Announced at Google's Marketing Livestream on May 27, 2021, Performance Max campaigns arrived in Beta in several countries, including France.
According to Google, this new type of goal-based automated campaign significantly improves performance. Performance Max campaigns allow ads to be served across the full Google advertising inventory: Search, Display, Discovery, Gmail, Maps, and YouTube. They therefore allow you to reach a wider audience.
And because we are always looking for innovation in our acquisition strategies to achieve better performance, datashake tested these new campaigns. Here are our first takeaways one month after their launch!
How to launch a Performance Max campaign?
If you want to set up a Performance Max campaign, you first need to:
- Submit a request to Google
- Once your request is approved, create a Performance Max campaign by choosing the desired goal relative to what you are tracking (Online sales, Lead generation, Offline sales).
- Make sure you have campaigns with the same objective running in parallel (in order to best analyze performance).
- Provide several elements that will allow the algorithm to find the best ad combinations:
→ at least 5 versions of text assets (recommended: 3 headlines of 30 characters, 2 descriptions of 90 characters, 1 long headline of 90 characters)
→ Call to Action
→ at least five versions of image assets, videos of at least 10 seconds, logo.
- Bidding and targeting are fully automated. However, it's possible to define relevant audiences as signals to help the algorithm.
- Use Google Ads conversion tracking rather than Google Analytics conversion import to track video view / cross-device conversions.
- Ideally plan a budget of 10,000 euros per month for the campaign. Google's recommendation is to allocate a daily budget that is at least 3x your target CPA.
Once launched, you will notice that it is difficult to draw relevant insights from the results obtained. We do, however, have a few analytical leads to share.
Tips for analyzing this campaign's performance
According to Google, the new beta should run for a minimum of 6 weeks to gather enough data and be able to compare results. You should therefore give the algorithm 15 days to exit its learning phase, and then review campaign results the following month.
To best analyze performance, it is necessary to do a before-and-after comparison following the addition of the Performance Max campaign. Check whether the beta campaign achieves better performance than the overall account for the targeted goals.
In addition, it's possible to get a sense of how the assets provided are performing via the performance indicator, which shows whether the integrated assets are of "low," "average," or "good" quality.
This is what we applied to interpret the results of our Performance Max campaigns, and here are the first numbers!
Our feedback on the Google beta
We have tested Performance Max within the strategies of several advertisers, and our initial feedback is quite positive! Here is a quick snapshot of the numbers one month after the beta launches.
For one of our advertisers in the Finance sector, the CPA was cut in half and CTR increased by 50%. The conversion rate is higher with an attractive cost per conversion.
One of our advertisers in the Health sector achieved a cost per lead 4x lower, and a declining CPC. The conversion rate is above the account average. One question remains: is the CPA lower because the campaign is running heavily on lower-cost channels, or is it bidding on brand-related keywords?
Looking at the consistent performance on our brand campaign, there appears to be no cannibalization between the two campaigns, which is a good sign!
Finally, for one of our advertisers whose goal is lead generation, results improved day after day as the campaign exited the learning phase. The CPA started at 66 euros at the beginning of the campaign before eventually dropping to 13 euros. The average CPC is lower than the account average (0.50 euros vs. 1.60 euros), while the conversion rate is 2x lower. For this brand focused on lead generation, the Performance Max campaign is interesting for driving traffic to the website, but it brings in a less qualified audience.
Initial feedback is therefore fairly positive. That said, Performance Max campaigns do have some drawbacks.
The downsides of Performance Max campaigns
While these new automated campaigns claim to be more effective, it is currently impossible to know where our ads are being served. Automatically placed across Google's various channels via machine learning, Google does not communicate on the breakdown of activated levers. This makes it difficult to draw insights about our audience.
We also have little control over our ads, whose combinations are generated automatically, meaning very few optimizations are possible.
Another point to watch closely is campaign cannibalization with Performance Max. You need to remain vigilant about how your campaigns are served, and observe whether classic campaigns continue to generate consistent impressions and results.
Add to this the recommended budget for this type of campaign, which is 10,000 euros per month.
To sum up, we can say that initial tests are encouraging! And while Performance Max campaigns remain something of a black box that's hard to analyze, this beta version may still evolve.
Want support with Google Ads or, more broadly, with your digital marketing strategy? Don't hesitate to contact us!
Sources: Google analyses, Google source, BDM, Adcore article
.jpg)
.jpg)