🔥 Want your next Google Ads test to produce a clear winner?
In this video I break down five A/B tests I run in Google Ads to improve conversions and sales, covering keyword match types, bidding strategies, landing pages, and ad variations.
I also show how to set up Drafts and Experiments, choose the right metric, and read the test results before updating the original campaign.
By the end of this video, you’ll know which Google Ads tests to run first, how to set them up with Drafts and Experiments, what metric to track, and how to interpret test results so you can apply winning changes to your original campaign for more conversions and sales.
⏩ Skip to Section
0:00 Why you need to test in Google Ads
1:10 How to set up Experiments (Drafts and Experiments)
1:59 A/B Test 1: Maximise Conversion Value vs Conversions
4:38 A/B Test 2: Broad vs Phrase vs Exact Match
6:36 A/B Test 3: Landing Pages
8:07 A/B Test 4: Manual CPC Bidding vs Maximise Clicks
10:29 A/B Test 5: Ad Creatives and Audience Signals (Performance Max)
12:11 Recap: All 5 tests and what each one tells you
13:34 How long to run tests and get enough data
14:16 Final takeaway
📺More videos from The Conversion Clinic
📚 Related resources
❓FAQs
Q1: What is an A/B test in Google Ads?
I compare an original version vs a different ad or setting, changing one variable so I can see what drives more sales.
Q2: Can I test in Google Ads without rebuilding my account?
Yes, Drafts and Experiments lets me test changes inside the same Google Ads account and keep the original campaign intact.
Q3: What should I test first if I want more conversions?
I start where volume is highest, usually the landing page offer or ad copy in the top campaign.
Q4: How long should a split test run?
Long enough to get enough data and cover normal buying days, not just a random 2-day spike.
Q5: Which metric should I use to pick a winner?
I choose one primary metric per test, then check conversion value and cost so I do not overfit.
Q6: Should I test match types in the same campaign?
Yes, broad, phrase, and exact can behave like totally different intent buckets, so I control it with negatives and tight structure.
Q7: Can I test bidding strategies like Manual CPC vs Maximize Clicks?
Yes, but I keep everything else stable and judge it on lead quality and cost, not just clicks.
Q8: What counts as a good landing page test?
One clear change like headline, offer, layout, or speed, so I know what improved conversion behaviour.
Q9: How do I avoid misleading test results?
I test one variable at a time, avoid tiny budgets, and verify tracking before I optimise.
🌐 Connect with me
#googleads #splittesting #googleanalytics #abtesting