Over the past few months, I’ve consulted with a few eCommerce brands who have been struggling to scale PMAX up to levels they used to achieve with Smart Shopping. Each have had their own theories (and conspiracies) about the reasons why this is the case but none had taken the time to dig into what was going on with their campaign and underlying product catalogs.

My first move in these cases is to run this PMAX script built by Mike Rhodes (not sponsored) which surfaces a number of insights about an account’s PMAX performance. Included in this report is a breakdown of how all of the individual SKUs (Item IDs) that are being targeted are performing in terms of profitability.

(I also made a video walking through this all here if you'd prefer to watch/listen)

Below is a snapshot of one of the report outputs where you can see two things:

1. The percentage of products that are generating the most revenue for an account. In the example below, 3% of products are generating 50% of sales (!!)
2. A segmentation of products based on their performance and profitability (using your avg. COGS)

These buckets are incredibly insightful and useful because it highlights how the blending of products within a campaign can seriously limit performance and ability to scale.

Close to 60% of spend in this account has been going towards unprofitable transactions, leaving ample opportunity to re-adjust and immensely improvise efficiency.

Here’s a different view of similar data which better illustrates the problem at hand:

Let’s say this campaign is set with a 350% ROAS target and is consistently hitting that objective.

But as you can see, the way the investment and returns are spread across the catalog reveal what’s really going on…

The ‘Winning’ products in this catalog are achieving an 1100% ROAS but are unable to spend more because they are being pulled down by the ‘Losing’ products which are barely hitting over a 100% return.

Blend all of these together and the campaign is doing its job, driving a 350% ROAS but there is a TON of wasted spend pulling down the ‘Winning’ products.

So, my approach here is to split this catalog into four campaigns based on these buckets. With the key being to give each bucket its own ROAS target of 350% (for this example).

With that in place:

• The ‘Winners’ can likely spend 3x as much as they can in this current setup
• Spend going towards the ‘Losers’ will throttle AND will become profitable
• The ‘Maybes’ will be given a shot at become ‘Winners’
• Zombie products (those without any spend) will also be given a crack at things
  ​

And if you’re really eager, you can get more aggressive with the ‘Maybes’ and ‘Zombies’ in hopes of finding more ‘Winners’ from that lot.

To implement an approach like this, you simply need to attach a custom_label to Items based on their recent (30, 60, 90 days depending on your business) performance and segment the products included in each campaign accordingly.

Hit me back with any questions, suggestions, or if you try it out! If you'd like me to run this for your brand feel free to pick up an Express Audit from me here.
​