Key tools to run successful Performance Max campaigns

If you haven’t tried Performance Max yet or want to compare it to your standard Shopping campaigns before making the switch, experiments are available to help you run tests and compare performance. You can set up, manage, and optimise two types of tests in the Experiments page of your Google Ads account:

• Performance Max Uplift experiment: Learn how adding a new Performance Max campaign to your current campaign mix can drive incremental conversions or conversion value.

• Performance Max vs. a standard Shopping campaign: Measure the boost of switching from a standard Shopping campaign to a Performance Max campaign targeting the same products. Discover how the results may improve and whether you should start replacing your Shopping campaigns¹.

Next steps for experiment creation:

a) Link a clear hypothesis to your business goals and define the experiment’s purpose

This can help you better determine if your experiment was successful or not. Ensure you pick one or two metrics maximum, such as finding more converting customers or driving better performance against your goals, that are relevant to your experiment before the test begins.

b) Create your Performance Max test

Make sure you use the same settings, targeting bidding strategy, and Cost Per Action (CPA) or Return On Ad Spend (ROAS) as measurement data points in the campaign’s account. Comparing the exact conversion goals allows you to easily tie the impact of your changes to the test.

Allow a 1–2 week learning period and run your test for at least 4–6 weeks to get more accurate results. Remember to exclude existing campaigns from the experiment that are not optimising for the same goal as the one in the experiment, such as branding campaigns like YouTube for awareness and cost-per-thousand impressions display.

c) Analyse results and choose experiment winners

Analysing your results can help you understand and quantify whether the test outcome is due to what you’re testing or by chance. Let the campaign ramp up for at least one or two weeks and exclude this duration from the experiment evaluation. Also consider checking conversion lag time by visiting ‘Attribution’ > ‘Path metrics’ > ‘Average day to conversion’.

Remember: Keep records of your experiments. Documenting learnings from past experiments can help you tap into them for your next campaign, tracking and benchmarking the impact of your efforts over time. Following this method can help you decide whether to launch or pause your Performance Max campaign. When experiment results are statistically significant, you can feel confident that they’re driven by what you’re testing given the sample you ‘ve provided.

Several ecommerce companies have found success with Performance Max, including our client Bricozor.

Bricozor is a French hardware company with over 56,000 products for building, furnishing, bathroom equipment, heating, plumbing, and power tools. Bricozor’s main objectives were to grow revenue and improve their return on investment. To gain a competitive advantage, Bricozor decided to modernise their Shopping campaigns strategy by testing Performance Max.

The approach

The initial phase of Bricozor’s new Shopping campaign strategy was to set up an A/B test to compare the performance between standard Shopping campaigns with a target ROAS bidding strategy and Performance Max campaigns. By following the Google Shopping (CSS) team’s expert advice, Bricozor was able to test Performance Max in a way that aligned with their objectives while reducing risk. After months of testing their new automation strategy, Bricozor found that Performance Max campaigns had 10% higher ROAS and drove over 15% in revenue growth compared to standard Shopping campaigns with target ROAS. You can read more about their success story.

One of the biggest changes that Performance Max brought is the addition of creative assets. Performance is usually optimised when using a wide range of text, image, and video assets for each asset group within a campaign, so you can meet more customer preferences through more channels.

You can easily check if your asset group has a good spread of assets across all available types. By looking at the ad strength of your asset group, you’ll receive a score based on the relevance, quality, and diversity of your ad copy. This score can range from ‘Incomplete’ to ‘Poor’, ‘Average’, ‘Good’, and ‘Excellent’. Aim for ‘Excellent’² to ensure Performance Max can participate in all available channels with enough assets to generate ads.

As you add more assets³ – up to the maximum amount – and increase their diversity, your Performance Max campaign will have more content to combine and test. That way, you can find out which mix of elements can produce the best results. The score can give you a basic understanding of the Ad Strength of your asset group. To help you more with that, we designed the table below with specific suggestions as per the recommended number of assets and specifications for each asset type.

For more expert advice on Performance Max, you can contact your Google Shopping (CSS) team.