I’m at SMX Advanced, and I just saw a session by Mike Ryan on PMax, and let me tell you… it was a good one! Mike and I talk quite a bit about PMax, but the thing I love about Mike’s dataset is that he has so many accounts at SMEC so can really dig into cross-account data and share his findings. I may or may not have been guilty of asking him specific questions in the past with the hopes that he would investigate my theory or concern with his dataset! 😁

‍

Regardless, one thing Mike pointed out in his session that was totally new to me, is a specific aspect of the new PMax Channel Report. 

‍

The New PMax Channel Report Segments Out Product Data in Display, YouTube, and Non-Search Channels

‍

Now, let me start by saying what all of you are thinking: it sucks to have a Channel Report and no Channel Controls (and it sucks that Demand Gen has Channel Controls and no Channel Report… honestly, that’s such a Googley thing to do, I have to admit I laughed when that happened… 😆).

That being said, keep reading to the end because I have some practical ideas for how you can actually utilize the PMax Channel Report even if you have no controls…

‍

Ok, so here’s the new aspect Mike told us that blew my mind:

You can now see which of your non-SERP ads (think YouTube, Display, etc) are product feed-based and which are from your (or Google’s automated) assets. 

‍

HOLY COW THAT IS COOL!

Mkay, I don’t think you are as excited as me so lemme explain. Cause I think this is actually one of those tricky things that was fooling you and you didn’t even know it.

‍

In the past, PMax channels reported their Product clicks to us through the API like below, this being a screenshot taken of the well known Mike Rhodes’ script (in one of our accounts).

‍

‍

Here’s where someone may realize they’ve not been looking at what they thought they were looking at: 

‍

“Shop Cost” is commonly thought to be ONLY Google Shopping clicks of your products in a Search Engine Result Page (SERP) when in reality it is ANY click on any product feed based ad… which can include ads on the Google Display Network, YouTube, and more. These are typically referred to as Dynamic Product Ads (DPA) and can be either Dynamic Remarketing of Dynamic Prospecting Ads (example shown here): 

‍

So to summarize, you may have thought “gosh, we’re mostly doing shopping ads in our PMax campaigns so this is basically the same as running Standard Shopping campaigns.”

‍

Nope, and to be clear, I’m not here making some sort of qualitative statement (at this time), just more of an objective statement of fact: PMax and Standard Shopping can never be apples to apples because of DPA… and most people didn’t realize this is what their report was including. 

‍

BTW, this is also what made testing product attributes with PMax virtually impossible. We had no idea of even knowing whether those attributes were helping us rank better with searches when we had no way of knowing what KPIs in the PMax campaigns were for “true” Google Shopping ads and which were showing on additional channels. 

‍

With that said, it was exciting to hear about the new PMax channel reporting, as that allows us to now more easily tell when DPAs are showing and when Shopping Ads are showing… but it still doesn’t give us a ton of insight into those individual channels… or so I thought. 

‍

So what’s the big change? 

‍

Well, Mike Ryan just showed us that now we can actually see what showed as product feed ads and what showed us “other” (static, dynamic, auto-generated, etc). 

‍

Practical Ways You Can Actually Use the PMax Channel Report

‍

Ok, I’ll break this down, but there are some REALLY cool things we can do with this data.

For instance, remember when we learned that Feed Only wasn’t really Feed Only? Well now you can tell when Google auto-generates assets from your feed, like in this feed only campaign channel report for one of our clients. 

‍

So in this Feed Only campaign, 1.7% of all traffic is product feed based ads… that’s pretty good, honestly, so we have more confidence now that the Feed Only strategy is working in this client and that we can continue to push Shopping aggressively. Though certainly, keep an eye on KPIs to ensure Google doesn’t jump spend into any of those alternative channels, you know, for “testing purposes”.

‍

Speaking of testing purposes, excitingly, we can actually more accurately run tests on Product Attribute optimizations now that we can break this data out more fully! 

For instance, let’s say you want to see if changing your titles causes increased exposure in Shopping Ads in SERPs. Previously, PMax wasn’t exactly a suitable campaign type for product attribute testing since we had no idea how much of that was DPA and what was static assets and what was actually Shopping Ads in SERPs. Well, I sound like a broken record but… we do know that now!

‍

‍

Now you can track, at least impressions and interactions on just product data, over time to see if your title test has increased exposure… of course, never forget that the BEST way to test your title changes, would be to track the exact search terms you are testing in the Product Title as well (which can also be done easier in PMax now with the new search terms report). I’ve written a walkthrough on that previously you should definitely check out if that interests you. 

‍

Here’s another example Channel Report on PMax: 

We also need to be a little careful to move more slowly and really pondeing report all together. For instance, we can see that YouTube “other” (non-product) Data segment was lower than the “product ads” in traffic, but significantly HIGHER in interactions. We need to be a little more careful about this, since interactions here for video actions is not the same as clicks, but looking at this, one could make the case from this report that whatever video content is being utilized is causing a decent reaction in its audience, and one should probably test video more purposefully based on this (likely in YouTube only campaigns). 

‍

You can draw your own conclusions with all of these, my point being that there really are ways you can utilize this data, especially in being able to distinguish between product ads and all other assets. 

‍

What do you think? What are ways you’re going to utilize this report? I look forward to seeing what the PPC community comes up with!

‍