Our thoughts on Performance Max campaigns in Google Ads

The Google Ads Automation timeline started back in 2011 with the launch of Dynamic Search Ads. A few years later Smart Bidding entered the game. In 2018, Responsive Search ads appeared as well as Smart Shopping. And last year, Google forced advertisers to switch from Broad Match Modifier to Phrase keywords. So it’s not surprising that Google continues to expand its way towards automation with Performance Max campaigns.

Google wants to push its ultimate solution to answer customer demand and its ongoing shifts. For the last five years, Google’s strategy has been aiming toward creating a fully automated Google Ads platform. This is part of Google’s bigger vision: full automation.

Performance Max at a glance

Performance Max is a new goal-based campaign type and way to buy traffic on the Google Network. It allows advertisers to run one single campaign combining all of Google’s channels: YouTube, Display, Search, Search Partners, Shopping, Discover, Gmail, and Maps.

The campaign setup is pretty simple: you upload your assets (image, video, text), pick a conversion, set a target and Google does the rest.

By the rest, we mean:

• Unlock unexpected new audiences across Google’s channels and networks thanks to the real-time understanding of consumer intent coupled with audience signals
• Drive better performance against your goal (increasing conversions and value)
• Get more transparent insights thanks to Performance Max asset reporting that helps you understand which creatives are impacting performance and help you optimize campaign creatives to drive ROI.

Sounds pretty convincing, doesn’t it? At least, that’s the promise that’s being given to advertisers.

Let’s see what the actual performance is telling us after having run the test on one of our client’s Google Ads accounts.

Learnings and thoughts about our Performance Max test

With Performance max, you don’t have the possibility to deep dive into your results as you would with classic Search campaigns. However, as a PPC specialist, you like to have control over your campaigns and optimize towards the best performances. This includes adding negative keywords (we will develop that aspect in a few moments), being more aggressive on certain keywords, excluding placements, allocating more budget to a certain segment that performs well, etc.

A new ‘Insights’ tab

With Performance Max, you gain access to a new tab which is called “Insights”. It only gives you different search categories (by regrouping search terms into a theme) with metrics such as Clicks, Impressions, and Conversions. So, it clearly shows the lack of transparency with this new type of campaign. You still have to accept the results you get, there is almost no way to act on these “insights”.

Placements

Another big concern is the placements where your ads appear. You don’t even have a report that actually shows your placements’ performance. This also means that your ads are appearing on Search Partners sites and you can’t do anything about it. This is confirming the lack of control you, as an advertiser, have with Performance Max.

Difficulty in accessing data

When speaking about results, you don’t even have the opportunity to access the performance that you have in the “Insights” tab, if your campaigns are paused. You can only access data until 2 months in the past. You’re lucky if you paused your PMCs less than 2 months ago. Otherwise, you cannot access any of the Insights. Another example of the lack of transparency from Google’s side.

On the other hand, it’s not surprising that Google doesn’t give more control over this new campaign type. If they would give more transparency or insights, advertisers would be inclined to split out a certain keyword/targeting into a new campaign, which wouldn’t work. This is in line with the idea that we have to accept the results.

Adding negative keywords

Let’s get back to what’s mentioned above about the search terms exclusions. There is no way to exclude search terms yourself. As you know, this is the basics of Google Ads optimizations.

But wait, this is your lucky day! To add negative keywords, you can ask your Google Rep to insert them for you. Why wasn’t it included in the beta test? It’s such a basic feature that we are missing… However, Google is rolling out this feature in the future. However, these will be Account-level negatives which, depending on the preferred campaign structure, also come with limitations.

The icing on the cake about keyword exclusion: if the negative keywords are added, then the Performance Max Campaign will not serve on any Google inventory. It will affect all the networks.

Excluding brand name

What if you want to exclude your Brand name, as there’s a good chance this traffic doesn’t bring in incremental traffic/leads? See one of our latest articles about branded paid search. Would adding the brand name as a negative also exclude the contextual or audience-based targeting in Display, resulting from the brand name keyword as a Display-targeting option? At this point, we are not able to answer this question since the data are hidden.

Is Performance Max stealing traffic from your Brand Search campaigns?

You may think that Performance Max would be stealing Brand terms from your existing Brand Search Campaign. That’s where it becomes interesting!

If you have an existing Search campaign with keywords that exactly match a user’s search query (regardless of match type), the ads from those Search campaigns will be prioritized over Performance Max. For the test, we asked our Google Rep to “link” the two accounts for our client. In one, we ran Display campaigns, which included the Performance Max Campaigns, in the other, we ran the Search campaigns.

Here is the misunderstanding that we had. We thought that “exactly” was related to a match type, but it has nothing to do with the Exact match type. It means “if it is written in the same way”. In other words, if Google recognizes a synonym or a close variant.

It all comes down to saying that your Brand Search Campaign will be prioritized over PMCs.

This is theory and the way of working of Performance max. But let’s compare it with the actual outcomes of the PMCs we ran.

The client we ran these campaigns for, has Brand & Generic Search campaigns on top of the PMCs. Here are the results in terms of Leads that both types of campaigns did generate:

Seeing the results, Performance Max Cost per conversion isn’t that bad compared to the CPL of the Generic Search campaign, isn’t it? We took a closer look at the performances of the PMCs (let’s note that we have access to the results because we had them stored internally, otherwise we couldn’t have access to them).

According to the saved results, 132 clicks of the PMCs (7.6% of the total PMC traffic) came from Search and generated ALL the conversions (92). From the limited data we have, we could assume that Search had generated all the conversions. And knowing that most clicks didn’t come from Search, we would come to the conclusion that 92.4% of the traffic didn’t generate a single conversion.

Is it wasting money? Again, it’s impossible to know because all of that is an assumption…

The future of Google Ads: Performance Max only?

We already know that by the end of 2022, Smart Shopping and Local campaigns will automatically be upgraded to Performance Max campaigns. This new all-in-one style campaign is a further indicator of the future of Google advertising and the power of machine learning.

Let’s be clear, this won’t be the last time we are going to discuss Performance Max and its functionalities. We do believe that Google is going to continue developing this campaign type by providing more formats and inventory to expand campaign reach, smart bidding advancements, and gradually improve insights, reporting, and tools made available to marketers, adding negative keywords yourself for example.

New Customer Acquisition goals

Recently, Google announced that PMCs gained New Customer Acquisition goals. Performance Max is getting a new optimization feature aimed at driving new business. This will allow advertisers to optimize to those who are not previous customers. Google will use first-party signals:

• Customer Match lists: these lists are made of first-party data including email, phone, first name, last name, country, and zip code. When uploaded Google can evaluate how many of your customers have a matching Google account that can be used in targeting.
• Conversion tags: the use of conversion tags for purchases can help these new Performance Max goals to exclude people who are not new.

It’s a decent feature that Google brought to advertisers. Without that feature, you can be sure that the algorithm goes into a self-reinforcing feedback loop when it sees good performance on existing customers. This results in higher bidding on existing customers which is actually not creating new business.

Conclusion

All that being said, I’m truly convinced that Machine learning is great, but it’s always better when a human with common sense can push it in the right direction. As soon as such a campaign type will be easier to have “full” control over, on top of smart bidding, it could become a very interesting campaign type that could save you time (by managing all at once). Let’s be optimistic about the future and continue testing new features!