Pmax – The Next Best Thing? 

By: Yael Heffetz. Search Director, CvE


Harnessed with vast experience of working with Google (it doesn’t always feel like I’m working with Google, at times it feels like ‘finding a way to work around Google’), I know that every new development is going through a beta stage. 

It is also known that there are a few special ones whom Google deems, that play a crucial role in those beta stages and provides their feedback to Google with regards to the new development. 

This is meant to ensure that once the new feature is released, it is tight and free from bugs or complaints. When Performance Max was launched, I wondered whether those who participated in these beta stages were simply too biased.  

In the last couple of years, it seems like Google is doing all that they can, to take the control out of the advertisers’ hands and into the mercy of their AI algorithms. It feels like Google are trying to automate everything and anything they possibly can, leaving fewer and fewer opportunities to optimise the Google Ads activity. 

A Black Box

Performance Max campaign is one of the best examples of this trend. One of the blackest boxes Google has ever released. It is so black, that even Google understood that they took it a bit too far this time and they have already announced that they will enable a few features that we used to be able to control. The first in the list is account level negative keywords. Although to this day, this feature might have been released in the US, but it is still not available in the UK. 

New features Coming Soon

They also announced that many other new features will be coming soon. These include campaign-level brand exclusions, page feeds, a video creation tool, asset group reporting, budget pacing insights and recommendations. 

Whilst it looks like an improvement in the ability to optimise Pmax campaigns, the fact is, that the only significant change would be the negative keywords and the ability to receive asset group reporting – but even this is depending on the reported metrics. As if those reports will present metrics that can’t be modified, apart from being aware of them, there won’t be much that can be done with this information. As I have yet to see this report, I cannot share my thoughts about it. 

The last feature (which was yet to be released) is the ability to run experiments, measuring the performance of Pmax against other traditional shopping campaigns or even against display, video or discovery campaigns. This is very interesting, as from my experience, once Pmax is launched, standard shopping campaign don’t stand a chance. Hence, I wonder whether it is Google’s way to show the undeniable superiority of Pmax over almost any other campaign type – especially shopping.  

With that said, Pmax came to replace the Smart Shopping Campaign 

I heard of cases in which it was used without connecting it to GMC (Google enables it and considers it as an option), but this would result in an odd campaign which is more restricted than Discovery but has a wider reach, as Pmax covers more of Google properties. 

When examining it as a replacement for the Smart Shopping Campaign, Pmax is doing a fantastic job, performance-wise, with e-commerce accounts that generate substantial volumes of data.  

Need of a High Threshold of Data to Excel

I think that unlike other products and campaign types Google Ads is offering, Pmax requires a very high threshold of data to excel. Much more than any other AI-based Google product. 

As such, an attempt to use Pmax with smaller accounts is most probably doomed. Especially if it is not connected to GMC. I really wouldn’t attempt to do it, why should one really? Without GMC, Pmax is simply a super limited Display campaign.  
It is limited from a reporting perspective and limited from an optimisation perspective. If there is a need to use a display campaign, use TTD. Or even GDN, but by all means, don’t use Pmax. It wasn’t meant for that. 

I was lucky to onboard a client who has already implemented a Pmax campaign as I’m not sure I would have been brave enough to offer this campaign type to one of our clients.  The inherited Pmax campaign was set up with the most basic setting. It was a ‘catch-all campaign’. No segmentation of anything. And even with these super basic settings, once comparing the one single Pmax campaign performance to the former traditional (10!!) shopping campaigns that the account had, this is what we found: 

  • Pmax Increased the total number of conversions by 135% 
  • Decrease CPA by 57% 
  • Increase ROAS by 124%
  • Increase revenue by 126%

If I learnt anything in life, is that one should never argue with numbers and the above numbers are a dream coming true, for any advertiser. 


To conclude; First, Pmax has very limited to no optimisation options, which makes it one of the blackest Google Ads boxes that have even been created. Second, it doesn’t allow any type of helpful reporting as even with the reports that are currently available, no practical use can be applied post-report analysis. Third and despite all the above, providing there is at hand an account that generates a significant amount of data and transactions, and providing this account can be linked to GMC, Pmax can definitely generate significant incremental growth and as such, for those instances only, it is highly recommended. 

News & Blog