If Google Ads CAC is climbing while you “improve” structure, the constraint is usually self-inflicted: too many campaigns and ad groups starving Smart Bidding of clean conversion data.

If Google Ads CAC is climbing while you “improve” structure, the constraint is usually self-inflicted: too many campaigns and ad groups starving Smart Bidding of clean conversion data.

That sounds backwards. More segmentation feels like more control. But in 2026, control often means fragmentation—of budget, of learning, of signal.

And the macro pressure is real. WordStream’s benchmarks show the average Google Ads CPC was $4.22 in 2023, with 61% of industries seeing CPC increases year over year and 91% seeing CPL increases (WordStream, Google Ads Benchmarks 2023). When auctions get pricier, “pretty” accounts get expensive fast.

Here’s the nut graf: paid search is mostly demand capture. When teams try to turn it into demand gen—by sprawling into low-intent keywords, overlapping structures, and endless micro-campaigns—CAC creeps up, attribution gets noisier, and everyone starts blaming the channel. KlientBoost’s argument is blunt: overcomplicated Google Ads account structures can increase CAC, and simplification plus high-intent focus is the fastest path back to efficient unit economics (KlientBoost, Google Ads Account Structure To Reduce CAC).

The one move: consolidate into four campaigns to rebuild signal

If you only change one thing, change this: collapse your search program into a small number of intent-based campaigns, then force everything else (keywords, ads, landing pages, bidding) to serve data density.

KlientBoost recommends a simplified structure with four campaigns: Brand, In-Market (high-intent non-brand), Competitor, and Retargeting (KlientBoost, source). The point isn’t the number “four” as dogma. It’s the principle: fewer buckets, clearer intent, less internal competition.

There’s a second, very 2026-specific reason this works. AI Advantage Agency calls out that over-segmented campaigns, excessive ad groups, rigid exact match lists, and manual bidding can fragment conversion data and worsen CPL/CAC under modern Smart Bidding (AI Advantage Agency, Why Is My Google Ads Cost Per Lead Increasing in 2026). In other words: the machine can’t learn from conversions it never sees.

But the data question hangs there: does simplification actually move downstream metrics, not just dashboard CPAs?

KlientBoost says yes—at least in one documented B2B use case, a simplified structure reduced cost per opportunity by 61% (KlientBoost, source). That’s not a promise. It’s a directional proof that structure can hit pipeline economics when it stops fighting the algorithm.

What it looks like in practice (and where teams mess it up)

The four-campaign model is simple. The execution rarely is—because the failure mode is letting “simple” turn into “sloppy.”

Campaign 1: Brand. Protect it, but don’t worship it. KlientBoost suggests allocating 0–40% of budget to brand, depending on competitive pressure (KlientBoost, source). Brand can be efficient, but it can also be a last-click vanity mirror. Run incrementality checks where possible; don’t let it silently tax your CAC.

Campaign 2: In-Market (high-intent non-brand). This is the workhorse. KlientBoost suggests roughly 40–60% of budget here (and more if competition is low) (KlientBoost, source). The constraint: only buying-intent keywords. Category and capability terms that signal someone is shopping, not studying.

Campaign 3: Competitor. Useful, but volatile. KlientBoost suggests 10–20% budget allocation, adjusted based on conversion performance (KlientBoost, source). Expect lower CTR, different messaging requirements, and higher scrutiny from legal/brand teams.

Campaign 4: Retargeting (search + audience layer). KlientBoost frames retargeting as lower volume and roughly ~10% of spend (KlientBoost, source). The nuance worth copying: using a high-intent audience layer (like pricing-page or demo-page visitors) as a “protective barrier” when opening up match types in that retargeting context.

Now the part most accounts skip: ad group design. KlientBoost says they often use STAGs—Single Theme Ad Groups—built around tightly related keyword themes, then pair them with custom copy and landing pages where possible (KlientBoost, source). That’s the practical middle ground between SKAG-era micromanagement and today’s “one ad group to rule them all.”

Run it this week: setup, hypothesis, metrics, guardrails

Here’s the 5-minute version you can run this week:

Setup. Build the four campaigns. Move existing keywords into the right bucket by intent (brand, in-market, competitor). Remove obvious low-intent modifiers (jobs, careers, login). Keep the structure boring on purpose.

Launch. Start bidding with Max Conversions, per KlientBoost’s recommendation, and only consider moving to tCPA/tROAS when impression share is nearing saturation (KlientBoost, source). That sequencing matters because premature constraint can choke learning.

The hypothesis (make it falsifiable): If we consolidate into four intent-based campaigns and align bidding to Max Conversions, then cost per opportunity will decrease within one sales cycle because Smart Bidding will have denser, less-conflicted conversion data to optimize against.

Readout. Don’t grade this on week-one CPL. Measure down-funnel.

Next test. Once impression share is tight, test tCPA/tROAS using Google Ads Experiments so the comparison runs under similar market conditions (KlientBoost recommends using experiments to fairly test bid strategies) (KlientBoost, source).

The trade-off should be named: this can reduce volume before it improves quality. High-intent capture is finite, and simplification won’t fix a weak offer or a broken handoff. It just stops the account structure from making those problems worse.

HockeyStack Labs adds another piece of context: across 50+ B2B SaaS companies and $100M+ in Google Ads spend, average budget allocation to Google Ads declined from 55% (2022) to 47.72% (2023) to 39.85% (Q1 2024) (HockeyStack Labs, Google Ads Benchmarks from 2022 to 2024). The channel is still important, but it’s getting held to a higher standard. Structure that protects CAC is part of that defense.

And yes, automation is accelerating. Google launched AI Max for Search in May 2025, and Google reports 14% more conversions at similar CPA on average (as summarized by Involve Digital) (Involve Digital, Google Ads for B2B SaaS: Strategy Guide 2026). But automation doesn’t rescue messy architecture. It amplifies it.

That’s the circle back to the opening: when CAC climbs, the instinct is to add more knobs. The better move is usually subtraction—fewer campaigns, clearer intent, and a measurement chain that ends in pipeline, not leads.