In the world of digital advertising, guessing is expensive. A/B testing (or split testing) allows you to compare two versions of an ad against each other to see which one performs better, ensuring your marketing budget is spent on strategies that actually convert for your Brisbane or Australian-based business.
By testing variables like your image, headline, or audience, you remove the guesswork and let data drive your decisions. This guide will show you how to set up a formal A/B test using Meta Ads Manager's built-in tools.
Prerequisites: What You’ll Need
Before you start, ensure you have the following ready:- An active Meta Business Suite account.
- A Facebook Business Page linked to an Ad Account.
- A valid payment method (and your ABN registered in the billing section to avoid unnecessary tax issues).
- The Meta Pixel (or Conversions API) installed on your website if you are testing for conversions.
- At least two different versions of an ad element (e.g., two different photos or two different headlines).
---
Step 1: Navigate to Meta Ads Manager
Log in to your Meta Ads Manager. Ensure you are in the correct ad account for your business. On the main dashboard, you’ll see the green "+ Create" button.Screenshot Description: Look for the bright green button on the top left of your campaigns table. This is the starting point for all new ad activities.
Step 2: Choose Your Campaign Objective
Select the objective that aligns with your business goals. For most Australian small businesses, this will be Leads, Sales, or Traffic. Note: You can only A/B test ads that share the same objective. You cannot test a 'Brand Awareness' ad against a 'Sales' ad.Step 3: Enable the A/B Test Toggle
After selecting your objective and naming your campaign, look for the A/B Test toggle. It is usually located at the Campaign level settings. Switch this to On.Facebook will display a message explaining that after you publish this campaign, you will be prompted to create a version 'B' to test against this version 'A'.
Step 4: Set Your Campaign Budget (Advantage Campaign Budget)
Decide whether you want to set your budget at the Campaign level or the Ad Set level. For a clean A/B test, Meta often recommends using a Daily Budget. Pro Tip: Ensure your budget is high enough to generate significant data. If your daily spend is too low (e.g., $5 a day), it may take weeks to get a statistically significant result.Step 5: Define Your 'Version A' Audience
Progress to the Ad Set level. Here, you define who will see your ad. For your first test, we recommend keeping the audience broad or using your best-performing 'Local Brisbane' or 'Australia-wide' demographic.Screenshot Description: You should see the 'Audience' section where you select locations, ages, and interests. A map of Australia should be visible if you’ve selected the correct targeting.
Step 6: Create Your 'Version A' Creative
Now, move to the Ad level. Upload your first image or video, write your primary text, and add your headline. This is your 'Control' ad. Make sure your 'Call to Action' (CTA) button (e.g., "Book Now" or "Learn More") is clear.Step 7: Publish Version A and Start the Comparison
Click Publish. Once the campaign is published, Meta will trigger a pop-up window asking you to "Create A/B Test." This is where the magic happens.Step 8: Select the Variable to Test
You will be asked what you want to change for the second version. Common variables include:- Creative: Different images, videos, or ad copy.
- Audience: Testing two different age groups or interest sets.
- Placement: Testing Instagram Stories vs. Facebook Feed.
Step 9: Configure Version B
Meta will now create a duplicate of your campaign/ad set. Change the specific element you decided to test. For example, if you are testing creative, swap out the image for your second option. Everything else must remain identical.Step 10: Set the Test Duration and Power
Choose how long the test should run. Meta typically recommends a minimum of 3 to 14 days.- Estimated Test Power: Look for the 'Test Power' percentage. You want this to be above 80% for reliable results. If it’s lower, you may need to increase your budget or run the test for longer.
Step 11: Determine the Winning Metric
Select how Facebook should determine the winner. Usually, this is Cost per Result (e.g., the cheapest cost per lead) or Cost per 1,000 People Reached. For most service-based businesses in Australia, "Cost per Result" is the most meaningful metric.Step 12: Review and Launch
Review your settings one last time. Ensure the budget is split 50/50 between the two versions. Click Confirm or Publish Test.---
Common Mistakes to Avoid
- Testing too many things at once: As mentioned, changing the audience and the image simultaneously ruins the experiment. Keep it simple.
- Ending the test too early: It’s tempting to stop a test after 24 hours if one ad looks like it's winning. Resist this! Facebook’s algorithm needs time to "learn" and stabilise.
- Ignoring the 'Winner' notification: Meta will email you when the test is complete. Make sure you actually go back into Ads Manager to turn off the losing ad and scale the winner.
- Small Audience Sizes: If your target audience is a tiny suburb in Brisbane with only 2,000 people, an A/B test won't have enough data to be accurate.
Troubleshooting Common Issues
Problem: My test ended with "No Clear Winner."- Solution: This happens when the performance difference between A and B is too small to be statistically significant. This is actually a result! It means the variable you tested doesn't heavily influence your audience's behaviour. Try testing a more drastic change next time (e.g., a video vs. a static image).
- Solution: Check your bid caps. If you’ve set a manual bid too low, Facebook won't be able to win any auctions. Switch to "Highest Volume" (automated bidding) for more consistent spending during a test.
- Solution: If the toggle didn't appear, you can also create an A/B test after your ads are live by selecting two existing campaigns in the Ads Manager table and clicking the "A/B Test" button in the toolbar above the list.
---
Next Steps
Once you’ve found a winning ad, don't stop there. Take the winner of this test and test it against a new challenger. This process of "continuous improvement" is how top-tier Australian brands lower their acquisition costs over time.Need help interpreting your data or setting up a complex funnel? The team at Local Marketing Group specialises in helping Brisbane businesses maximise their ROI. Contact us today for a strategy session.