In early 2025, The Wall Street Journal published an investigation into Google’s advertising division that revealed an uncomfortable truth for business owners and marketing managers: Google Ads “strategists” are not trained optimisation experts, they are sales representatives measured on how effectively they grow Google’s revenue.

The Journal found that Google recruits many of its Ads staff from outsourced sales organisations including Teleperformance, TTEC, Accenture’s offshore teams and Cielo.

Their onboarding and KPIs emphasise:

• product-adoption quotas,
• budget-expansion targets, and
• quarterly revenue growth metrics.

Almost none of the documented training covered search intent, segmentation, negative keyword strategy or diagnosing wasted spend, the fundamentals of effective search advertising.

These representatives are not incentivised to improve ROI. They are incentivised to increase budgets.

Across hundreds of audits, JMarketing has found the same pattern: whenever a Google rep has shaped an account, automation is overused, visibility is reduced, and a meaningful portion of the budget, often 30% to 80%, is misallocated to irrelevant or low-intent traffic.

This is not an isolated issue.

It is structural.

And it stems from deeper pressures inside Google’s business model.

Google’s Business Is Facing
Structural Decline

For two decades, Google’s dominance rested on one engine: search advertising. According to Alphabet’s 2024 annual filing, **over 80% of Alphabet’s global revenue still comes directly from ads served through Google Search.**¹

But that engine is now under sustained pressure.

1. AI is replacing search queries

Traffic analytics from Similarweb, SparkToro and Gartner-tracked sources all reported **consistent month-on-month declines in total Google search volume throughout 2024.**²

The reason: AI systems such as ChatGPT, Perplexity, Claude and even Google’s own Gemini now answer questions directly, summarising, interpreting and recommending without requiring a search query at all.

This is not competition.

It is displacement.

The search query itself is being bypassed.

2. Younger audiences are no longer
starting on Google

A 2024 Axios survey found that only 46% of U.S. users aged 18–24 begin information searches on Google.³ The rest start on TikTok, YouTube, Reddit or AI tools.

This is the first generation in modern history for whom “Googling” is not the default.

3. Amazon has captured commercial-intent searches

Jungle Scout’s 2024 Retail Index found that over 60% of U.S. product searches now begin on Amazon⁴. These were historically Google’s most profitable queries.

Under this pressure, Google has pursued a clear strategy:
extract more revenue from the advertisers who remain.

DoJ Testimony Confirmed Google Quietly Raised Ad Prices

During the U.S. Department of Justice antitrust case, internal Google documents showed that the company used algorithmic controls internally referred to as “pricing knobs” to raise ad prices by 5–15%, in some categories more, without informing advertisers.⁵

Executives admitted under oath that these increases were designed to meet quarterly revenue targets.

The Wall Street Journal reported on an internal email in which a senior Google executive discussed the need to “shake the cushions” a reference to extracting additional revenue from the ad auction system.⁶

In September 2024, a federal court ruled that Google had illegally maintained monopolies across crucial layers of the digital advertising stack.⁷

This is the ecosystem in which advertisers operate today:

• Google controls the auction
• Google controls the pricing
• Google controls the visibility
• Google controls the reporting
• Google controls the incentives of the people advising you

Google Has Systematically Reduced Advertiser Visibility

A 2024 analysis by Search Engine Journal found that Google now hides a “significant percentage” of search terms from advertisers, often the very queries where irrelevant spend accumulates.⁸

Performance Max hides even more.

Responsive Search Ads generate thousands of combinations with no transparency.
Smart Bidding uses signals advertisers cannot inspect.
Broad match routinely expands into loosely related terms.
Ad Strength scores compliance, not quality.

This fragmentation of visibility benefits only one party.
But the complexity runs deeper.

The Hidden Complexity in Google’s “Smart” Ad Formats

Responsive Search Ads (RSAs) allow advertisers to provide up to 15 headlines and 4 descriptions, which Google claims will be “automatically optimised.”

A standard RSA can generate over 2,400 potential headline–description combinations.

Testing these combinations properly would require an advertiser to generate statistically meaningful data for each variant, an impossible task for all but the largest enterprise budgets.
So while Google’s optimisation promise is technically true, it is practically impossible.

Most combinations are never really tested. Advertisers cannot see which combinations are shown. Google’s algorithm relies on broad signals that often prioritise impressions over quality.
And advertisers are encouraged to “trust the system” rather than understand it.

This illusion of sophistication encourages inexperienced advertisers, often the most profitable segment for Google, to surrender control, accept opacity, and increase spend to make up for inefficiencies they cannot see.

This is not framed as a risk in Google’s messaging.
But it is one of the largest contributors to hidden waste.

Why Following Google’s Recommendations Consistently Underperforms

Most advertisers assume that following Google’s recommendations is safe. Google built the platform; surely they know best.

But the evidence suggests otherwise.

A clear example comes from Maple Bear UAE, a premium early-education provider operating across several highly competitive locations. Prior to a structural audit, Maple Bear followed all Google-recommended best practices: broad match, Smart Bidding, Performance Max and increased budgets.

Yet performance plateaued, and more than 80% of enquiries were classified as low-quality, despite significant investment. Google repeatedly described the account as “performing well.”

It wasn’t.

When JMarketing rebuilt the account manually, and without implementing any of Google’s recommendations, performance shifted immediately:

• Total qualified enquiries increased by several hundred percent
• Unqualified enquiries fell from above 80% to below 20%
• Cost per lead more than halved
• Overall spend decreased significantly, while results improved
• Historically difficult, highly competitive locations saw strong YoY growth in all commercial metrics
• Tour bookings (the true commercial indicator) increased by triple-digit percentages year-on-year

The same pattern appeared in another UAE education provider immediately before Maple Bear.

When the same structural corrections produce the same dramatic improvements in different organisations, the conclusion is clear:

Google’s recommendations increased spend.
Correct structure eliminated waste.

What Waste Actually Looks Like Inside Real Accounts

Waste is not a high CPA. It is budget quietly funneled into irrelevant queries that never convert.

Case: New York Tennis Club – $250,000 Misallocated

This tennis club had been assured repeatedly by Google that its account was “healthy.” A manual audit revealed thousands spent on searches such as:

• “law courts near me”
• “boat party venue NYC”
• “birthday party on a jet”

Google’s broad match had equated “tennis court” with “law court.” No Google representative identified the issue over nearly a decade. A five-minute review revealed approximately $250,000 in wasted spend.

Case: High-End Dating Platform – The Illusion of Performance

This client targets high-achieving singles. One subgroup was Jewish singles. In Google’s interface, the segment looked successful:

• high lead volume
• low CPL
• excellent Ad Strength

The underlying search terms told a different story:

• Jewish-specific keywords were matched to generic dating searches
• Ads written for Jewish audiences were shown to non-Jewish users
• High-intent audiences were not being reached at all
• The appearance of performance was caused by automation drift, not genuine success

The owner believed the account was performing because Google had repeatedly told him it was. But the underlying data told another story.

The Silent Cost: The Revenue You Never Realise You Lost

The most damaging inefficiency in Google Ads is not the irrelevant clicks you pay for, it is the high-value customers who never see your ads, and the profitable campaigns you never know existed.

From the outside, Google Ads appears to be working. Traffic arrives. A percentage converts. Some enquiries look healthy.

But what advertisers cannot see, and what Google’s interface is designed not to reveal, is which audiences are being served, who is being missed, and where budget is actually flowing.

Performance Max is the clearest example. Google chooses the networks, the placements, the inventory and the audience allocation. Advertisers get outcomes, not insight. You cannot tell:

• whether your highest-value, bottom-of-funnel users are seeing your ads,
• whether budget is being diverted to people who were never a match,
• or whether a previously high-performing segment has stopped firing entirely.

Every year, Google removes more control “for privacy.” Each reduction makes it harder for advertisers to see what is happening underneath, and easier for Google to justify automation that nudges budgets upward.

With AI eroding search volume and revenue under pressure, this opacity is commercially convenient.

For Google, it keeps the machine running. For advertisers, it hides the truth.

Experts who understand the remaining levers of control, and who know how to interpret what little signal remains, can restore clarity.

Without that expertise, advertisers are flying blind.

A Rational Next Step for Any Advertiser

At this point, the situation is clear.

Google’s advertising system has become too complex, too opaque and too heavily automated for most advertisers to determine where their budget is going or why results behave as they do. Google’s own representatives cannot diagnose these issues, they are not trained to, they are not incentivised to, and the platform does not expose the underlying data.

For any business where Google Ads represents a meaningful line item, the only responsible way to understand the true state of the account is through a manual, independent review, one that examines the search terms Google does not surface, the intent pathways automation has distorted and the structural decisions that quietly redirect spend.

Some advertisers will discover their campaigns can be rebuilt profitably. Others may find that Google Ads does not suit their market or buying cycle.

Both outcomes are commercially valuable. What matters is clarity.

Sources Mentioned (Newspaper Style)

¹ Alphabet Annual Report, 2024 – Advertising revenue share
² Similarweb, SparkToro, Gartner (2024) – Search volume decline
³ Axios (2024) – Gen Z search behaviour
⁴ Jungle Scout Retail Index (2024) – Amazon search dominance
⁵ DOJ vs Google (2023–2024) – “Pricing knobs” testimony
⁶ The Wall Street Journal – Internal “shake the cushions” revenue email
⁷ U.S. District Court Ruling – Google monopoly findings
⁸ Search Engine Journal (2024) – Hidden search term analysis