Mike Rhodes P-Max script unlocks many data points only available through the Google Ads API. In addition, it presents the data in an easy to analyze format. In theory, this should allow advertisers to better optimize their P-Max campaigns. But how useful is all this information really?

In this article I’m going to dig into the various reports and discuss what’s informational vs. what’s actionable.

Disclaimer: Ten Thousand Foot View does use the paid MCC version of the script

How Complicated is it to Install and Maintain?

The MCC version is by far the most complex edition. Installing the script in Google Ads is pretty straightforward but you do have to run the script a few times to get the initial results loaded up. From there you need to copy over the accounts of interest and configure some basic settings for each one. If you have a hundred accounts running P-Max this task is not trivial.

While I haven’t done it, I expect installing the paid or free single account version should be both faster and much easier.

There are quite a few settings to configure at the account level, plus advanced features, plus the “whisperer” if you plan to use the AI analysis tool. It’ll take you some time to go through each setting, figure out what they do, and then remember what they do weeks or months later when you return back to the advanced settings screen.

It took me several hours to get everything configured the way I want… we do have quite a few clients on P-Max.

On an ongoing basis I need to add new accounts and configure them, and share the sheets with the PPC managers. Also, the script needs updating every month or two when new versions are released. If you’re running the MCC versions and unable to perform an “update” which only works for fewer than 20 accounts you are basically doing a fresh install all over again. This needs to be resolved as it’s frustrating to go through this long process over and over again.

What Reports are Available?

MCC Level Reports

The MCC version offers a suite of account level reports including weekly and 30-day trends as well as performance summaries. Sadly, at least at my agency these reports aren’t very useful. While there is loads of information, we do not run P-Max exclusively for any of our clients. Thus looking at cross-account reports that only contain a portion of spending and results doesn’t benefit us at all and in fact can provide a false sense of account performance. We have developed our own reporting dashboard that includes comprehensive performance KPIs. I suppose if you have P-Max only accounts this would be more useful… but who’s doing that?

Digging into individual account reports, there are many.

Account

My favorite report, this tab displays the breakdown of spending and returns by channel for each of your campaigns. This data is completely obfuscated in the Google Ads platform. There’s no easier way than this script to get this information. One interesting thing to note before I continue, is this report includes standard shopping campaigns as per this example.

Okay, great… so what? At a high level it can be helpful to understand where spend is going and what channels are driving conversions. You do have to be thoughtful though wherein video and display will sometimes seem to spend money without much return… but those channels may be raising brand awareness which is driving more sales across shopping ads or branded search.

One way we do use this data practically at my agency is to evaluate traffic sources when we have a lead quality issues. For example, when you compare lead quality between paid search and P-Max you find that you’re closing 25% of search leads compared to say only 10% of P-Max leads. Furthemore, when you look at your P-Max report you may notice that 90% of your leads and spend are from display. This is a strong indicator that you might be suffering from click fraud or simply lower quality placements.

Armed with this information you can run a P-Max placements report and start blocking those with high impressions that subjectively don’t look relevant. You could also block all the app categories in one shot to kill what’s known to generate a lot of bad clicks.

Another insight would be if you are barely getting any spending on video ads. It’s probably worth investing in some new creatives.

One thing to remember when you are running shopping ads in P-Max, you should expect to see a pretty small portion of spend going towards search ads. This aligns with the normal expectation that shopping ads will tend to dominte your ad spend as almost nobody scrolls past those only to click on a search ad.

Campaign

This tab provides quite a lot of detail for each P-Max campaign you run. You can select which campaign to view from the drop-down menu.

Besides overall channel breakdowns similar to the account report, this one also graphs historical performance. This can be useful for spotting changes such as a sudden increase in display spending that correlates to a drop in lead quality.

This report also introduces search category data.

A breakdown of impressions, clicks, conversions and value are provided by category type (brand, non-brand, etc.). And the search category details are provided by label and sortable by KPI. These reports are way more user friendly than thumbing through categories one by one in Google Ads. Importantly, it can be very helpful to know how much “branded” searches are skewing your conversion metrics to look great.

While it’s certainly possible to identify “bad” categories and add negatives based on Google’s “Insights” page it’s a little more tedious for sure. Likewise, this reporting format makes it easier to identify winners to add back to your search campaigns (mainly useful when running P-Max for lead generation).

Title

The title report is specifically for shopping ads and segments individual products by title (it’s possible to have multiple products with the same title). Performance “buckets” include profitable, costly, flukes, meh, zeroconv, and zombie.

To optimize your product ads it’s important to understand the implications for each category and how to deal with the products within each category.

As an example, products that are “costly” eat up a lot of budget while not driving much return in terms of ROAS. This is obviously bad since higher performing products won’t get as much ad spend to work with. In a scenario like this you might consider moving the costly items into a separate campaign with a restricted budget or just stopping them altogether. Importantly, these results are based on the threshold cost and ROAS figures from the advanced settings tab. I’d caution against using the default values here or your report may not be very meaningful.

Is this any easier to work with than what Google spits out? Probably, yes. In Google Ads you need to sort and filter in the “products” area to output the same buckets. You can, of course, save those filters so this isn’t a huge time saver really. But it’s a nice to have report for sure.

nGram

The idea behind nGrams is to identify root words that lead to either poor or good performance and then action those accordingly. There’s a bit of a learning curve to properly understand and action nGrams despite different versions of the tool being around for many years. The truth, also, is that this tool is far less important than it used to be as we move towards automatic query targeting along with broad match keywords. Whenever I’ve used nGrams in Google Ads I’ve found that 2-3 word nGrams are most useful. nGrams with fewer words tend to produce meaningless results.

Mike Rhodes’ spin on it displays one word nGrams for product titles. Presumably you can use this data to help refine/optimize your product titles to utilize higher performing root words. If you have a few thousand SKUs you’d better clear your calendar!

I’ll let Mike explain all the features of this tab.

S. Cat

This is extremely similar to the last tab but runs nGrams for search matrix (categories) instead of product titles. In this case, then, you would use the report to identify negatives to add to P-Max or positives to add to search campaigns or your audience signal keywords.

Comp

As of right now this report is undocumented. It appears to be another type of nGram analysis. But until Mike documents I’m hesitant to say what I believe it does.

ID

This report is very straightforward and simply displays all your product IDs and their respective “buckets.” You can copy items that belong to a specific bucket here and then action them in bulk using a variety of tools to optimize your P-Max campaigns.

Geo

This is a brand new report that dropped this week. The report breaks down spend by location at this point. Not very useful honestly since almost any reporting tool including Google Ads report offers this. There are a lot of KPIs here that are blank, however. So maybe more is coming soon.

Assets

My second favorite report, the assets tab displays all the missing KPIs per asset for your creatives. Yes, Google did add “conversions” a few months back and it has been very helpful. But just having impressions and conversions along with a subjective performance rating is lacklustre at best.

Mike unlocks all major KPIs with this report and it works for every single asset type. The report even pulls in thumbnails to make optimization of images and videos a snap (just check the “turn on thumbnails” box).

With this information in hand you no longer have to guess how an asset is performing. I wish we had this data for RSAs!

Asset Groups

As the name suggests, this report offers a summary of performance by asset group across your entire account. This report is useful for analyzing performance when you have a large number of P-Max campaigns and asset groups per campaign. Most importantly, it makes it very easy to spot outliers that need attention such as when an asset group stops spending or returns are extremely low.

What’s the AI Whisperer?

The AI Whisperer is an experimental feature that automatically analyzes and summarizes performance of your P-Max campaigns. This is provided in a results tab and can also be sent via email or enter read aloud to you. It’s early days for this feature and I don’t see much benefit for my agency until/unless it is much more stable and reliable.

If you want to learn more you can see Mike’s explainer video here.

Does it Save Much Time?

Some of the reports simplify extracting data and thus streamline analysis. One good example of this is the title report which would require custom filtering for each of the 6 “buckets” and flipping between each of those to pull up the data. Some reports facilitate new analysis that’s simply not possible without the script or similar other tool. This can actually add to your workload as this isn’t something you previously assessed. But doing that work may be quite valuable. For example, looking through the account/campaign reports can help you identify correctable wasted ad spend.

There is quite a bit of setup and maintenance for the script if you use the MCC version.

My verdict is the tool probably doesn’t save you any time and might add a bit of management time. That’s okay, of course, if it helps you improve campaign performance in a meaningful way.

Does it Improve Campaign Performance?

My short answer to this is, probably. In order to take advantage of the script several things need to be true:

You have to learn how to configure and use the script to properly analyze your P-Max campaigns.
You need a large enough data set, particularly in some reports, to find actionable trends.
The insights you get from the script have to be uniquely offered in the script and not something you can get natively from Google Ads – otherwise why bother.
You need to know how to bend P-Max to your will. It’s one thing to identify that you are, for example, wasting money on video ads. It’s quite another to address it in the black box solution that is P-Max.

Is the Script Useful?

In a word, yes. You might have guessed this up front when I stated that we use it at my agency.

We get a lot of mileage out of the assets report, optimizing copy and images is so much more efficient now. The account and campaign reports are really handy when lead quality or where you may be spending comes into question. The other reports are less useful to us as we have other ways we already deal with shopping data sets. But that doesn’t mean those particular reports won’t be useful to you.

For $400 MCC or $200 single-account the script is well worth the cost of admission. Thank Mike, keep up the great work!