UpriseUp - Up
UpriseUp - Rise
UpriseUp - Up
Back to EventsBack to Blog

Google’s Performance Max Campaigns: The Need to Knows

Robot hand holding up a tablet with a lightbulb, representing machine learning in Performance Max campaigns

Performance Max or Mad Max?

Recently, there has been a concerted effort by Google to move Paid Media advertisers towards more automated solutions, from bidding strategies to responsive search ads. One element of this automated push is the Performance Max campaign type. Google claims that this campaign is the crowning achievement of automation; it’s able to perform complex advertising tasks with minimal input and produce strong results for advertisers. We therefore want to give you the inside scoop on what a Performance Max campaign is, how it works, and share our initial impressions are after testing them ourselves.


What is a Performance Max campaign?

Performance Max campaigns are essentially a merging of many of the different “smart” campaigns that previously existed in the Google Ads system. It allows an advertiser to combine search, display, video and shopping in a single campaign, which will utilise those channels to try and maximise results. Depending on the creative you supply to the campaign some of these channels may not be used (video requires video creative for example).

Additionally, many of the smart campaigns previously mentioned, such as smart shopping and smart video, no longer exist. The only way to advertise on these channels in an automated manner is through Performance Max.


How Does a Performance Max Campaign Work?

Asset groups are at the core of performance max campaigns. These are assets that you supply to the campaign to use in advertising. This includes text, such as headlines and descriptions, images, videos and product feeds.

Table of asset groups for performance max campaigns

These are then mixed and matched to create the ads used by the campaign. For example, an image might be combined with a headline and a call to action to create a display ad.

Very recently, internal documents from Google suggested that they were testing using generative AI to change your submitted asset groups based on audience signals. This is very new tech, and likely won’t be implemented for months, if at all, but keep a close eye on this. If it does get implemented, it may mean that your asset groups become more of a suggestion than a rule.

In addition to asset groups, you can add targeting options such as audiences to the campaign. You cannot manually set bid adjustments on these audiences, but any audiences added will be used by the campaign to optimise performance.

Once your campaign is live, rather than seeing results broken down to individual keywords or ads, you will be viewing the insight page:

Screengrab of the performance max campaign insights page

These insights will change depending on the data of your specific campaign, but they will always be broken into aggregated buckets, such as the search term report above. Don’t expect to be drilling into the data in the same fashion as you would in a manual campaign.

It’s important to note that advertisers cannot manually bid using Performance Max campaigns – you must use one of the available automated bidding strategies. The amount of granular data and targeting options available to you will be less than a manually ran campaign (for example, you will not have as much search term data as you would from a manual paid search campaign).


Our First Impressions

From our tests, Performance Max campaigns had a rocky start. Initially, results didn’t seem to be able to compete with manually managed campaigns. However, as is often the case with new Google Ads features, after around a year of development the system appears to have improved significantly. Now, they have reached the point where a well-managed performance max campaign can at least equal manual control, whilst saving on the need to do granular management.

On the surface, Performance Max campaigns inhabit a strange spot in the Google Ecosystem, requiring large amounts of preparation to set up, but low amounts of time to manage. However, we found that significant changes needed to be made before any of our campaigns were performing similarly to manual campaigns. Do not treat these as a set and forget solution to marketing, they still need optimisation on a manual level.

We also found that Performance Max campaigns don’t necessarily suit all advertisers and may only fit best for certain scenarios. So, let’s dive in to who these campaigns are for, and some strategies to maximise their effectiveness.


Who Are Performance Max Campaigns For?

The first key requirement for running a Performance Max campaign is that you must be an advertiser wanting to run ads over multiple channels. Although you can run a Performance Max campaign through a single channel, we have found that performance is never as good as manually running a single channel campaign.

Performance Max campaigns also require enough data to optimise it’s algorithm, like any machine learning solution. We wouldn’t recommend Performance Max campaigns for advertisers with a tight, minimal budget, as you won’t be able to provide the data necessary to feed the algorithm to achieve effective learning. Once again, manual campaigns would be the better option for you.

If you are an advertiser willing to run over multiple channels and are happy to push a reasonable amount of spend through your campaign, you’re ready to go! Let’s move on to how to make these campaigns shine.


Tips for Successful Performance Max Campaigns


A Well Fed Algorithm is a Happy Algorithm

The single most important factor in Performance Max success is ensuring that you are giving the algorithm as much data as you possibly can. This starts with asset groups.

Making sure you are adding as many assets here as possible is a vital step in ensuring performance. The more assets added to the campaign, the more variations of ads the campaign will be able to test, and the more optimised the ads will become.

Similarly, the more targeting options you can give to the algorithm, the better it will be at targeting the most relevant users. Things like locations and audience definitions should be included, even if you’re not looking to target those segments specifically. If the Performance Max campaign sees that users from these segments convert more readily, it will target them, but if they do not it will ignore them. The important thing is to give the campaign this lever to pull.


Organisation Still Matters

Although a Performance Max campaign says it can run all your activity through one campaign, that doesn’t mean you should let it. We’ve found that performance Max campaigns have a tendency to just focus on the top performing products or audience segments in its targeting, often completely ignoring others. This is fine if those other options are not performing well, but often they were performing just fine.

The solution to this is to operate using multiple Performance Max campaigns, splitting out your targeting and ensuring that you are not condensing your spend into a few sets of targeting options.

This also helps with tailoring creative. You only have a certain number of creatives you can add per campaign. Often you won’t have a one size fits all creative for your ads. Splitting your campaigns up can also allow you to tailor images and text in a similar way to ad groups in manual campaigns.


Take Advantage of the Insights Tab

The insight tab is the one window you get into the inner workings of your Performance Max campaign, and it’s critical to monitor it if you want to maximise your performance. Early on in a campaign’s life, this can be a great way to identify issues with the types of targeting the campaign is doing, whether that is audience segments, search terms or creative combinations. You should be able to nip these in the bud before they become a big issue.

Later on, the insights tab can show you any changes in the targeting, as the campaign reacts to shifts in the environment, or to new assets being added. Make sure to check in on this tab every few weeks, it will often have new information to show you.


Final Thoughts

Performance Max campaigns are, at least at the moment, not for everyone. They require a level of investment, both in media spend and in creative creation, that some users may not be able to supply. However, for the users who can support such an investment, they appear to be a viable alternative to traditional campaigns when managed well.

We’re still not fully comfortable with many of our normal optimisation tools being taken from us, but it is difficult to ignore that this seems to be the future of Google’s advertising efforts. Optimisation is shifting from directly improving performance to improving the data fed in to machine learning, which will end up with more conversions at the other end. Whether or not Performance Max is the final form of this sort of system is yet to be seen, but it is worth at least testing if they can work for you, because it very well might be.


Want to Experiment with Performance Max Campaigns?

Are you interested in trying out Performance Max campaigns for your charity? We’d love to help you! Send us a message on our contact page or email hello@upriseup.co.uk and we can have a chat about whether Performance Max campaigns are right for you, and how you can get the most out of them.

    Did you enjoy this blog post?

    Share this article:

    We use Mailchimp as our marketing platform. By clicking below to subscribe, you acknowledge that your information will be transferred to Mailchimp for processing. Learn more about Mailchimp's privacy practices here.

    Contact us