Summary
- In 2023, use Performance Max with a focus on Shopping campaigns.
- Use Blended ROAS to track total revenue and marketing spend as a key performance indicator.
- Adopt a “wartime” mindset in advertising to focus on efficiency and remove waste in poor-performing accounts.
- Prioritize the right products, revise campaign setups, fix website leaks, and react quickly to poor performance.
- Prepare for possible changes in how Performance Max is utilized on display networks.
Intro
In 2022 most of us struggled with figuring out how to use Performance Max in accounts. The best practices from Google were a bit one-sided, and then the rest of us had to figure out what worked best for the different scenarios.
With all that behind us, there are some core things we should try to revisit in Google Ads in 2023.
This will not be a list of wild “keywords are going away!” predictions.
I will instead focus on explaining what we in SavvyRevenue will start doing (change!) and continue doing (double down!).
But first — let’s get our plans for Performance Max out of the way from the get-go, as we have spent too much time discussing this in 2022.
1) How to Work with Performance Max
Double down!
You have two choices with Performance Max: Go all in, or take a step back to focus on Google Shopping. This depends on your scenario, competencies, and marketing mix.
In Savvy, we will continue to ensure that Performance Max campaigns (around 30% of our Shopping spend is in Performance Max) are Shopping only campaigns.
I believe that we haven’t seen Performance Max campaigns in their full “glory” yet. I think they are holding back on YouTube and other Display moves. Based on the data we are able to pull from Performance Max campaigns, we see that it generally focuses on Shopping with a few exceptions but very little direct Display / YouTube activity.
I have two thoughts on this:
- Google is holding back on these networks to prove Performance Max is a high-performing campaign type
- Google’s algorithms can’t spend on Display while hitting ROAS targets
Both are scary because why are we pushing so much video / image creative if the Display part of Performance Max isn’t really spending any money?
If it’s the case of 2), then I predict we will all have a unified push from all our Google Reps to push for maximize conversion value in the campaigns.
2) Going Beyond Revenue Tracking: Blended ROAS & Micro Conversions
Double Down!
Overall, the eCommerce brands we worked with that saw the most success somehow used Blended ROAS to maximize their marketing potential.
The first thing is to understand that Google Analytics isn’t an accurate way to see what channels drive the most revenue. Most eCommerce brands simply can’t think of their marketing in such simple terms.
The second thing is to set up Blended ROAS mechanisms where you compare total revenue to total marketing spend as your North Star KPI. From there, you can start setting optimal ROAS targets for your individual marketing channels.
3) Defense Mode Due to The Economy
Change!
2022 was the first year since 2018 that we experienced client bankruptcy, and we experienced two:
- A client that had been with us for 5 years terminated our cooperation, and went bankrupt two months later
- A client that had been with us 5 years ago, came back in February, and went bankrupt in November (shitty timing!)
I’ve spoken with both founders, and they said the same thing: You can’t spend your way out of a decreased demand…
So if we can’t just increase spend and revenue, what should we do instead?
We like to work with two mindsets in Savvy: Peacetime and Wartime, which is unapologetically stolen from the excellent book The Hard Thing About Hard Things and covers how we like to think about accounts in the current environment.
When it’s peacetime, you can afford to spend freely, not focus on efficiency, and perhaps have a lower ROAS than you ideally want because your overall business is doing well.
When it’s wartime, you have to be ruthless:
- Remove all waste in your account — aka. anything without a demonstrable healthy ROAS
- Maximize your potential — are all products advertised for? Are the correct products prioritized, etc.
- Revise your setups: With less data, perhaps you need to combine into fewer campaigns
- Make sure you have no CRO leaks on the website
- React to poor performance! Don’t just let it run and think it will be fine due to “attribution”
- Don’t spend your way out of it to try to get revenue back. Failing with this tactic can put your company under duress.
- Restructure your cash conversion cycle by renegotiating all and any agreements you can. Don’t lose due to cash flow issues.
To conclude, I must say one thing: I hate LinkedIn sometimes. It’s full of marketing gurus in their 20s telling you “the best time to do marketing is in a downturn” because they:
- Read it somewhere
- Truly believe it
- Think it sounds cool
I’m not sure which of the three reasons is the worst, but no — don’t spend your way out of a recession. Fix all faults that make your accounts less effective, and live to fight another day.
4) Understand Your Ecosystem
Change!
The biggest denominator for struggling eCommerce businesses in 2022, or outright bankruptcies, were inflated overhead and a lack of SKU expansion.
Meaning the eCommerce businesses who did not perform well in 2022 all experienced a degree of:
- Having added too much headcount or other fixed costs
- Didn’t expand the products/categories they sold
So when the demand for their products decreased, they were left with a business where the demand for their products simply wasn’t high enough to support their headcount, fancy offices, software inflation, or poor marketing efficiency.
5) Start Managing Your Account Again
Double down!
Now, this is a funny one. After eCommerce golden days were towards the end of 2021, we saw an increase in incoming accounts where it was evident that it had just gotten permission to run.
Here are a few examples:
- An account ran with double tracking (Google Ads tracking and Google Analytics), which meant the actual ROAS was half of what was “reported”
- An account had had great success just running a Smart Shopping campaign, but once everything had normalized, the performance dropped by 70%
- An account with 600k SKUs outright excluded a product if it wasn’t profitable over a 7-day period: decent idea, but horrible execution.
Another case is similar but different. An agency that logged 50 hours a month for 5 hours worth of actual work. The rest were meetings, reports, emails, and what I believe to be forged time-tracking sheets.
The advertiser was very worried about changing things up to Black Friday and was hoping to just hit the same revenue as in 2021. We tried convincing them to allow us to make account structure changes, but they wouldn’t budge.
At the end of the day, we did make some structural changes but also managed the setup better.
We went from trending 20% below last year’s revenue to having our best Q4 ever.
Here is the gist and recommendation:
Going into a downturn in the economy, it’s essential to understand that accounts can’t be “left alone” anymore — they might reach a death spiral that results in everything going south — or the things that used to work simply don’t anymore. Hence, you need to change things up.
Action Items
- Create a dashboard with targets vs. actual performance
- Get into the account on a weekly basis
- Deep dive into the campaigns to check if everything runs as it should
- Develop a list of routine optimizations you need to do
- Decide on 3 projects (bigger action items) per quarter to implement in the account
If that seems too much, hire an agency to do these things.
6) Rebuild your Search Campaigns
Change!
Especially if you have neglected Search in the last couple of years, then it is time to get it up and running again.
A healthy relationship between Search and Shopping is 25/75 or 50/50 in terms of ad spend, depending on the industry you are operating in.
Smart Bidding is constantly getting more intelligent, and based on what I’m seeing, there are plenty of opportunities on the Search side.
This might mean we need to start thinking about our account setups more in a Paid Social fashion. One of the most recommended setups for Paid Social is to set the algorithm to target as broadly as possible, then focus on developing creatives (video, image, copy), build landing pages with funnels, and improve the website/product in general.
Sound familiar?
What if we took a similar approach to Google Ads?
One of my favorite activities in 2022 has been to simplify account setups. So often, they get bloated and impossible to manage because someone has an idea to do something that ultimately has no effect.
2022 Was the Year of Performance Max
This most likely means that most of you have neglected your Search campaigns due to two things:
- Not wanting to rock the boat while Shopping undertook a significant change
- Not enough time to invest in rethinking Search
Both are correct, but with Performance Max migrations behind us, it’s time to start thinking about Search again.
Recommended Actions
Review your account structure for the following:
- Is Dynamic Search Ads spending >20% of your Search spend?
- If yes, then you need to create keywords for the highest-volume search terms
- Are all categories, brands, and products covered?
- If not, then you know what to do
- Is it overly complicated?
- Do you still split campaigns in device, region, match type, etc.? Combine them
- Do you run SKAGs? Consider grouping more keywords
Simplify your account structure:
- Group similar-sounding keywords together. Here is an example I used for a client:
Taking one ad group, I added similarly themed keywords together:
This produced two things:
- I could focus more on writing ad copy
- I didn’t spend much time on this
The old Andrew would argue that I could have written the same ads even if I had split up the keywords in multiple ad groups:
- Back Pain
- Lower Back Pain
- Back
- Orthopedic xxx for back pain
- Spinal pain
- Orthopedic xxx for spinal pain
And then I could have gotten an EVEN BETTER keyword to ad relevance.
But the problem with this is twofold:
- There isn’t space to write about spinal pain or orthopedics in the headlines, so it’s a moot point.
- Performance isn’t better just by adding the keyword
Throughout 2022 I’ve seen this again and again.
Of the two following ads, there was always one winner:
Just adding the keyword to a headline never produced the best results. Having contextually correct ads are important, but your messaging is much more important today.
Again, this depends on your business. Running thousands of SKUs, then it might not be as relevant.
But running ads for a DTC brand requires better copywriting than just having the keyword in the ads.
7) Time to try Broad Match (again)
Change!
As mentioned above, I feel that Search campaigns have been massively under-prioritized by many advertisers who just rely on Google Shopping. Search can and should be about 25-50% of the total spend in your account, ideally closer to 50% than 20%.
With a combination of broad match, simpler campaign structures, Smart Bidding, and Responsive Search Ads, it should be possible for even the biggest eCommerce companies to build a strong Search program.
I have updated my thinking on this and I am planning to challenge it in 2023.
Here is what I have changed my thoughts on.
Dynamic Search Ads are superior to Broad Match. I used to think this, and I still feel it’s true in some cases. For long-tail searches in high-SKU eCommerce websites, Dynamic Search Ads are still better to match search term to ad to landing page.
Broad match can’t do that as well as Dynamic Search Ads.
But for keyword variations, I have noticed that broad match keywords do it better than Dynamic Search Ads.
This is partly because the ad copy is better in Responsive Search Ads that you create yourself compared to what Dynamic Search Ads can pull in from the Page Title some SEO copywriter wrote.
Recommended Actions
- Try adding Broad Match keywords across your entire account
- Not as an experiment. Try to add them to your existing campaigns and see what happens
- It has to be in combination with Smart Bidding
- Let it run. The longer it runs, the smarter the Broad Match / Smart Bidding combination becomes
8) Everyone is Running Smart Bidding. Now What?
Change!
I’ll keep this short as I’m not entirely sure what we are planning to do here.
But with everyone running algorithmic bidding, some opportunities might open up.
Like asking the questions:
- What does that mean for the auction ecosystem?
- How can you find pockets that Smart Bidding ignores?
- What are 90% of advertisers not doing today?
Now, this doesn’t mean you shouldn’t run Smart Bidding.
You definitely should, but it’s always interesting to think about zigging when everyone is zagging.
9) Responsive Search Ads: Please STOP!
Responsive Search Ads seem to have gone straight to everyone’s worst tendencies.
Way too many Responsive Search Ads are built inadequately.
Adding 15 headlines with just slight variations only results in the following:
- Google taking forever to learn what headlines work best
- Google showing underperforming assets too much, which lowers your conversion rate, ROAS and lowers your overall volume
But the worst part about this is that the pressure to create 15 headlines makes every single Google Ads manager write inferior assets.
This is not a generalization. This is literally everybody, and it happens in Savvy too, whenever someone thinks they want to try writing 15 headlines.
Just stop it right now. It does not work.
The method is writing 1-2 headlines that are “keyword” focused and 3-4 headlines that are “feature/benefit”-focused.
Keyword-focused examples:
- Nike On Sale Now
- Nike Exclusives
Feature/Benefit/Social Proof focused examples:
- Memory foam to avoid injuries
- Slip-on technology
- Breathable mesh
- Allows Your Feet to Breath
- Trusted by 1 Million Runners
Only by allowing fewer examples in each Responsive Search Ad can you write better assets.
What We Will Not Do
- Leave everything 100% to Performance Max
- Blindly investing in new creatives (videos & images) for Performance Max
Conclusion
Besides all the things I’ve listed today, I want to reiterate something I spoke about in Amsterdam, Berlin, Copenhagen, and San Diego at conferences in 2022:
As Google Ads marketers, you must move beyond button-pushing. Just knowing how the platform works don’t cut it any longer for anything but small business advertising.
You must be able to combine strategy, business insights, AND your platform know-how into one competency to keep doing well in Google Ads.
If you don’t start learning how to work on all three cylinders, then you will be left behind and only able to work with small businesses.
6 Comments
Here you go: https://savvyrevenue.com/blog/shopping-ads-performance-max/
🙂
Another great post Andrew. Every time I read your stuff it feels like we are having a fun Google Ads talk that will never end. For that reason, I always have an espresso while reading it.
One quick question about RSAs: are you in favor of pinning the keyword-oriented headlines in the 1st position or do you just let it open?
“The method is writing 1-2 headlines that are “keyword” focused and 3-4 headlines that are “feature/benefit”-focused.“
Really? No call to action in the headlines? Google has long advised this, and I’ve taken it at face value (without ever really testing) …
Google says many things, but many of there recommendations are general best practices (and some are outdated or translated poorly to different languages) 🙂
Call To Actions can be powerful in certain industries but for eCommerce people generally know they are clicking on something to buy it. So adding “Buy” as a call to action is redundant for eCommerce, and in my general experience haven’t outperformed other assets.
It’s different if it’s lead gen and it’s sign up for demo, get free evaluation, etc. Then it works really well!
The article is very informative and detailed. Got many points that i use in my campaigns. Thanks a lot andrew!
This is mind blowing, thanks for sharing this valuable content. this help me a lot.