If paid search is “working” but CAC still feels like a dare, the constraint usually isn’t bidding—it’s what you told the algorithm to value.

If paid search is “working” but CAC still feels like a dare, the constraint usually isn’t bidding—it’s what the account is being trained to value. In B2B SaaS benchmarks, paid search CAC is cited around $802 per acquisition (Oliver Munro). That’s not cheap. So when a program is optimized to “conversions” that never become pipeline, the math gets ugly fast.

Here’s the uncomfortable baseline: Position Digital cites average B2B SaaS PPC conversion rate at ~1% (vs. ~2.1% for SEO). Low conversion rates aren’t a moral failure. They’re a signal that measurement has to be tighter than your instincts want it to be.

So the one move in this playbook is simple: stop optimizing paid search to leads, and start optimizing it to a CRM-defined, stage-based conversion that actually correlates with revenue.

Nut graf: This matters in 2026 because automation is doing more of the steering in Google Ads and Microsoft Advertising, while privacy changes and GA4 migrations make it easier to lose the thread between click → lead → opportunity. Holini’s guidance is blunt: measure paid search against qualified leads, opportunities, and revenue by connecting ad platforms to CRM/analytics—don’t stop at CTR, CPC, or CPL. If the conversion signal is wrong, smart bidding will get “smart” about the wrong outcome.

The core problem: the platform can’t optimize what you don’t define

Most B2B teams don’t have a paid search problem. They have a conversion definition problem. The ad account is counting things that are easy to generate and hard to monetize: content downloads, low-intent form fills, “contact us” spam, students doing research. Busywork. Expensive busywork.

The source material behind this prompt describes audits where a meaningful share of “conversions” were ebooks/whitepapers and where large chunks of spend produced zero high-intent actions. Those specifics can’t be independently verified from the research brief, so they shouldn’t be treated as universal facts. But the pattern is common enough that it’s worth treating as a diagnostic: when volume looks fine and pipeline doesn’t, the conversion mix is usually the culprit.

But the context is more complex. Teams often keep those low-intent conversions because they need something to feed automation. And with a ~1% PPC conversion rate benchmark (Position Digital), there’s real fear that if the “easy” conversions are removed, the account will starve.

That fear is rational. It’s also fixable—if the conversion reset is staged and tied to CRM reality.

The tactic: import a revenue-proxy conversion and make it primary

The better approach is to treat paid search as a full-funnel channel operationally, even if it’s still mostly a capture channel strategically. Holini recommends connecting Google Ads/Microsoft Ads to CRM and analytics so reporting and optimization can use qualified leads, opportunities, and revenue outcomes—not just platform metrics.

In practice, that means picking one “revenue-proxy” event that (a) happens early enough to have volume and (b) is strict enough to correlate with pipeline. Call it what you want—SQL, SAL, qualified meeting held—but make it a CRM stage with rules, not a vibe.

Then: import it back into the ad platform as an offline conversion and set it as a primary conversion. Everything else becomes secondary. The ad system still sees the other signals, but it stops treating them as the finish line.

This is also where the glossy ROI claims get filtered through reality. Averi.ai cites B2B SaaS paid ads ROI around ~$1.80 returned per $1 spent (benchmark summary), while Powered by Search cites a Google Ads example with 423.31% ROI on $215,500 spend, 91 customers, and LTV:CAC of 4.23. Both can be “true” in their own context. The difference is usually measurement quality, offer clarity, and whether the conversion signal matches revenue.

Run it this week: setup, launch, readout, next test

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

Setup (Day 1–2): Pick the stage-based event.

Launch (Day 3): Import and reclassify conversions.

The hypothesis (make it falsifiable): If we switch the primary conversion from lead-submit to a CRM-qualified stage (SQL/opportunity), then cost per qualified pipeline event will decrease and qualified pipeline per $ will increase because smart bidding will stop optimizing toward low-intent form fills.

Readout (Week 2–4): What to measure (and what not to over-interpret):

Next test (Week 4+): Once the conversion signal is stable, then test automation changes (bidding strategy, keyword expansion, match types). Marcel Digital’s position is the right posture here: use automation to enhance management, but keep humans responsible for keyword quality and budget allocation. Translation: don’t hand the keys to the algorithm until the dashboard is telling the truth.

The trade-off (and when this is wrong)

The trade-off is immediate and predictable: reported “conversion volume” will drop. Some stakeholders will interpret that as failure. It’s not. It’s the measurement getting more honest.

When this is wrong: if the business has extremely low stage volume (for example, only a handful of SQLs a month from search), the platform may not have enough signal to optimize against the stricter event. In that case, the practical workaround is to choose an earlier proxy that still has intent (for example, “demo request with required fields” that Sales agrees is real), then tighten it over time as volume grows.

Paid search doesn’t need more dashboards. It needs one clean definition of success, wired end-to-end. Because with CAC benchmarks like ~$802 (Oliver Munro) and PPC conversion rates around ~1% (Position Digital), there isn’t much room for pretend wins—only the kind that show up in pipeline.