of the upstream ad to Google in the referrerAdCreative parameter whenever the traffic comes from a source they control, so Google can compare the promise with the page.
Cloaking
Cloaking means showing one version of an ad or page to the platform's reviewers and automated checks, and a different version to ordinary users. The reviewer sees a harmless article. The public sees the aggressive version that would never have been approved. Every major ad platform bans it, and it is banned in search-feed policies too, because it defeats ad review by design.
This lesson will not describe how cloaking is built. What matters for a reader is how to recognise that it is happening around you, because it is often done by a partner, an affiliate or a "white-label" supplier, and the account that gets closed may be yours.
| Red flag | What it suggests |
|---|---|
| A supplier offers "aged" or "pre-warmed" ad accounts that "do not get banned" | They expect bans, which means they expect to break rules |
| The creative text you report to the feed differs from what users are seen sharing or complaining about | Something different is being shown to the public |
| The page behaves differently depending on where you open it from | Content is being switched by visitor type |
| Very high CTR on the ad, very short time on the page, many complaints | People expected something the page does not offer |
| A partner refuses to show you live creatives or lander URLs | You cannot vouch for what you cannot see |
Deception on the lander itself
Misleading does not stop at the ad. A page can also push people into ad clicks. Policies on AdSense for Search and RSOC prohibit encouraging clicks, disguising ads as navigation or content, and placing ads where they will be hit by accident. Forced keywords and keyword stuffing are related abuses: the page sends the feed a high-paying search term that has little to do with what the visitor came for, so that expensive ads appear. The visitor did not search for that term in any real sense, which is why it is sometimes classed with query fraud.
Compliant journey
- Ad says: "Compare home insurance options"
- Page is an article about choosing home insurance
- Related searches are about home insurance
- Sponsored results are clearly labelled as ads
- Visitor who clicks is plausibly shopping for insurance
Misleading journey
- Ad says: "Government gives free roof repairs, check eligibility"
- Page is a thin article with no programme or form
- Related searches are unrelated high-paying terms
- Ads are styled to look like the next step of an application
- Visitor who clicks is looking for a benefit that does not exist
Why this counts as invalid, and why consumer law cares
For the advertiser, a misled click is as useless as a bot click, and often more expensive because it comes with a confused phone call or a complaint. Search engines therefore deduct for it, restrict the publisher, or close the feed. In 2025 Google tightened RSOC considerably. Trade press reported that from August 2025 many RSOC publishers were placed on restricted features, including a limit on the number of related search terms and blocks per page, with restrictions lifted case by case for publishers in good standing.
There is also a legal layer. In the United States, deceptive advertising is unlawful under the FTC Act, and regulators in the EU, UK, India and elsewhere have equivalent consumer-protection rules. A policy violation costs you a feed. A deceptive ad aimed at vulnerable people, about debt, benefits or health, can cost far more.
Key takeaways
- Clicks from real people who were misled are treated as invalid, because they carry no genuine interest for the advertiser.
- A misleading creative promises something the destination page cannot deliver; cloaking hides the real ad or page from reviewers.
- Google requires RSOC traffic sources to describe the destination accurately and, since 1 November 2025, to pass the upstream ad text via referrerAdCreative.
- Forced or stuffed keywords misrepresent what the visitor was looking for and are a policy breach.
- Deception risks more than the feed: consumer-protection law applies to the ads themselves.
Questions people ask
What is cloaking in advertising?
Cloaking is showing one version of an ad or landing page to a platform's reviewers and a different version to real users, so that content which would be rejected gets approved. It is prohibited by Google, Meta, TikTok, native networks and search-feed policies. Accounts and feeds caught cloaking are normally closed, and revenue already earned can be withheld.
Why are misleading ads a problem for search arbitrage?
Because the feed's advertisers pay for clicks from people who want their product. If the visitor arrived expecting a free gift or a government payout, their click on a sponsored result is worthless to that advertiser. Search engines detect this through poor conversion rates and creative checks, then cut payouts, restrict features or terminate the feed.
What is referrerAdCreative in RSOC?
It is a parameter in which a publisher tells Google the exact text of the ad that brought the visitor to a Related search for content page. Google's help page says that from 1 November 2025 it must be provided, precisely and completely, whenever the traffic comes from a source under the publisher's control. It lets Google check the ad against the page.