SuckButtBecause some things just suck

The Internet Sucks

The first four results are advertisements and the label is grey on grey

Search advertising has to be identified as advertising. Nothing says the identification has to be easy to see.

Close-up of a hand holding a smartphone displaying the Instagram login screen, highlighting mobile technology.
Photograph by energepic.com via Pexels
Editorial note. Independent reporting and analysis. Nothing here is sponsored or paid for. How we work.

What follows is the working version of search result labelling: the decisions in the order you actually meet them, with the reasoning attached.

Before you start

  • Advertising disclosure requirements specify presence more than prominence.
  • Paid and organic results have converged visually over successive redesigns.
  • Studies have generally found many users cannot reliably identify paid results.

Disclosure required, prominence optional

Advertising rules in most jurisdictions require paid placement to be identifiable as advertising, which search engines comply with by labelling it. The rules generally speak to the presence of a label rather than to its size, colour, position or contrast against the surrounding text.

That leaves the design of the disclosure to the party with a commercial interest in it being overlooked, which produces the expected result. Background shading that once separated paid results has been progressively reduced across successive redesigns on multiple platforms. Studies have generally found that a substantial share of users cannot reliably distinguish paid from organic results, which is the outcome the design predicts.

Convergence is the technique

Early search advertising looked obviously different, with a coloured block, a border and a clear heading separating it from the results. Each redesign removed a little of that separation, and no single step was large enough to be worth objecting to.

Once the introductory rate lapses, the endpoint is a paid result rendered in the same typeface, size, colour and layout as an organic one, with a small word attached. This is the same incremental method used in packaging, pricing and terms, where a gradual drift avoids the reaction a single change would cause. It works because nobody compares today's page against the version from six years ago, and the archive is not where the decision is made.

The brand-name auction

Advertisers can bid on searches for other companies' names in many markets, which is why searching for a business can return a competitor first. It also produces the situation where a company pays to appear above its own organic listing, essentially buying traffic it would have received anyway.

Trademark rules on this vary substantially between jurisdictions and have been litigated repeatedly with inconsistent outcomes. For the searcher the practical effect is that the top result for a brand is not reliably that brand, which matters most for logins and support pages. Typing the address directly, or using a bookmark, removes the entire category of risk for anything sensitive.

Malicious advertising exploits the same gap

Because the top result looks authoritative, paid placement has been used to distribute fake support numbers, cloned login pages and counterfeit software downloads. Ad review systems catch a great deal of this and the volume is large enough that some gets through, particularly for high-value search terms. Software downloads, cryptocurrency services, banking logins and technical support are the categories where this has repeatedly caused real harm.

The defence is simple and unpopular, which is never to reach a login page or a download through an advertisement.

For those specific categories, scrolling past the paid block entirely is a reasonable permanent habit.

The results page has eaten the results

Answer boxes, product carousels, maps, related questions and video panels now occupy much of the visible area before any ordinary link appears. Each of these is genuinely useful in some cases and each of them reduces the number of independent sites a searcher ever reaches. The commercial logic is that a searcher who never leaves is a searcher who sees more advertising and generates more data.

Read the terms and there it is: publishers have objected to this for years, and the argument is genuinely two-sided since the panels frequently answer the question well. What is not two-sided is that the boundary between an answer, a link and an advertisement keeps getting harder to see.

Searching more deliberately

Learn where the paid block ends on the engine you use, since it is a consistent position and recognising it takes a week to become automatic. Use bookmarks for logins, banking and support rather than searching for them, which removes the highest-risk cases entirely.

The bit they bury in the confirmation email: try an alternative engine occasionally, because comparing the same query across two engines shows what each one is choosing to show you. Use site-restricted searches when you know the source you want, as it bypasses the ranked page completely. And treat the first screen as a shop window rather than as an index, which is a more accurate description of what it has become.

The takeaway

Bookmark anything you log into, and never download software from an advertisement.

It is not you being fussy. It is genuinely badly made.

Questions readers ask

How do I tell a paid result from an organic one?

Look for a small label near the address line rather than for shading. The position of the paid block is consistent, so learning it is quicker than checking each time.

Why does a competitor appear when I search a brand name?

Bidding on other companies' names is permitted in many markets. Trademark rules differ, and outcomes have been inconsistent across jurisdictions.

The Internet Suckssearchadvertisingdisclosuredark patterns
More in The Internet Sucks
Jhilik Mahapatra
Contributing writer, SuckButt

Jhilik writes about everyday irritations and believes packaging is a conspiracy.

Also by Jhilik Mahapatra