Why Web Media Ad Revenue Stagnates: Dissecting 3 Structural Problems

Web media ad revenue is stalling. There are three structural reasons behind this phenomenon that no amount of on-the-ground effort can fix.

1. A negative spiral in which adding ad slots drives down CPM
2. Declining ad impressions and targeting accuracy caused by ad blocking and privacy regulation
3. The concentration of ad budgets into Google, Meta, and Amazon

I have been working on both sides of web advertising and web media since 2015. At a major publisher's media arm I led monetization and data analysis for large-scale sites, and on the SSP and ad-network side I helped teams monetize through programmatic advertising. I am the person who has sat across from media owners and told them, more times than I can count, "Add one more ad slot here and you'll see roughly this much monthly revenue uplift." I am, in short, an industry insider whose livelihood has been built on advertising.

So why am I now writing about the limits of ad revenue? As someone who has been part of building this structure, I want to lay it out from the inside.

Ad revenue is determined by "impressions × price"

The revenue equation is simple.

Most web media still depend on advertising income for the bulk of their revenue. Even media properties with monthly page views north of ten million often have no meaningful revenue pillar beyond ads.

Ad income is basically determined by "impressions × price." An impression is a single instance of an ad being displayed on screen. In other words, the only ways to grow revenue are to increase page views, increase the number of ad impressions per page, or raise the price.

Of these, growing page views takes time. You invest in SEO, raise the quality of your content, and wait for it to reach readers. It is a six-month, twelve-month commitment. Raising the price is not something the media side can control on its own, either—the number is set in an auction-based ad market and is subject to supply and demand.

That leaves adding more slots.

Add one banner above the article. Add another below it. Drop a rectangle in the sidebar. Slip an in-feed ad into the body. Set video ads to autoplay. Each of these takes only a small code change, and the revenue projections are, as a rule, accurate. Produce a number—"add this slot and you'll see roughly this much per month"—and it will generally be right.

So slots get added.

I have made this "add one more slot" proposal many times myself; the full firsthand account is in Why We Built an AI Voice SaaS for Web Media. When I was on the media side running ad operations, adding ad slots was the quickest way to hit the monthly revenue target. It was not that the person in charge had bad intentions. There was a target number, the means to hit it were right there, and there was no reason not to take it. That is all there was to it.

Reason 1: Oversupply of ad slots pushes down CPM

On top of this sits another unavoidable fact.

When you increase ad slots, the number of ad impressions per page goes up. Impressions go up. But at the same time, ad exposure frequency per user also rises. The same person ends up seeing ads on the same page over and over.

From the advertiser's perspective, showing the same user an ad repeatedly has diminishing returns. Both click-through rate and conversion rate decline from the second exposure onward. So advertisers will only pay a low price for "duplicate impressions to the same user."

In the programmatic market, this adjustment happens in real time. In RTB (real-time bidding) auctions, the value of an ad slot is evaluated instantaneously. What is actually pushing down prices is the easing of supply-demand balance from the increase in total impression volume, declining viewability on slots placed low on the page, bid exclusion triggered by frequency caps, and the advertiser-side CPA/ROAS optimization algorithms that automatically lower bids on slots with thin performance. Adding slots does generate impressions, but all of these factors work in concert to drag down the average price.

In other words, when you add slots to generate impressions, prices fall in response, and revenue ends up growing far less than expected. So you add more slots. The same loop keeps turning, with growing force.

I have written about this loop—"CPM drops so we add slots → UX degrades → dwell time falls → CPM drops again"—with the firsthand account in Why We Built an AI Voice SaaS for Web Media. Both the media side and the ad-tech side are simply making rational decisions to protect their own businesses, but the cumulative result is collectively irrational.

As someone who worked on the SSP and ad-network side, I have seen the structure that hollows out media revenue from the inside. Adding slots does lift revenue in the short term. But over the medium to long term, falling prices eat those gains. Plenty of people on the ground are aware of this structure. The reality is that with quarterly targets staring them in the face, there is no room to be having long-term conversations.

Programmatic concentration is intensifying

There is another figure that accelerates this negative spiral. According to projections from eMarketer, more than 90% of US display ad budgets already flow through programmatic channels. The share of direct deals that bypass auctions has been shrinking year after year.

The advance of programmatic means that price-setting has become more mechanical. There is less room for human judgment of the "this is a high-quality slot, let's pay a premium for it" variety; prices are set instantaneously on pure performance metrics (click-through rate, conversion rate, viewability). Even when media owners argue "our readership is high-quality so we deserve a higher price," it is increasingly unconvincing in the face of machine verdicts.

Reason 2: Ad blocking and privacy regulation are shrinking impressions

There is another pressure that is almost invisible from outside the industry.

Ad blocking adoption. According to GWI research (Q2 2025, compiled by Backlinko), 29.5% of internet users worldwide use ad blockers at least occasionally. In absolute terms that reaches 1.77 billion people. That is a 40-fold expansion in a little over 13 years, from 44 million in Q1 2012 (DataReportal). Even limiting to habitual users puts the figure at around 20%, but either way a substantial number of impressions simply never come into existence.

By country the rates are Indonesia 40.1%, South Africa 35.6%, Vietnam 34.6%, China 33.7%, and the United States 32.5%. In most major countries, roughly 30% of users rely on ad blocking in some form. Japan sits below 20%, on the lower end, but even so one in five people uses some form of blocking.

Privacy regulation erodes targeting accuracy

Even more consequential is Apple's ITP (Intelligent Tracking Prevention). The feature, which restricts third-party cookies in Safari, has been strengthened in stages since its introduction in 2017. In 2020 it began restricting even CNAME-based tracking, fundamentally undermining the accuracy of ad measurement and targeting.

As someone who was working in ad operations at the time, I experienced ITP as a serious body blow. Retargeting performance declined, conversion measurement became inaccurate, and advertiser budgets began shifting away from the open web. In April 2025, Google withdrew its policy to deprecate third-party cookies in Chrome altogether (it also shelved the introduction of a new choice prompt). However, the restrictions imposed by Safari's ITP and Firefox's ETP remain in place, and a state of fragmented measurement environments across browsers has become the norm. With the single hard deadline of deprecation removed, migration decisions have if anything been pushed further down the road.

In Europe, GDPR, and in Japan the revised Act on the Protection of Personal Information, continue to tighten the rules around tracking. The cost of complying with these regulations ultimately becomes another factor compressing media revenue.

Ad blocking shrinks impressions, privacy regulation lowers targeting accuracy, and as a result ad prices fall further still. This double pressure is what now sits on top of the media revenue structure.

Reason 3: Ad budgets are being siphoned into Google, Meta, and Amazon

Looking at the trajectory of CPM (cost per thousand impressions), most media have seen it flat or slightly declining over the past few years.

The causes are compound. As noted above, oversupply of slots is pushing down prices. On top of that, the concentration of market share among platform companies is significant.

According to projections from eMarketer, 2026 is expected to mark a historic turning point. This year Meta is projected to surpass Google in global digital ad revenue for the first time. Meta's net ad revenue will reach $243.46 billion (a 26.8% share), against Google's $239.54 billion (26.4%).

The heavier fact, though, is the combined total. Meta + Google + Amazon together account for 62.3% of global digital ad spending. That share is projected to keep rising through 2028. The remaining 37.7% is what all other platforms and every general web media property in the world have to split among themselves.

Concentration of budgets on the platforms means that the budgets reaching general web media are shrinking in relative terms. Media owners are left to survive on whatever spills over from the platforms.

Around 2020 there was a surge in online ad spending. The pandemic accelerated the shift to digital and temporarily pushed CPMs up. But in the backlash that followed, ad spend growth slowed from 2022 onward, and at many media properties CPMs returned to—or fell below—previous levels. The media that had built their revenue plans around that temporary boom are now in an uncomfortable position.

Furthermore, as of 2025, the IAB revised its US ad spend growth forecast down from 7.3% to 5.7%, citing macroeconomic uncertainty and the impact of tariffs. The industry as a whole is being forced to acknowledge the wave of slowing growth.

To be candid, I see no factor that would dramatically recover CPM from here. If anything, I think a gradual decline is more likely.

Observation: The top of the advertising pyramid is taking a larger cut for itself

What follows is not a fact but an observation from someone who has spent a long time in the ad industry watching the field. Please read it as such.

Google, which sits at the top of the advertising pyramid, has visibly strengthened the exposure of its ad formats over the past few years. Two concrete examples.

First, ads within AI search results. In October 2025, Google announced it would expand ads inside AI Overviews beyond the US to English-speaking markets such as the UK. As of March 2026, AI Overviews appear on roughly 48% of all searches, and ads show up in about 25% of those. A structure in which AI summarizes the answer at the very top of the search results, with ads placed above, below, and inside it, is becoming the norm.

Second, the expansion of ad formats on media pages. In February 2026, Google added new triggers for full-screen vignette ads in AdSense. Scrolling back after reaching the end of an article, performing an action after 30 seconds or more of inactivity, and using the browser's back button were all newly added as triggers for full-screen ads. Opportunities for full-screen ads to appear on media pages, triggered by reader behavior, were intentionally increased.

The player at the top taking a larger cut looks to me like a sign that the industry as a whole has reached a plateau in its growth.

That said, other readings are possible. It may simply be Google redesigning its revenue sources during the transition to AI search, or it may be pure revenue maximization. I do not have hard evidence. But as someone whose livelihood has depended on ad revenue, the anxiety that the platform's format strengthening "can only work in the direction of pushing down the media's share" does not go away, as a matter of field intuition.

When do we start building revenue pillars beyond advertising?

To summarize the three structures and one observation laid out so far:

• Adding slots lowers CPM, and the same loop keeps turning
• Ad blocking (29.5% globally) and privacy regulation erode impressions and targeting accuracy
• Meta, Google, and Amazon absorb 62.3% of global digital ad spend, and the amount reaching general media is shrinking in relative terms
• The top player is strengthening its own ad-format exposure, which is likely to compress the media's share further (observation)

None of this is because someone is doing something wrong. Advertisers are paying fair value for performance. Platforms are providing efficient ad delivery infrastructure. Users who adopt ad blockers are making a rational choice. And regulation is a social consensus to protect consumer privacy.

But as a result, web media ad revenue has plateaued structurally.

If that is the case, then what a media operator should be thinking about is not how to add more ad slots, but how to cultivate revenue pillars beyond advertising.

Subscriptions, membership communities, events, merchandise, content licensing. There are several options. None of them grows overnight, and none is as simple as "just add a slot." They require the work of rebuilding the relationship with readers—moving it away from the indirect metric of page views toward a more direct connection.

There is one more approach that tends to be overlooked. Delivering content in forms other than "reading" can move dwell time and repeat-visit rates in certain cases. Turning articles into audio is one example, but it is not a solution to the structural problem of display CPM itself. It is one option for changing the quality of the touchpoint with readers.

In closing

I have no intention of writing a prescriptive conclusion about what the industry should do. I am myself an industry insider who has been sustained by ad revenue. Advertising remains, even now, an indispensable source of income for many media.

What I felt, though, as someone on the ground, was a quiet anxiety: "Is it really okay to keep going like this?" Add slots, prices fall, add more slots. Inside that loop, the scenery readers see gradually becomes saturated with ads. Even knowing that this is the result of accumulated rational decisions, something about it always felt off.

Perhaps I just wanted to put that sense of dissonance into my own words.

I do not have a firm answer as to what the next revenue model for web media will be. But there is no doubt that the time has come to go looking for one.

Related articles

• What changes when web media articles are turned into audio — How dwell time and repeat rate moved with audio
• A field comparison of web media article audio services, by people on the ground — A comparison of BeyondWords, PUBVOICE, ReadSpeaker, and others
• Why we built an AI audio SaaS for web media — Why I built an audio product coming from the ad side
• Imitation in the AI era and delivery beyond text — What it means to hold a delivery route other than "reading" in the AI search era