Why Facebook advertisers should test Performance Max
With more levers to ensure brand safety and profitable ad spend, it's a great time for DTC brands to test Performance Max. Here's how.
Since Performance Max’s rollout in November 2021, adventurous PPC marketers (me included) have been giving it everything we’ve got to make these campaigns work.
The early days of a new Google Ads product often come with some hiccups. The lack of control and visibility into Performance Max felt incredibly disappointing, given its potential and innovation.
That’s why this year’s Search Ads Week felt a lot more reassuring, thanks to some new updates for Performance Max campaigns that include campaign-level negative keywords and asset group reporting.
With an exciting product now offering more levers to ensure brand safety and profitable ad spend, there’s never been a better time for DTC brands (and other Facebook-first advertisers) to test a Performance Max campaign meaningfully.
In this article, I’ll walk you through:
- What Google offers that Facebook can’t.
- The latest Performance Max updates and what they mean.
- How to set up a “social-style” Performance Max campaign.
- Additional learning resources.
What DTC brands are missing without Performance Max
After running Google Ads campaigns since 2003 and seeing every iteration and innovation the product offers, I can confidently say that advertisers ignoring Google Ads are missing out.
The Google AdWords of yesteryear was about the intent behind a keyword or search query.
It sought to answer what people were searching for at a given moment, why they were searching for it, and how to differentiate it from similar but different things.
When I think back to 2015, Google put out many marketing materials about “micro-moments” – that is, you’re serving an ad to somebody searching for that exact thing at that exact time.
But to scale past a certain point, you’ve got to think broader. Only so many people are searching for a given keyword, particularly at a transactional level.
Once you’ve achieved a high impression and market share of that business, how do you continue to make money?
This is where you add a layer of audience targeting.
Google works hard to connect keyword intent to audiences by looking at users’ previous search history, the types of websites they browse, the types of purchases they make, and so much more.
With Performance Max, there are two sides:
- Search intent: Action-oriented, clearly defined, based on user behavior.
- Audience signals: Awareness oriented, less defined, based on your data or Google’s.
This campaign type tries to get in front of people who are (or might soon be) at the convergence of those two paths.
Let’s say you sell pickleball rackets. Performance Max will target people who’ve told Google through their actions that they’re interested in pickleball.
Maybe one user watched a video on YouTube about it. Another searched for pickleball gear or lessons. And now, those people will start to see ads on YouTube, in the Discover app, in Gmail, and across other Google properties.
Indeed, Google has constantly been shifting toward machine learning and automation.
So how do you augment a keyword with who a user is and what they might be interested in, similar to what Facebook does?
Google has an advantage over Facebook in this area because it has your search history.
They know what you’re looking for across different sites and properties, which means piecing things together is not predicated on guesswork and estimation.
What’s new in Performance Max campaigns
On February 23, there were rumors of several new capabilities for Performance Max, followed quickly by an official Google announcement.
While Google has given top billing to combining broad match keywords with Smart Bidding (what they call the Ads Power Pairing), I’m more excited about these four updates, most of which Google Ads community liaison Ginny Marvin chose to highlight.
Campaign-level brand exclusions
It’s the feature everyone has been asking for (kind of).
While it’s currently limited to a drop-down list of brands, PPC marketers will still appreciate the ability to exclude branded traffic (their own and that of competitors) from Performance Max.
Though it’s important to note that branded traffic isn’t always worth excluding. (My team makes this decision on a case-by-case basis.)
Page feed support
Performance Max includes an option to enable final URL expansion, allowing Google’s machines to decide which pages on your website it should send users to.
Now with a page feed, you can input a group of URLs to prioritize. And with URL expansion turned off, you can still send users to one of several pages instead of a single one.
Asset group reporting
This is my favorite announcement and arguably the most important. Being able to see which creatives are driving the best performance allows you to iterate even better ones that speak to customer desires.
While brand exclusions are undoubtedly helpful, success in advertising is still largely down to the quality of your ads.
Experiments for incremental lift
The new experiments feature in Performance Max embodies the spirit of “Always be testing.”
I look forward to seeing how some of my Standard Shopping campaigns measure against Performance Max. You can operate using either conversions or conversion value as your guiding metric.
Performance Max uses Google’s AI to drive results in three ways:
- Multiply your customer connections by finding new customers across all of Google’s channels and inventory.
- Multiply conversions and conversion value by optimizing for the highest-ROI conversion opportunities.
- Multiply your expertise by sharing insights to help you answer strategic questions about your business.
For more details on these and the full list of Performance Max updates, read the official Google Ads blog post.
Get the daily newsletter search marketers rely on.
How to set up a 'social-style' Performance Max campaign
There are three ways to run a Performance Max campaign:
- The authentic Performance Max experience as it was intended, which is a combination of search, shopping, and social display inventory.
- The attempt to replicate Smart Shopping, which is only to provide a data feed and no creative assets.
- The social-style campaign, which only uses creative assets taken from high-performing ads on Facebook, Instagram, and TikTok.
I only hear people talk about the first two, but the third can be very powerful for brands that have proven themselves on Facebook and other social media platforms.
When you use those high-performing assets from your other ad networks, they'll appear on:
- The Discovery network, including the Google app, Android news feed, and Gmail.
- The Display network.
But for this to work, you must tell the system who to show it to. Unlike the shopping component based on keyword intent, Google needs to know your audience signals.
Remember, these are not a list of users you want to target directly. You simply want to share the characteristics associated with your specified audience.
Let's say you're selling an impulse purchase product. It doesn't make a difference whether it's $40 or $100. You're doing great on Facebook and getting a ROAS that you're happy with.
On Facebook, the pixel can do automated targeting based on who else buys from there. But when you start a brand new Performance Max campaign, it doesn't have that data. And so I shout from the mountaintops all day – use your data!
If you have 100,000 customers in your Shopify database, upload that list to Google as an audience signal. This will tell Performance Max who buys from you and, more importantly, allows it to find more people like them.
You can also create an audience signal around search terms, competitor names and websites, demographics, interests, and affinity groups. When you combine these with your winning ad creatives and remove the data feed, you have a "social-style" Performance Max campaign.
Embrace the opportunity
I wasn't always accepting of Google's move toward machine learning and automation.
However, I've realized over the last few years that resisting it negatively influences my mentality and the performance of my campaigns, team and business.
So I spend less time trying to hack or game the system and more on sourcing the data I need to drive performance.
But when you consider a campaign type as automated as Performance Max, it's critical that you seed it with high-quality data and then continue to monitor performance.
You should embrace it, but not blindly. Trust but verify.
In addition to providing audience signals and getting your campaign structure right, you have to make peace with the learning period.
Even with the most relevant data, every brand must spend a little money while the campaign runs in trial and error mode. With the right moves, it will be a very affordable entry price for additional profit.
So, have I convinced you yet? Ready to give it a shot? If so, check out these other articles from my series on Performance Max:
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.
New on Search Engine Land