Optimising for Google Performance Max: Manipulate your input to maximise your output

We’re now a year down the line since the launch of Google Performance Max and it has certainly proven it's here to stay, near doubling its revenue contribution (respective to total Google Ads revenue) from 22% to 40% YoY (internal client data). Whilst it does still have its nuances, the roll-out of account-level negative keywords and future updates such as budget pacing insights and campaign-level brand exclusions on Search & Shopping inventory are set to have notable impacts. 

However, when it comes to optimising Performance Max campaigns there’s still a lot we don’t have control over, including;

• Bidding on certain keywords or even see what those search terms are that we’re serving for
• Controlling how much our budget goes towards each placement
• View the performance of individual asset groups to see which are performing best, being prioritised etc. 

Therefore, brands need to rely on the information they are putting into the campaign, in order to maximise what comes out of it. 

Knowing which levers to pull can be difficult, so we asked our Head of Paid Advertising, Nicki Eggie, to share her top three tips for successful Performance Max campaigns, as tested on our clients.

The keywords you appear for rely solely on two methods of input - your product feed and your website content.

What search queries you appear for in the shopping carousel is entirely dependent on your feed. It is therefore imperative that we fuel the feed with high quality content to maximise impression opportunity for the most relevant queries. 

The best way to optimise your feed is to invest in a feed optimisation partner. At Reload Digital, we are great fans of product feed management platforms, which allow us to set up rules to optimise the feed and fill in missing attributes, make full use of character counts and even test different images, as what resonates in Google Shopping may not be what resonates on your product page.

Given 80% of search terms are matched to your product title alone, our biggest recommendation would be to start here. Google reads left to right, placing the strongest emphasis on the first few keywords in your title. So think about what keywords you want to prioritise to align to your brand goals. 

We’ve pulled together a feed health checklist (which can be downloaded here) for the key attributes that are going to have the biggest impact on your Performance Max campaign serving capabilities.

Feed Health Checklist.

Just as we need a good feed to support the shopping element of Performance Max, we need good SEO and on-page content to support the DSA element. 

My first recommendation would be to jump to Google Search Console and review the queries you’re ranking for. 

• What are we serving for organically? 
• What’s the ratio of unbranded to branded clicks and impressions? 
• What’s unbranded CTR like?

Generally speaking as a goal we want to be making sure that unbranded queries are contributing around 30-40% of all impressions. Any less than this means that your site doesn’t have enough content to support generic queries and the flow on effect of this will be that your Performance Max campaign will struggle to attract new customers.

However, there is a second element to this – clicks.

If you’re getting a lot of unbranded impressions but very few clicks, then it’s clear you aren’t appearing for the right terms. As a result, you’re feeding this directly into the DSA element of your Performance Max campaign and paying to be serving for the wrong terms. 

We therefore want to make sure that around 20% of our clicks are unbranded. If it’s not, then it’s time to reach out to your fellow SEOers to help support you here!

Performance Max is entirely responsive – meaning it will work with the campaign assets you upload (and the ones you don’t) to inject into certain placements to maximise your inventory with minimum input.

Given such a huge element of this campaign is creatively led, we have to make sure we’re giving Google the highest quality content to work with. For example, we noticed that 9/10 of lifestyle brand Emma Bridgewater’s top performing products were Coronation specific, so we uploaded Coronation assets which saw the campaign CVR increase +29% YoY.

Emma Bridgewater's portrait & landscape creative assets for Performance Max.

Considerations for producing A+ creative:

• Make full use of all 20 images
• Refresh ads as frequently as Paid Social
• Upload YouTube content to override auto-created videos
• Create HTML5 creative to control the look and feel

If you don’t upload YouTube creative to your Performance Max campaign then Google will automatically create content for you which may not resonate with your brand image, particularly for premium brands. A way to see what Google has created is to go to your Asset Groups and then go to ‘View Details’ and scroll through the list until you see the video content. 

If you’re also finding that the responsive display element of Performance Max is not aligning with your brand image, we’d recommend investing in HTML5 assets which allow you to have complete control over the end look. 

Your feed likely contains a plethora of products, all with varying levels of interest, conversion propensity and bottom line profit. Rather than funnelling your entire inventory into one campaign, we’d recommend segmenting your feed (via custom labels) and then creating separate Performance Max campaigns that hinge off these custom labels. 

There are 3 different ways you can segment your inventory:

Segmenting your feed by product category, sale versus non-sale etc. This strategy is great if you have certain hero categories that you’d like to push (or restrict) and is particularly useful for sales. However, it is a static approach that doesn’t consider individual product performance. That’s when dynamic performance labels come into play. 

Dynamic performance labels allow us to look back on the last 30 days of performance and segment the feed into different buckets (High Revenue, Low ROAS, No Conversions, etc.), meaning we can segment our campaigns and put the money behind what’s going to move the needle the most. This can be set up through your shopping optimisation partner, who will use an API to connect the feed platform to your Google Ads account, pulling in that historical data at the SKU level. 

One limitation of this strategy however is that it doesn’t consider COGS. Your highest revenue product might actually have the lowest margin and therefore contribute half the profit of a lower volume, higher profit product. This is where you can look at a Profit Margin strategy.

Profit margin labels go a step further as to segment the feed based on profit, allowing us to have more bottom line impact.

Here we can actually import the COGS into our feed and calculate the profit margins on our products at the SKU level and bucket these into High, Medium or Low profit. We can then use these buckets to form the basis of our campaign segmentation, setting different ROAS allowables for each. 

For instance we might set up the Low profit campaign to have a minimum ROAS of 600%, meanwhile the High Profit campaign can afford to have a lower ROAS of 300%. This will help Google to prioritise the products that are going to move your bottom line product, rather than just going after surface level metrics such as revenue and ROAS. We essentially move from a ROAS to POAS model (profit over ad spend).

As we lose more and more visibility and control over our campaigns and the ability to reactively optimise, we need to be ensuring that we are feeding those campaigns with the best content from the onset. When it comes to optimising PPC campaigns, now more than ever, brands need to think about the levers they can pull from a cross-channel perspective (SEO, Creative & Content) in order to maximise their success.

Have you got a question about Performance Max and how to integrate it as an effective part of your Paid Search strategy? Get in touch to speak to Nicki or a member of the Paid Ads team.