Why Websites Have So Many Ads in 2026

Open a major news site in 2026 and you will see the same pattern almost everywhere: a thin article wrapped in an enormous superstructure of ad slots, sponsored content blocks, autoplay videos, newsletter modals, cookie banners and "you may also like" feeds. Many users assume this is what publishers want. Almost no publisher wants this. The reality is messier, and worth understanding if you want to be more than a confused participant in it.

This article explains why the modern web looks the way it does, who the players are, and what readers and publishers can each actually do. Companion reading: how ad blockers work and adblock detection.

The short version

Average revenue per ad impression has fallen for more than a decade. To keep total revenue roughly stable, publishers show more ads per page. Each ad fires several requests through a long supply chain. The cumulative weight, latency and clutter is what users see and feel. Privacy regulation, especially in the EU, has reduced the value of personalised ads, which has further pushed rates down. Many large advertisers have shifted budget to platforms like Meta and TikTok that capture user attention more efficiently, leaving publishers with worse advertisers at lower rates.

That is the economic engine behind the visual mess.

The supply chain in one paragraph

A brand wants to advertise. It works with an agency. The agency works with a demand-side platform (DSP) that programmatically bids for ad slots in real time. The publisher works with a supply-side platform (SSP) that exposes their slots to DSPs. A data management platform (DMP) provides audience segments. An ad server delivers the creative. Often there is an ad exchange in the middle, and verification tools alongside it. By the time the ad reaches your browser, six to ten companies may have touched the request.

Each one takes a cut. None of them write the article you came to read.

Why each ad is so heavy

A single ad in 2026 is not one HTTP request for an image. It is typically:

• A few requests to the ad exchange to start the auction.
• A handful of bid requests from DSPs.
• A creative request once the auction settles.
• Several verification and viewability tracker requests.
• A click-tracker stub regardless of whether you click.
• Audience segment cookies set or read.

For a publisher with five ad slots per page, multiply by five. For a page with autoplay video at the top, add another twenty requests for the video ad infrastructure.

The page weight implications are dramatic. Across our 25-site benchmark in 2026, blocking the ad chain typically reduces page weight by ~62 percent. Total HTTP requests drop by half. Time to first paint drops by ~41 percent. The article itself, the actual content, was always a small fraction of the total.

Why ads load after the content

This is "cumulative layout shift", and it is one of the more frustrating UX failures of the modern web. The article renders, then a paid block above it loads with no pre-reserved height, pushing everything down. You are halfway through a sentence; suddenly the sentence is elsewhere.

The cause is technical and avoidable. Publishers can pre-reserve the slot's height; many do not, because doing so means committing to a height before the auction settles, which sometimes wastes ad inventory. The user pays the cost.

Why personalised ads got less profitable

For a decade, the bet was that personalised ads would command premium rates because they would be more relevant. The reality has been more nuanced:

• A growing share of users opt out of personalisation where they can.
• EU regulation has restricted what cookies and identifiers can be used for personalisation.
• Brands have learned that personalised ads do not always perform as well as their pitch implied.
• Privacy-respecting browsers (Safari, Firefox) restrict third-party cookies by default.

The net effect is lower personalisation, lower per-impression rates, and publishers compensating with more impressions per page.

Why publishers do not just charge for content

Many do. The Times, the Atlantic, the Wall Street Journal, the FT all have meaningful subscription bases. The reality is harder for smaller publishers because subscription willingness is concentrated in the top brands. Many small publishers tried subscription models and saw signups in the low single-digit percent of their audience. Ads remained the only viable revenue line.

Hybrid models, with ads for most readers and subscriptions for power readers, are the steady-state most publishers have settled on.

What readers can do

A few options, ranked by impact.

1. Use an ad blocker. NovaBlock removes the worst of the experience. Pages load faster, layout stops shifting, your data stops flowing to dozens of intermediaries.
2. Subscribe to publishers you read regularly. Two or three subscriptions for the publishers whose work you actually value is a fair deal.
3. Whitelist the smaller sites you want to support. One click in the NovaBlock popup.
4. Use reader mode. Built into Firefox, Safari and Edge. Strips the page to its content.
5. Accept that not every site is yours to read. When a wall appears and the publisher is unfamiliar, leaving is also fine.

What publishers can do

This is not our domain, but a few directions visibly work better than others in 2026:

• Treat performance as a feature, not a tax. Heavy pages cost real traffic.
• Reserve ad slot heights to eliminate layout shift.
• Reduce the number of third-party scripts. Each one extracts cost from the user.
• Offer a clean subscription. Soft walls with a clear pricing page convert better than hard walls.
• Be honest about the trade. Readers respect publishers that explain the relationship.

Comparison: typical news article in 2016 vs 2026

Pros and cons of the current arrangement

Pros (for the ecosystem)

• Independent journalism still gets funded in some form.
• The marginal cost of an additional reader is near zero.
• Targeting (where it works) helps small advertisers find audiences.

Cons

• The user experience is degraded.
• Smaller publishers feel the squeeze first.
• Privacy is the casualty of the targeting pipeline.
• Trust in the web economy is at a low.

Conclusion

The web looks the way it does in 2026 because the economics demand it, not because publishers enjoy crowding pages with ads. Users have two effective levers: install a blocker like NovaBlock for the daily experience, and subscribe to the publishers whose work they want to keep existing. Publishers have more levers but slower changes to make. Either way, the easiest single thing to do today is take the user-side step: a clean blocker, a calmer web, and a fairer relationship with the journalism you actually choose to support.