Most B2B accounts run responsive search ads the same way they ran expanded text ads in 2019: write a few headlines, let Google figure it out, check the Ad Strength score, move on. The result is predictable. Clicks go up, conversion rates drift sideways, and nobody can explain which ad copy actually influenced pipeline.
The problem isn't the format. RSAs work. The problem is that B2B marketers optimize for the wrong signals, feed the algorithm incomplete data, and then blame automation when CAC payback stretches past 18 months.
Ad Strength Is Not a Performance Metric
Let's start with the metric that wastes the most time. Google's own documentation states that Ad Strength measures the relevance, quantity, and diversity of your ad assets. It does not factor into Ad Rank. It does not determine whether your ad serves. It is a setup checklist, not a performance indicator.
Amy Hebdon's analysis at SMX Advanced put it plainly: Ad Strength reflects what has worked for the masses in the past, but it doesn't care about what works for you. An ad with "Poor" Ad Strength that converts at 4% is more valuable than an "Excellent" ad that converts at 1.2%. Yet teams spend hours chasing the green checkmark instead of running holdout tests on actual revenue outcomes.
Google does report that advertisers who improve Ad Strength from "Poor" to "Excellent" see 15% more clicks and conversions on average. That's a population average across all verticals, all deal sizes, all sales cycles. For a B2B account with a 90-day sales cycle and a five-person buying committee, the correlation between Ad Strength and closed-won revenue is far weaker than that headline suggests.
The Real Optimization Lever: Conversion Signal Quality
Here's where the math changes. SaaS Hero's 2026 analysis found that 67% of B2B Google Ads campaigns still optimize for lead volume metrics rather than downstream conversion to SQL or closed-won deals. The algorithm gets better at finding form-fillers every month you feed it the wrong signal.
The fix is offline conversion tracking. You capture the GCLID on every click, store it with the lead in your CRM, and send stage-change events (meeting booked, opportunity created, deal won) back to Google Ads. Customer Impact's implementation guide explains the mechanics: the click identifier travels from ad to deal and back, so Smart Bidding learns which keywords and audiences produce pipeline, not just which ones produce forms.
DataCops' September 2026 breakdown clarifies the two paths. Offline conversion import matches on the GCLID you stored, with a 90-day attribution window. Enhanced conversions for leads matches on hashed email or phone, with a 63-day window. They fail in different situations, so running both covers each other's gaps. The GCLID handles the heavy lifting when it's present; the email match catches sales when the click ID leaked somewhere in your form stack.
For B2B accounts with sales cycles longer than 63 days, the practical advice is to send an earlier stage (qualified lead, meeting booked) while the click is still in range. A signal that arrives inside the attribution window beats a perfect signal that arrives too late.
Asset Strategy: Variation, Not Volume
Pace Ads' 2026 guide identifies the most common RSA mistake: entering 15 versions of the same headline with slightly different phrasing. Google then has no real variability to test. The algorithm needs range across tone, intent, message, and funnel position.
A functional asset set includes:
- Value propositions ("Same-Day Results. No Delay.")
- Trust signals ("Used by 10,000+ Businesses")
- Calls to action ("Book Your Free Audit Today")
- Objection handlers ("No Hidden Fees or Contracts")
Each headline should serve a distinct function. Near-duplicates like "PPC management," "expert PPC management," and "professional PPC management" add nominal quantity without giving the system a meaningfully different message.
Spires Digital's July 2026 guidance adds a critical constraint: read each headline beside every other headline. Remove combinations that repeat, contradict one another, or create an unsupported claim. Since Google's February 2024 update allows RSAs to display a single headline when predicted to boost performance, every asset must be comprehensible without relying on the one before it.
Pinning: When Control Beats Automation
Pinning gets a bad reputation because it lowers Ad Strength. But Ad Strength, as established, is not a performance metric. Lunio's expert interview with Amy Hebdon explains when pinning makes sense: when you need brand name in position one, when compliance requires specific disclosures, or when testing has shown that a particular CTA in headline two consistently outperforms alternatives.
The tradeoff is real. Pinning reduces the combinations Google can test. But for B2B accounts where brand recognition matters in a crowded SERP, or where legal requires certain language, the control is worth the score penalty. The question is whether the constraint improves conversion rate enough to offset the reduced impression share. That's a test, not a rule.
The Reporting Gap That Matters
Jordan Fry's July 2025 LinkedIn post flagged a significant update: Google Ads now surfaces real, actionable data for RSA assets, including cost, conversion, conversion rate, and CPA at the headline level. Check your accounts under Assets → Performance → Headline → Load new columns. This changes how you evaluate which headlines actually drive outcomes, not just which ones Google shows most often.
Before this update, the combinations report showed impressions but not conversions. You could see what Google preferred to serve; you couldn't see what actually closed. Now you can. The implication for B2B accounts is that you can finally tie headline performance to downstream metrics, assuming you've connected your CRM and are importing offline conversions.
A Two-Week Pilot Plan
Week One
Audit your current RSA setup. Count distinct headlines (not variations of the same message). Check whether you're importing CRM conversions or still optimizing for form fills. Verify GCLID capture across all form paths. Review the combinations report to see which assets Google is actually serving.
Week Two
Rebuild one ad group as a test. Write 10 to 12 headlines across four categories: value prop, trust signal, CTA, objection handler. Set up offline conversion import if you haven't. Add the new conversion action as secondary first, so it reports without steering Smart Bidding until you've validated the data. Pin brand name to position one if brand recognition matters in your market.
Run for 30 days. Compare cost per SQL (not cost per lead) against your control. If the test wins, roll out the structure. If it doesn't, check whether the conversion signal is actually reaching Google inside the attribution window.
The goal isn't to chase Ad Strength. The goal is to make the algorithm smarter about what actually produces revenue. That requires better signals, not more headlines.
