B2B copy · CTAs · longevity · lead quality
Quick Answer
- Product-led CTA language appeared 2.74 times as often as demo-related language in the analysed SaaS Google Search ad clusters.
- “Free trial” was the largest individual SaaS-native phrase, appearing in 1.33 million archived ads.
- In a separate 2,000-ad sample, 50–125-character stored description-copy strings had the highest 30-day observed survival rate at 8.4%.
- The most common copy length was not the most persistent, showing why frequency alone should not decide your ad strategy.
- Use AdSpyder’s lead generation CTA benchmark study to compare product-led and sales-led copy patterns.
- These are ad-copy prevalence and longevity benchmarks—not competitor CTR, CPC, conversion-rate, CPL, or ROAS benchmarks.
B2B Google Ads benchmarks are often reduced to average CPC and conversion-rate tables. Those numbers can help with forecasting, but they do not answer a more practical creative question: what language are B2B advertisers actually using in Google Search ads?
- This AdSpyder research focuses on observable ad copy rather than private account performance. It compares product-led and demo-led CTA language, individual SaaS keyword themes, description-copy length, and observed ad longevity.
- The findings provide a market baseline for writing new ads. They do not tell you which copy will win in your account. Final decisions must still be based on qualified leads, opportunities, customers, and revenue.
What This Report Benchmarks
Ad-copy behaviour in the market—not private advertising performance.
Use the data to choose stronger hypotheses. Use your Google Ads account, analytics, CRM, and sales data to select the actual winner.
B2B Search Ad Benchmark Report
Methodology and Dataset
This report combines two AdSpyder Google Search archive analyses. The datasets answer different questions and were not merged into a single performance score.
Dataset A · CTA Language
167,750,873 ads
Google Search archive snapshot analysed in July 2026.
Measured exact and related CTA phrases appearing in SaaS-relevant archived ad copy.
Dataset B · Copy Longevity
2,000 random ads
Sample drawn from a 167,505,019-ad Google Search archive snapshot in June 2026.
Compared stored copy length with the gap between first-seen and last-seen dates.
Supporting category context: AdSpyder also reviewed a manually curated sample of 30 SaaS brands across enterprise software, collaboration, ecommerce SaaS, automation, AI, support software, and other B2B-oriented categories.
The 30-brand sample supports the B2B interpretation of the CTA findings. It is not a complete classifier for every B2B advertiser or industry.
Executive Findings
AdSpyder Google Search archive snapshots, June–July 2026.
2.74×
PLG-to-demo ratio
Product-led CTA cluster versus demo cluster
1.33M
“Free trial” ads
Largest individual SaaS-native phrase
8.4%
30-day survival
Leading stored description-copy bucket
4.4×
Longevity difference
8.4% versus 1.9% at 30 days
B2B Google Search CTA Benchmarks
The clearest split in the SaaS-relevant data is between product-led acquisition and sales-assisted acquisition.
Product-led CTA cluster
2,078,027 ads
Free Trial · No Credit Card · Try Free · Start Free · Sign Up Free
Demo-related CTA cluster
758,881 ads
Demo · Book a Demo · Book Demo · Watch Demo and related phrases
What this means: Product-led copy is more prevalent in the archive, but that does not make it the correct model for every B2B campaign. Enterprise, regulated, customized, or high-contract-value products may still require qualification and a sales conversation.
Most Visible SaaS and B2B Copy Themes
Phrase groups can overlap. One ad may contain “pricing,” “free trial,” and “no credit card,” so the counts should not be added together to calculate market share.
Google Search Copy-Length Benchmarks
The separate 2,000-ad longevity sample found that the most common stored description-copy bucket was not the most persistent.
30-Day Observed Survival
Under 50
7.9%
50–125
8.4%
125–250
5.3%
250–500
1.9%
Current RSA note: Google currently allows up to 15 headlines and four descriptions in one responsive search ad. Each headline supports up to 30 characters and each description up to 90 characters. The archive study measured stored historical description-copy strings, so its larger buckets should not be treated as current single-asset limits.
See the full methodology and cross-platform comparison in AdSpyder’s 211M+ ad copy-length study.
What the Benchmarks Mean by B2B Buying Motion
Self-Serve or PLG SaaS
Reduce the risk of trying the product
Useful copy: Free Trial · Try Free · No Credit Card · Get Started
Best page: Fast signup, clear activation steps, and transparent limitations.
Enterprise or Sales-Led SaaS
Earn the sales conversation
Useful copy: Book a Demo · Talk to Sales · Request Assessment
Best page: Industry proof, implementation details, security, integrations, and qualification.
Hybrid SaaS
Separate small and enterprise buyers
Useful copy: Start Free for teams · Book a Demo for enterprise
Best page: Two clear conversion paths based on complexity, company size, or buying need.
B2B Agency or Service
Make the deliverable specific
Useful copy: Request Audit · Get Proposal · Book Assessment
Best page: Scope, process, qualification, proof, timeline, and what happens after submission.
Patterns and Anomalies in the Data
What B2B Marketers Should Do Next
Define the buying motion first
Decide whether the user can start independently, needs sales qualification, or should choose between both paths. Do not select “Free Trial” or “Book a Demo” because it is common.
Build distinct RSA asset families
Create unique assets around the problem, outcome, differentiator, proof, risk reducer, category, and CTA. Avoid filling 15 headline slots with minor rewrites of the same claim.
Keep each description focused
Use one main benefit, one proof or risk reducer, and one appropriate next step. Current RSA descriptions allow up to 90 characters each.
Match the CTA to the destination
A “free trial” ad should lead directly into a credible signup path. A “book a demo” ad should explain who the demo is for, what it covers, and what happens after submission.
Test a commitment ladder
Compare suitable actions such as View Pricing, Start Free, Get Assessment, and Book a Demo. Keep the audience and offer stable so the CTA comparison remains interpretable.
Return qualified outcomes to Google
Import qualified-lead and converted-lead outcomes so bidding is guided by commercial progress rather than every form submission receiving equal value.
Use AdSpyder’s impression-to-revenue KPI framework to separate diagnostic metrics such as CTR and CPL from CPQL, opportunities, CAC, and revenue.
How to Benchmark Competitor Copy With AdSpyder
The global data provides a starting point. The more useful benchmark is the one built from advertisers selling to the same audience, in the same country, with a similar buying motion.
STEP 1
Select comparable B2B domains
STEP 2
Filter Google Search ads
STEP 3
Tag the CTA and offer
STEP 4
Inspect landing pages
STEP 5
Create an original test
Recommended Research Fields
Competitor · country · keyword · audience signal · headline angle · proof · CTA · risk reducer · landing-page type · observed dates · original test idea
- Start with AdSpyder Ad Analytics to compare visible domain activity, recurring keywords, active campaigns, country distribution, and advertising patterns.
- Then use the Google Ads Spy to examine headlines, descriptions, CTA language, destination pages, dates, and competing advertisers around each priority keyword.
Build a B2B ad-copy benchmark for your own market
Compare competitor headlines, CTA language, keyword themes, observed duration, and landing pages before writing the next RSA.
Limitations and Methodology Notes
Use the benchmark to build a test—not to declare a winner
AdSpyder shows what language is visible across the market. Your Google Ads, CRM, and revenue data should decide which message earns more investment.
Frequently Asked Questions
What are B2B Google Ads benchmarks?
They are reference points for assessing B2B search campaigns. They may cover ad-copy patterns, CTR, CPC, conversion rate, CPL, lead quality, pipeline, or revenue, depending on the dataset.
What CTA appears most often in SaaS Google Search ads?
“Free trial” was the largest individual SaaS-native phrase in this AdSpyder archive analysis, appearing in approximately 1.33 million ads.
Is Start Free better than Book a Demo?
Not universally. Start Free fits a self-serve product, while Book a Demo fits products requiring qualification, customization, security review, or sales assistance.
What is the ideal Google Search ad-copy length?
There is no guaranteed ideal. In AdSpyder’s 2,000-ad sample, the 50–125-character stored description-copy bucket had the strongest observed longevity. Current RSA descriptions remain limited to 90 characters per asset.
Does an ad running for 30 days prove it converts?
No. Observed longevity is a research signal only. The advertiser’s conversion rate, CPL, lead quality, and profitability remain private.
How many headlines should a responsive search ad include?
Google allows up to 15 headlines and four descriptions. Add distinct, useful assets rather than filling every slot with repetitive variations.
Which metrics should B2B advertisers optimize for?
Track CTR and CPL for diagnosis, but prioritize qualified-lead rate, CPQL, opportunity rate, cost per opportunity, CAC, and revenue.
How does AdSpyder help benchmark Google Search copy?
AdSpyder lets marketers research competitor ads, keyword themes, CTA language, observed dates, domains, countries, and landing pages before creating original tests.
