Posted by & filed under PPCChat.

Here is the complete screencap of this week’s PPCChat session. The theme of the discussion was “Performance Max & Demand Gen in Google Ads” and the session was hosted by Julie F Bacchini.

Q1: Are you currently using Performance Max in your accounts? If so, for how long and in what industries?

Yes ecomm and lead gen. @runnerkik

I have not done a ton with PMax. I am exclusively lead gen. @NeptuneMoon

Using them on ecom. Been about almost 3 years using them. @duanebrown

Yes. Been using it for several months now with great success. Lead gen for local services. @williamhboggs

Yes in both lead gen and e-commerce. @navahf

Yes. Ecomm for a couple years. Lead gen in the last year or so. @robert_brady

Ecom only. @JanineSEM

Same here @williamhboggs– Lead gen for local services and it seems to do well. @NicoleGreggo

Yup. Have been doing it a lot but recently moved to sales primarily so been a pause on PPC. I guess, I gotta refresh the latest information here. @1tagupta

I’ve been doing PMax since September 2020. Most successful in ecommerce, but had some success in lead gen. @ScottOmiller

We’re using it for eCom, but it hasn’t done as well/doesn’t seem to maeke sense for B2B SAAS. @revaminkoff

Doesn’t even feel like 3 years have already passed with Performance Max. Is it just me or does Performance Max lose the gossip spotlight recently? @1tagupta

It is definitely an established product at this point @1tagupta and the Google trial has taken up the gossip space! @NeptuneMoon

@1tagupta, it hasn’t been 3 years for most. It only became publicly available 2 years ago. @ScottOmiller

No. B2B clients only. @beyondthepaid

Google Trial? Gotta catch up on a lot of news then. @1tagupta

@ScottOmiller I remember it was first there in October or November 2021 in our accounts. I started it since then.@1tagupta

Been using for about a year in primarily ecomm accounts. @McKenzie Davis

@1tagupta I remember. My claim to fame in the PPC world is that I was incredibly fortunate to launch the world’s first PMax campaign EVER during its pre-pilot. (At least, that’s what the Google product engineer told us, that we were the first.) @ScottOmiller

Yes, in ecomm, first campaign launched Jan ’22. @DanielLuise

Predominantly this year, fully DTC within apparel, kitchen appliances, food, and outdoor gear. @teabeeshell

We haven’t implemented it across all accounts as budget is something that’s being considered but we have used it for both eCom and lead generation. @adwordsgirl

Q2: How has PMax performance been for your accounts lately?

I feel like in the platform performance looks amazing in reality so, so. @runnerkik

Google would say amazing when you forget PMax has brand and remarketing conversions basked in.At least for some brands, looking at removing brand starts to have you really understand what PMax is and is not doing for the ad account. It does work but at different levels. Don’t set it and forget it people. @duanebrown

When left unchecked, PMax looks like it’s AMAZING, but for straightforward prospecting/top-of-funnel marketing, it’s just slightly better than “okay”. @ScottOmiller

Crushing it! Our PMax campaigns see CPAs anywhere from 50-75% lower compared to search campaigns for the same local services. I’ll add that I have not yet excluded branded searches from these, per @duanebrown ‘s comment. I’ll be interested to see how this changes things going forward. @williamhboggs

Google Analytics and Google Ads show two very different numbers for PMax performance – makes it really hard to judge. @revaminkoff

Consider me a spectator here as I hadn’t run it in a while but I saw demand gen campaign emails from Google. @1tagupta

After excluding brand, it took us many months to rebound. We still haven’t gotten back to the client’s target ROAS but it’s gotten a lot closer and the client appreciates that it’s not brand search sales they feel they would’ve captured anyway. @JanineSEM

I tested it for my lead gen local accounts since about last year – we excluded brand and found CPA’s to be higher than search and lead volume low. The client base we had at my agency just left and now we’re in the middle of a big push to get other clients on board with it. My team members have always had solid numbers using P-Max and when I asked them if they excluded brand, they said no, so. @robert_brady

I mean, Google does not want us excluding brand… I’m always super curious when they trot out household brands to talk about PMax if they are including brand. Cause if they are, of course it looks amazing. @NeptuneMoon

It’s still working for us with brand excluded, but it’s so hard to know what is exactly working — I feel like I could optimize if I knew where the conversions were actually coming from. @revaminkoff

@NeptuneMoon a lot of big brands have budget to burn and are spending big bucks on branded search anyway, so they might as well include it. @ScottOmiller

It’s all about peeling back the layers:

  • What % of spend goes to Shopping inventory?
  • What % cannibalizes / overlaps with Branded Search efforts?
  • Is YouTube, Display, Discover, and Gmail work for or against the account?

In general, it’s a boon to business, but it deserves scrutiny and all control still available to advertisers. @teabeeshell

It’s been a little bit of a hit or miss for us. The numbers are definitely more impressive in Google ads than they are outside of Google. @adwordsgirl

Q3: Have you tried anything new in Performance Max or do you plan to in Q4 2023? If so, what and why?

I wouldn’t say it’s anything new, but I’ve often wondered if the reason others have been unable to get PMax to perform successfully may be because they aren’t utilizing all of the targeting features available in PMax. @williamhboggs

Which ones do you think are underutilized @williamhboggs? @NeptuneMoon

Nothing new and in e-commerce treading lightly with any major reorganization efforts. @runnerkik

Since we do a lot of ecom, not testing anything yet but we have some ideas set up for 2024. @duanebrown

@NeptuneMoon Potentially using only one of the 4 options in Audience Signals. I am using 3/4 (the fourth is demographics and doesn’t much pertain to our audience) @williamhboggs

The biggest execution decisions (settings) are: @teabeeshell

  • Customer acquisition – Bid only for new customers OR Bid higher for new than existing
  • Feed listing groups – Group similar priced products and/or priority products, not all inventory, all together
  • Bidding – tCPA (control for acquisition cost) vs. tROAS (prioritize AOV & select checkout behavior)
  • Opting out of defaults – URL Expansion + Automatically created assets
  • Brand exclusions – Understanding the “true” cost of acquisition, without the inflation of branded search/awareness

I am always weary about making big changes in Q4 because it is such a huge season for businesses, but also it’s the time of year where all platforms get somewhat finicky. Based on experience, something doesn’t go our way within these platforms (disabled accounts, random violations etc) for at least an account or two. So my focus in Q4 is to avoid more issues, if that makes any sense. @adwordsgirl

Recently started testing the “target new customers only” setting, but included not only our customers purchased list, but all site visitors as well to really try and see if Google can find “new” people. We also set the value of a new customer to .01 so there is nothing inflated with reported conv value. @DanielLuise

This is the way. @teabeeshell

Q4: Have you tried Demand Gen campaigns yet? If so, how have they performed? And if not, why not?

I have not I have total FOMO just following along here in this thread. @runnerkik

I haven’t either! @NeptuneMoon

Not yet. Can do this in Q1 2024. All hands on deck for Black Friday and Christmas. @NeptuneMoon

Not yet, but are pitching them to several clients. Was quite disappointed to see Google’s budget recommendations for them though (15x your target CPA) @beyondthepaid

No. I am that type of person who would rather let everybody else try it out first for a few weeks, take their learnings and then implement it myself on our end. @adwordsgirl

Not yet, but I have an opportunity for one specific client to launch for the next month, right up to BFCM. I’m not yet 100% sure on how we’ll gauge “success,” but we have the budget and flexibility to explore a bit. @teabeeshell

I’ve got one client that’s been running Discovery and we’re switching them over to Demand Gen soon. @robert_brady

I have a few clients that are using them; so far, I’m pretty impressed with the product. Pretty much all you could ask from a new product (and definitely better than PMAX at launch). @DigitalSamIAm

Same, haven’t run them yet but am really looking forward to trying them. We’re taking a break from all top of funnel advertising next month so we can test how that affects “organic” traffic. Will likely start them back in December or Jan along with Demand Gen. I’m excited for Brand Lift, Search Lift, and Conversion Lift measurement. @williamhboggs

I ran Discovery Ads on Multiple clients and now marginally testing Demand Gen as they have migrated, but I’ve not added a feed… as I’m too scared. Some worked and others didn’t also it really depends on what it is e.g. for one client I tested different product ranges and one did nicely and the other just drove Impressions, it’s interesting. @MeriemNacer

@DigitalSamIAm what type of creative have you found success with? I have a client that I wanted to test it for in the next week or two, till end of year but they don’t have a lot of video creative formatted for different demand gen formats. Wondering how static imagery will do? @timmhalloran

We’re mostly playing w/ our top-performing social creatives right now. It’s working pretty well. @DigitalSamIAm 

Q5: What is your Performance Max wish list? What would you like to see happen with its development?

Asset group level data. @JanineSEM

More granular metrics. @williamhboggs

Better built in video ad creation tools and templates – I’d roll PMax out to more of my clients, but most of them don’t have a budget for video production. @williamhboggs

Channel level data but I won’t hold my breath. @JanineSEM

Make it easier for us to exclude keywords e.g. without having to ask Google to add negative lists to the account. @MeriemNacer

I thought you all would have more wish list items for PMax…@NeptuneMoon

Are we just so beaten down we don’t bother wishing for anything anymore? lol @JanineSEM

It’s so lacking that we just don’t use it @beyondthepaid

We recently got data by Asset group level as long as you change it to the table format which was nice..@MeriemNacer

I use scripts to learn channel data and Search terms…@MeriemNacer

I mean, can I have what the scripts do but in the interface? @MeriemNacer

Metrics, metrics, metrics the ability to not have to see old asset groups. @runnerkik

More metrics, particularly on assets so that we can help the algo do better. @robert_brady

I feel like the crux of the tension with PMax is that we want insights and Google wants us to just make a few inputs and “trust the AI” @NeptuneMoon

Better Google Ads Support in general. @williamhboggs

“Advanced Mode” for advertisers that appreciate the ability to analyze data. Less “Below Average, Good, Great” columns (specifically with creative analysis but all of it). Being able to see all the metrics, like everyone else mentioned. I’ve been wanting to see more conversion breakdowns by location with my multi-location national brand but can see nearly nothing that’s helpful. @timmhalloran

My top list is: @teabeeshell

  • Ability to steer spend between inventory types, even if minimums are required across each
  • Specific query insights, not just grouped themes
  • Full confidence that assets are not created without explicit advertiser instruction/approval
  • Quicker algorithmic learning curve, lower cost to achieve stat sig

PPCChat Participants

Related Links

Stop wasted ad spend with Karooya

Stop the wasted ad spend. Get more conversions from the same ad budget.

Our customers save over $16 Million per year on Google and Amazon Ads.

Leave a Reply

Your email address will not be published.