Looking to create your first Performance Max campaign? Learn how
Google suggests running Performance Max campaigns for a minimum of six weeks. This is to give Google’s machine learning algorithm sufficient time to learn and collect enough data to compare performance.
Performance Max campaigns use asset groups (similar to ad groups in standard Shopping campaigns). These assets include things like text and images to better represent your inventory. Once your campaigns have been running and you’ve acquired enough data, it’s useful to split your asset group into multiple asset groups, each containing different assets, so that each group represents a different part of your product inventory. For example, you could have an asset group with the theme of ‘hats’, and a separate one for ‘shoes’.
Adding different versions (at least five) of both text and image assets will provide Google with more options for possible combinations for ads. This allows the algorithm to try to find the most effective combination out of your provided assets. On that note, providing as many assets as possible is suggested to improve performance - here are Google’s minimum asset requirements to get your content created and shown across the Google network.
Supplying your campaign with audience signals can help to accelerate the machine learning process and help to guide the campaign toward your target audience.
Google’s Smart Bidding algorithm will optimise the bidding for your campaign, but adding relevant data to audience signals will aid in overall performance optimisation.
You can look at your calendar view to see if there are specific times or days when your Performance Max campaign is getting a low return for ad spend.
Once you’ve highlighted a pattern, you can edit the ad schedule to only serve ads at profitable times. This is a good way to help manage your ad budget.
On your Performance Max campaign page you can see how much your campaign is spending per day and make budget adjustments.
For example, if your campaign is spending your full budget every day, you could test out increasing your budget to help scale performance.
Performance Max can use ad extensions to help you expand your ad with information, such as additional links and your company’s address, making your ads more appealing.
This expands the size of your ad without it being intrusive to users. With your extensions, you should address common queries that users may be looking for. You can even include specific information about your products or any special offers that you’re currently running.
We'd recommend checking your data between one and three months. Your spend can have an effect on this: the more you spend, the quicker the algorithm will move from the learning stage. You then gain access to more accurate results at a quicker rate.
If you're looking for a general view of how your campaign's performing, have a look at cost, conversions, and ROAS. When using different asset groups you can see how each individual group is performing and which need improving.
Keeping an eye on your ad spend is useful when it comes to optimisation. We also recommend identifying your most popular products and consider making individual asset groups or campaigns for different brands and/or product types, giving you more control.
Another aspect you should monitor is your asset performance. Under the 'Performance' column, you'll find a ranking of your assets. These are rated from 'Low' to 'Best' and are based on performance against other assets of the same type.
Ideally, all of your assets would be ranked as 'Best'. To do this, you can:
When it comes to video assets, Performance Max can auto-generate videos for you based on your assets. However, we recommend using your own video if possible. You can use an existing video you have on YouTube, grab the link, and use that as an asset in your campaign.
This is highly dependent on your spend. In theory, changes to campaigns go through immediately so you may see results right away. If you've made a more drastic change, the algorithm will need to go through another learning phase which could take a few days.
When it comes to ROAS and conversions, delayed attribution will have an effect so keep that in mind when reviewing your campaigns.
To avoid any potential drops in performance, Google recommends setting a similar budget for your Performance Max campaign as any existing Smart Shopping campaigns. This is to ensure that your Performance Max campaigns can capture previous shopping inventory.
If you’re still running Smart Shopping campaigns, then it’s recommended to stop them completely and allow Performance Max to take over and allow the machine learning algorithm to begin collecting data. It may take two to three days for your new Performance Max campaigns to fully take over your Smart Shopping campaign inventory.
As with any advertising campaign, looking at reporting and data is crucial. While Performance Max doesn’t give retailers access to all data, it does show you some actionable insights on how the automation is working. You can view consumer interest insights, audience insights, and search trends. This will be leveraged automatically by Google's AI learning process to better optimise campaign performance.
Be aware that Performance Max campaigns will be prioritised by Google over Smart Shopping campaigns that run the same products in the same account. This will cause spending in your Smart Shopping campaigns to significantly decrease, and Performance Max will increase in its place.
