Performance Max Exposed: Wasted Spend, Hidden Data

Source: YouTube auto-generated captions, lightly cleaned (repeated caption lines removed, misheard brand names corrected).

0:00 — Intro

Pmax Pax Pax Pax performance max subreddit it's a ban term of PPC on LinkedIn man it goes down in terms of complaints I if you could thumbs down on that baby you would so here are the most common challenges that folks are experiencing brand is taking all my conversions or taking all your conversions I don't know where my ads are serving. How do I measure to make a smart decision? And why in the world does they keep telling me I can spend more money?

And then I mean you can throw in shopping. You can talk about um feed only. You can do social style. Uh you can do all the things to it. But here's the reality check. We're all talking about Claude and AI and ChatGPT and all these M dashes and MCPS and all these things, but we still haven't just figured out the basics and how to deliver, when to deliver, why to deliver um our ads using Performance Max. So, here's a couple quick fixes for brands,

1:12 — What’s wrong with Performance Max

add them as exclusions, campaign level, account level, get dirty, get gritty with it. I suppose if you're using the full stack of uh performance max, my hope is is that you're doing retargeting with those same images using display. If you're not, that's a big miss. But um those placements and exclusions and audiences that you're using from display can also be used for performance max to help get the quality that you're looking for for exclusions. If not, go pull a placements report and add them babies to it.

testing, understanding, measurement. Thankfully, Google did something crazy. They gave us some insight uh into being able to see where our ads are delivering within that ecosystem, whether it's shopping or a search or Yeah. And so now we can kind of do some incrementality, lose incrementality. What are you talking about, John? And we can split test. We can do geo, we can do timing, geo, exclude um a specific zip code and measure the loss that you have over a three to four week period and say, "Okay, if I don't advertise in this particular zip code, this is the loss that I'm going to experience." vice versa only advertise in that geo or that zip code and don't advertise in others for a set period of time and say wow okay this is what I would achieve in this in this uh the other options are um

you can do um day parting and the the other and the last one which I I think is is is great uh right is you can do a branded PMAX campaign and a non-branded uh PMAX campaign to tell the difference between the two. So you can turn off your advertising at a specific time, you can um exclude a zip code, right? Or you can split between non-brand to measure the impact of what performance max is having on your business and then you can make a smart decision in determining if you want to invest more or not. uh in

3:20 — How to fix brand cannibalization

terms of doing shopping only with standard shopping you got to do what works best for your budget and works best for your business. Outside of that to answer those initial questions on is Performance Max just eating up all of my brands? Is it worth investing more? How do I uh measure the impact of this account? Well, we went through that. um exclusions for your brand, exclusions for your placements, um incrementality testing, turning things off, turning things on, and then lastly, that'll help inform the decision on whether you invest more in performance max. Anyways, it's happy Wednesday. I hope you can leverage some of this information. About to go inside. Just got out the gym. It was a quick thought. Peace and love.