A few weeks ago Google updated its Customer Match policy to restrict the types of advertisers that can take advantage of this targeting capability. This caused a bit of a backlash from some paid search marketers who felt the new restrictions unfairly constrained the ability of some advertisers to use Customer Match targeting.
Here we’ll outline the new changes, as well as why restricting access to Customer Match was probably a move Google needed to make in some way.
New Restrictions Seem Targeted at Limiting Bad Actors
Customer Match targeting is now only available to advertisers with:
- A good history of policy compliance.
- A good payment history.
- At least 90 days history in Google Ads.
- More than USD 50,000 (or other currency equivalent) total lifetime spend.
Additionally, advertisers looking to begin using Customer Match targeting must now go through Google account managers to request access.
This is a pretty significant shift from Google’s original Customer Match policy, when the main restrictions were that:
- Email addresses, phone numbers, and physical addresses loaded into Google Ads must be acquired firsthand from users.
- Customer Match audiences must be made up of at least 1,000 identifiable members to qualify for use in targeting.
Advertisers were responsible for ensuring these criteria were met, though Google alluded to the ability to detect when information that was not acquired firsthand from users was uploaded.
Of course, I’ve spoken on Customer Match at conferences several times, and the first question that inevitably gets asked is ‘How would Google know the difference?’ It’s a good question, and one I’ve never really had a good answer for, though it also explains why Google is now locking the Customer Match gate a bit tighter.
With Google’s previous enforcement essentially an honor system, there were no doubt advertisers that took this freedom to abuse the system. For example, smaller advertisers just getting their businesses off the ground likely saw little potential for consequence in buying email addresses from relevant groups and using these to target searchers through Customer Match to better qualify traffic.
These changes in policy seem aimed at such actors, ensuring that brands already have a meaningful history of spending on Google Ads, following policy, and are fully aware of the responsibilities they take when creating Customer Match audiences via conversations with account managers.
The update also indicates that Google never really had a reliable way to accurately identify good and bad actors, and thus felt the need to create hurdles for all advertisers to prevent abuse.
It is a shame that some smaller advertisers won’t be able to meet these criteria. At the same time, Customer Match is a small part of the overall paid search ecosystem at present.
As mentioned earlier, only audiences that contain at least 1,000 identifiable emails, phone numbers, or physical addresses are eligible for use in targeting. While Google’s match rate for Gmail addresses is typically above 90%, email addresses from other domains are matched at much lower levels. Thus, some smaller advertisers have trouble creating even one usable Customer Match list.
Even for Merkle advertisers, most of which are large well-known brands, Customer Match audiences accounted for only 3% of all Google search ad clicks in Q3 2018.
These shares can certainly be higher for some advertisers. For example, lesser-known brands might have trouble making non-audience paid search traffic profitable and thus lean more heavily on targeting qualified audience members.
Still, if those brands are so small as to fail to spend $50,000 in ad spend over the course of their Google Ads history, it stands to reason that they’re also not spending very much on Customer Match audiences specifically. As such, this probably seems a small price to pay for peace of mind from Google’s perspective.
Advertisers that no longer qualify for Customer Match do still have Remarketing Lists for Search Ads (RLSA) available for targeting and can create specific audiences for email signups and converters to target many of the users that might have been included in Customer Match audiences. It’s not quite the same as being able to use CRM data associated with email addresses for optimizations, but it is a way to better qualify traffic from those users most familiar with a brand.
It’s a shame Google had to implement these new restrictions, but the old honor system was ripe for abuse. In digital marketing, as in many facets of life, a relatively small number of bad actors are a big reason why we can’t have nice things.
Still, in the big scheme of things this probably shouldn’t have a huge effect on most advertisers acting in good faith within the restrictions, and there are RLSA options available for smaller advertisers to use to target at least some of these individuals in a qualified manner.
On Google’s end, it makes sense to put in some safeguards given the current climate surrounding use of personally identifiable information (PII) in marketing. Bad actors using emails inappropriately may have caused headaches for the search giant at some point and snuffing out at least some of that activity almost had to come in some form.
Whether the new criteria were the best way to go about such quality control is up for debate, and perhaps some sort of machine learning option might have accomplished the same goal without uninviting smaller advertisers from the party. However, only Google would know the details of the pros and cons of this update versus other possibilities. At the end of the day, Google clearly thought this was the best way to protect itself.