SEO A/B Testing That Proves
Which Changes Actually Worked
Run controlled SEO experiments on your real Google Search Console data. Compare the pages you changed against a control group, and get a statistical verdict with lift, confidence, and caveats instead of a before-and-after guess.
Title rewrite: price range in category titles
Test: /category/* (40 pages) vs suggested control
Likely winner
Test pages gained clicks relative to the control after the change. No Google core update overlapped the test window.
Estimated lift
+18.4%
95% range
+9.1% to +27.9%
Confidence
97%
Control fit
r = 0.86
Daily clicks: test vs control
Why Before-and-After
SEO Reporting Lies to You
Traffic moves for dozens of reasons that have nothing to do with your change. Without a control group, you can't tell a win from a coincidence.
Seasonality Hides the Truth
A title rewrite in November looks like a win when holiday demand is the real reason traffic went up. The same change in January looks like a loss.
Google Updates Skew Results
A core update lands in the middle of your change window and you can't separate its impact from yours.
Gut-Feel Decisions
Teams roll out changes site-wide based on a single chart that looked good for a week, then can't explain why it didn't hold.
No Idea How Long to Wait
Stop too early and you're reading noise. Wait too long and you've lost weeks you could have spent on the next experiment.
Spreadsheet Statistics
Exporting GSC data and building your own comparison model takes hours, and one formula mistake gives you the wrong answer.
Results Nobody Trusts
Stakeholders push back on SEO wins because "traffic went up" is not the same as "our change caused it".
Real Experiment Design,
Without the Data Science
Pick the pages you changed, let SERPView suggest a control, and get a verdict that updates every day as new Search Console data arrives.
Flexible Test Groups
Define the test group by URL pattern, a set of queries, an existing content group, or a saved custom filter.
Suggested Control Group
SERPView finds the page that tracked your test pages most closely before the change and suggests it as the control. You can also pick your own.
Difference-in-Differences Analysis
Measures how the gap between test and control changed after your change, so sitewide swings and seasonality cancel out.
Confidence and Lift Range
Every test shows the estimated lift in percent, a 95% interval from 10,000 bootstrap samples, and how confident you can be that the change helped.
Power Estimator
See the smallest lift your test can reliably detect and how many days it needs, before you commit to a test window.
Core Update Detection
If a Google core update overlaps your test window, the result is flagged with a caveat so you don't over-read it.
Real-World
Use Cases
See how teams use SEO A/B Testing to back their changes with evidence.
Situation
Rewrote titles on 40 product category pages to include the price range
Result
Compare clicks on those pages to unchanged categories and see if the lift is real before rolling it out to all 400.
Situation
Updated a batch of old blog posts with new sections and FAQs
Result
Measure the impact on impressions and position against posts you didn't touch, with seasonality removed.
Situation
Client asks whether last quarter's on-page work actually paid off
Result
Show a verdict with lift, confidence range, and a clear chart instead of a before-and-after screenshot.
Situation
Added contextual links from top pages to a group of underperforming guides
Result
Track the guides as the test group and see if the links moved clicks, not just rankings.
Situation
Added FAQ or product schema to one section of the site
Result
Test CTR against a similar section without schema to find out if rich results are worth rolling out everywhere.
Situation
Need to prioritize which experiment to run next
Result
Use the power estimator to pick tests that can actually show a result in the time you have.
Frequently Asked
Questions
Everything you need to know about SEO A/B Testing in SERPView.
SEO A/B testing (also called SEO split testing) compares a group of pages you changed against a similar group you didn't change. Because Google can't be shown two versions of the same URL, you test on groups of pages instead and compare how each group performs over the same period.
SERPView uses a difference-in-differences method. It looks at the daily gap between your test group and control group before the change (the baseline window) and after it (the test window). The change in that gap is your lift. It then resamples the daily data 10,000 times to estimate confidence and a 95% range for the lift.
A URL pattern (for example all pages under /blog/), a set of queries, one of your existing content groups, or a saved custom filter.
No. SERPView can suggest a control by finding the page whose traffic moved most closely with your test pages during the baseline window. You can also define your own control with filters.
Clicks, impressions, CTR, or average position from Google Search Console. Clicks is the default.
The baseline and test windows are set in whole weeks, from 1 to 13 weeks each, with 4 weeks as the default. The power estimator tells you how many days your test needs to detect a meaningful lift based on how noisy your data is.
A test can be a likely winner, likely loser, inconclusive, or show no detectable change. If the test and control groups didn't move together before the change, the test is marked as a low-confidence control so you know not to trust the number.
SERPView checks Google's official update history. If a core update overlaps your test window, the result shows a caveat so you can decide whether to extend the test or rerun it.
Yes. Tests update daily and conclude automatically when the end date passes. You can get the final result by email or in Slack, and start and end annotations are added to your charts automatically.
SEO A/B Testing is a premium feature and is not available on the Free plan.
Related Features
Tools that help you set up, track and explain your SEO tests.
Content Groups
Group pages by section or topic and use a content group as the test group in one click.
Custom Annotations
Mark site changes on your charts. SEO tests add start and end annotations automatically.
Custom Filters
Save filters for queries and pages, then reuse a saved filter to define a test group.
Stop Guessing,
Start Testing
Set up your first SEO experiment in a few minutes and get a verdict you can defend, backed by your own Search Console data.