Performance max campaign – lights and shadows. Read how we achieved ROAS of 1134% in two months

Judyta Lis Judyta Lis

Performance Max campaigns are a new type of Google Ads campaigns that were loud in the last quarter of 2021. At the beginning of 2022, Google announced the automatic transformation of smart product campaigns into Performance Max campaigns from July 2022 and local campaigns from August 2022. The ability to manually update smart product campaigns to the Performance Max campaign is expected to be available in the second quarter of 2022.

In the face of major changes related to the automatic transformation of smart product campaigns into Performance Max campaigns, we are constantly looking for new opportunities and optimization of a new type of campaign to prepare in advance for the upcoming changes. 

What are Performance Max campaigns?

Performance Max campaign, as indicated by Google, allows advertisers to access all of Google Ads’ advertising content within a single campaign. We are dealing here with a kind of “all in 1”, this campaign has the possibility to use all advertising channels – search and advertising network, YouTube, Gmail, Google Maps and Discovery. Therefore, it combines product campaigns, search network campaigns, advertising network campaigns, Discovery campaigns and local campaigns. Also external price comparators (CSS) allow you to create Performance Max campaigns.

As Google points out, in the Performance Max campaign, optimization is based on the interaction of different types of automation from Google: setting rates, budget optimization, creation, attribution and list of recipients.

We can use Performance Max campaigns not only for e-commerce accounts, but also for accounts whose purpose is to obtain leads. It can use 4 automatic rate optimization strategies: maximize conversion, maximize conversion with fixed CPA, maximize conversion values, and maximize conversion values with fixed ROAS.

Performance Max campaign, using and analyzing the output data, tries to increase the number of conversions or the value of conversions. Such data are primarily:

  • signals concerning recipients – remarketing lists, customer lists, market segments of recipients, segments of recipients with similar interests, custom lists, similar recipients and demographic data,
  • high-quality images, videos, ad texts, extensions (new or available in your account) created in the so-called resources,
  • conversion data, for which we optimize a given campaign, such as transactions or sending a contact form.

One of the most important issues related to the performance of the Performance Max campaign while other types of campaigns are running at the same time is that:

  • Performance Max campaign has a higher priority than the Smart Shopping campaign, if we have the Smart PLA campaign and the Performance Max campaign targeting the same products,
  • campaigns in the search network:
    • when using [exact] matches – priority will always be given to campaigns in the search network with this match,
    • in the case of using broad or “phrase” matches – higher quality campaigns will take part in the auction,
  • video campaigns – higher quality campaigns will take part in the auction,
  • Discovery campaigns – higher quality campaigns will take part in the auction,
  • local campaigns – higher quality campaigns will take part in the auction,
  • advertising network campaigns – if we use the dynamic remarketing list in advertising network campaigns – people who have previously viewed the product – the Performance Max campaign will have priority, in case of other targets (also other dynamic remarketing lists) – higher quality campaigns will participate in the auction.

Performance Max campaign – best practice

The question we are currently facing and are actively looking for answers to is how to prepare the Performance Max campaign so that it generates satisfying results for us. We will try to present the best practice for this type of campaign based on Google guidelines and our own experience.

When preparing the Performance Max campaign, make sure to use the maximum amount of high-quality resources (including video) and if we have appropriate signals on our Google Ads account regarding recipients – customer lists, remarketing lists, lists of similar recipients, custom lists – it is best practice to use lists with high historical effectiveness in the past. Similarly, if you have historical data on the effectiveness of the list of recipients in the market and recipients with similar interests, it is also recommended that you add the lists to the Performance Max campaign.

If you do not have historical data on the performance of your audience, Google recommends that you do not add such signals. Then the campaign, using machine learning, will try to reach the most effective group of recipients.

Google recommends optimizing Performance Max campaigns using either ads conversions with value or conversions taken from GA4.

Bearing in mind that the Performance Max campaign has priority over the Smart Shopping campaign when targeting the same products/product categories, it is good practice to divide products between two campaigns at the start of the Performance Max campaign tests. The products can be divided e.g. on the basis of data about the lower number of clicks in the past and the number of views in the smart PLA campaign, and these products can be separated into the Performance Max campaign on the basis of product categories with lower effectiveness, or on the basis of product categories with higher effectiveness (e.g. bestsellers) in the past in smart product campaign.
If we have and actively run a blog on our website, due to the fact that Performance Max also uses the capabilities of DSA (dynamic ads in the search engine), it is worth remembering to exclude such subpages :

Performance Max Campaigns - settings

Google helps to exclude branded keywords at the Performance Max campaign level if we do not want the campaign to display phrases related to our brand.

Case Studies

Win – so much to win

Our winning case concerns a brand from the clothing industry. We have used all available resources to the maximum extent possible. The result? ROAS 1134% within two months of campaign operation.

What did that look like?

In mid-January, we created Bestsellers with separate products of women’s and men’s clothing (a total of 465 products). The campaign was optimized for ads conversion – Shopping. In the first month, the communication in the resources included Sale -50% / -70%. In February, there was a communication of the Valentine’s Day promotion and the new collection.

Performance Max Campaigns - results

For the first month and a half, the campaign was optimized to maximize values without ROAS. We set ROAS 900% at the beginning of March.

Performance Max Campaigns - results

The results of the campaign in separating the products were very satisfactory. In this case, the campaign also found a kind of niche, which it used to increase the value of conversions – mainly men’s products were sold on the account, and the Performance Max campaign prioritized women’s products in views, which increased the value of conversions in these periods. Importantly, the campaign did not affect the effectiveness of the Smart Shopping, Brand or DSA campaigns.

Fail – so much to lose

Unfortunately, Performance Max does not work always and everywhere. In our second case – a little-known brand from the home furnishing industry – the choice of this type of campaign turned out to be a fail.

What did that look like?

The campaign was launched in February and initially its results looked good:

Performance Max Campaigns - results

Unfortunately, in March the results deteriorated significantly:

Performance Max Campaigns - results

For this account, statistics over two months indicate that with low brand awareness and no other marketing efforts, Performance Max was unable to produce satisfactory results, and even, over time, the results dropped significantly.

Despite the fact that the campaign had a lot of data (almost 2 years) and in the past there were many more conversions, the use of Performance Max did not improve the account results.

This example also indicates that the Performance Max campaign may benefit to a large extent from the brand movement, as a result of which, with a weak brand, its efficiency may be very low.

What now?

Each of us will be transitioning to Performance Max campaigns in the near future, especially in the case of e-commerce accounts.

Looking at the indicated examples and best practices when creating Performance Max, for appropriate optimization of a new type of campaign, it is already worth testing various configurations by, for example, separating a certain group of products, creating various advertising resources and analysing reports and statistics.

Conducted tests will allow the campaign to obtain the appropriate amount of historical data, and teach us to analyse the results and increase the efficiency of Performance Max. This is all the more important because in 2022, smart product campaigns will be automatically transformed into Performance Max campaigns.

Find out more about
Performance Max

Write to us