It’s not always necessary to complicate the strategy that will ultimately scale a client’s eCommerce account.
I’m going to show how to grow subscriber rates and sales while maintaining a significant Return on Advertising Spend (ROAS) using YouTube and Performance Max Ads.
Keep reading to learn how I improved two important aspects of this account including:
Let’s start by looking at the ad account here. The results show a 10x ROAS from November 1st to January 27th of this year. For those unfamiliar, ROAS is a marketing metric that measures the power or effectiveness of a digital advertising campaign. There’s a 128k conversion value with nearly 5K in conversions at 12K in spend.
The first problem with this account lies in the structure of the Performance Max campaign before I took it over.
Under Performance Max Campaigns, I click on Asset Groups and it’s clear the structure needs some attention.
There’s only one asset group with multiple signals lumped together. These were affinities, inMarkets, and remarketing for all their products. When the campaign has too many signals irrelevant to the images and products displayed, it’s a mismatch of images to audiences. Especially considering where the audience exists in relation to the funnel.
I fixed this low-performing setup by segmenting the creative, signal, and products being targeted to perform best.
This signal (above pic) is a remarketing signal that’s mixed up with multiple products. These viewers are already familiar with the brand, so they’ll now see general branded images as well as products they’ve already seen.
Next, I created a more targeted segment.
This change takes into consideration the client’s key demographic along with keywords that viewers would be searching. Then Google can find this cold audience easier. Plus, I took creative hyper-specific to this audience and product mismatch so that it could find this group the simplest way possible. This would essentially cut the learning curve.
Performance Max and any other machine learnings will eventually figure out who the audience is, but that’s going to burn money and slow down efficiency.
I did the same thing here. I created three different versions of this to attract cold audiences for the client’s key product categories.
Typically, I recommend at least one asset group for each product category on a particular site. One should be a remarketing signal and one can be creative including videos and images. This makes the assets scalable.
If we click in asset groups, there’s no video, so Google would automatically create it.
This is why it’s important to create relevant videos to the products being advertised. The ads created will reach potential customers who browse websites similar to the URLs entered. This doesn’t mean a client’s ads will show on those URLs.
Also, make sure that the final URL correlates to the product page that is the most relevant.
I also developed a competitor’s look-alike audience. This is similar to remarketing where you can have general branded products or specific creative that shows a unique selling proposition against the competitors.
I also tested ticket and inmarket ticket audiences. But, I put those on pause because those weren't performing as well.
Reviewing the listing group breakdown shows which particular group was performing better over this time range sorted by cost.
The biggest difference here is capturing which groups are performing best based on return on ad spend.
So, we can see the competitor’s audience targeting all products (pic below). It has a 9x return versus some of the other keyword audiences that aren’t performing as well - they aren’t getting those clicks.
This group got a big order. And we can see ‘basketball’ is performing well. These insights allow me to scale or work on refreshing creative and low-performing segments. Then I can push scale on those select groups that are performing better.
Youtube wasn’t doing much for this client. It's important to turn this around since the platform is extremely efficient in pushing awareness of clients’ products which helps build branding and maximize sales.
So, I set up a number of specific YouTube campaigns.
Typically, I like to construct a ‘warm remarketing’ YouTube campaign,
And a ‘cold’ competitor’s YouTube campaign, targeting visitors of competitor’s channels. This works well in Performance Max. And then I set up a custom segment keywords campaign from the best performing insights and performance maps.
The trend is shown here (above and below). We’re consistently using the data insights from Performance Max to supplement other campaigns, then reviewing Performance Max accordingly.
Looking at this cold competitor’s ad group (below) - it’s a custom signal similar to Performance Max as previously mentioned.
We’re testing various videos - I typically recommend at least three to five videos depending on the niche, as well as the amount being spent. It’s beneficial to determine which competitors are spending the most, and which ones have the best cost-per-view and view rate.
We can see which campaigns are driving the most subscribers for this account. The numbers rose to over 35 thousand subscribers from zero. This information shows what’s performing as what has the lowest cost-per-subscriber being targeted. It’s also revealing competitors’ placements.
The placements report is a good stop for analysis as well.
It’s confirming that the look-alike signal being relayed to Google is relevant. We see where the ads have placed and negative match certain aspects as well.
Next, I have an audience for custom segment keywords. We’re creating an audience within ad groups because YouTube has updated its ads strategy. Now topics queue, display keywords and placements can no longer be targeted. It has changed to a conversion-focused bidding strategy.
For example, if I want to target a specific sport keyword, I would go to Audience and recognize that the audience being used in Performance Max is clearly successful on YouTube as well.
It’s driving significant subscribers, helping this account scale, as well as scaling another campaign I’m testing.
These are the main campaign types and audiences. As I mentioned with creative, it’s important to be constantly refreshing it. That can be done by focusing on Insights.
I go to columns and click on conversions. This will show a view rate as one metric. When it’s showing videos with lower view rates, it’s an indicator that there’s an audience mismatch.
However, we can also look at the Views tab to see a percentage of videos viewed. This may be handy in deciding if a creative should be swapped out so other videos can get more clicks and impressions.
In this sports eCommerce case study, it was clear that Performance Max and YouTube ads could be strategically altered to increase awareness and sales of the brand. By fixing the structural setup of Performance Max as well as utilizing testing and placement of YouTube ads, it was possible to take this client’s account to new heights while attracting thousands of new subscribers.
Shorten your learning curve, make the most of your resources, an maximize your impact both online and off.
Ted is the founder of TGQ Marketing a PPC, Analytics and CRO agency focused on client results.