Learning to trust the lack of transparency from ad platforms
Ad platforms are clear as mud in helping you look under the hood. But at some point, we have to trust them and focus on what we can control.
As I write this article, I see the irony in the headline…
Learning to trust three major platforms – Google, Bing (I refuse to call it Microsoft Advertising), and Facebook (I still hate calling it Meta) – which just in the past month have laid off tens of thousands of employees.
And then on the first working day back in the office since Google’s mass layoff, the Google Ads UI goes down for 3 hours – the longest I can recall in my 18 years of doing this.
Let’s just pretend to ignore that elephant in the room and discuss how we can learn to trust what we cannot see (anymore) on the platforms.
With a series of changes on the platform side (Performance Max, Facebook Advantage+ or Bing’s change to networks to include “select traffic” in it), the current state of transparency of where, when and how our ads will show is fuzzier than ever.
And don’t even bother asking Yelp for detailed reporting (in their defense, they have always remained constant though).
Meanwhile, the ability to quantify some of this data smoothly on the analytics side (hello, GA4), has become as successful and viable as being a starting New York Jets quarterback for more than two seasons.
You can say, “Watch me, Jon! I will pull out of Google or Facebook and be just fine.”
Once I am eventually done laughing and after you’ve explained to your CMO that you stopped one of your website’s largest traffic sources because the big bad conglomerates won’t give you more details, then we can have a serious chat about reality.
Here’s the thing:
As much as I complain about the platforms’ lack of transparency, the reality is that we, as advertisers, do not have a choice in the matter. In a privacy-obsessed era, ad platforms bend to such demands – and certain foreign and domestic government bodies.
So upon facing the reality that the level of transparency and insight we loved is coming back a week after never, it is time to learn to trust the new normal.
Platforms are not wanting your money just today
Despite what we think of them, not being able to identify how many Performance Max impressions are coming from GDN placements is a pain in the butt, but not evil. They don’t plan on leaving us broke with garbage performance in a month and running away.
Every ad platform under the sun knows, its number one goal is to make money for themselves.
But if they sell in an ad capability with no transparency and poor performance, then the advertisers pull out and they won’t make the revenue (a fun reminder of Yahoo Gemini).
So this has to instill the belief in an advertiser that just because the platform wants your money, doesn’t mean they don’t want to grow you profitably. Remember folks, ad platforms look at the lifetime value of an advertiser rather than just the quarter.
With that in mind, just because you can’t see everything pre or post of a launch, doesn’t mean it is wrong. It does mean they operate with a bit of a black box though.
A platform, with all its sophisticated algorithms, in an absolute best-case scenario (i.e., ideal volume, conversions, seasonality, budget, tagging, linking, etc.) and under non-“unique to me” scenarios, will likely produce inherently better performance within their black box, than anyone on manual.
They do this not because they are taking your cash and running, but because they believe in growing your business.
Get the daily newsletter search marketers rely on.
Despite what you believe, you are still in control – sorta
We as advertisers have the ability to turn off an ad unit. But one should only do it if they have first evaluated the following:
- Do I understand what the ad unit is doing?
- Do I understand how the ad unit is delivering its results?
- Are the results it produces heading in the appropriate direction for my goals?
So, this is where things get sticky. You may evaluate all three of these, but if you can't answer the first two questions and the last one is a "yes," it may just outweigh everything. Or as I often call this concept, "If it ain't broke, don't fix it."
We've all taken this approach at one point or another.
That moment, when you can't explain to the CMO where your ads are, what they show for, and not entirely who they show to, but that 1,000% is looking mighty fine.
Unless you've hit a brand safety issue (we'll get into that in a moment) but you can produce quality and quantity at the same time, sometimes you may have to accept that your target (e.g., women 25-44, living in 06117, who graduated from *my alma mater* Hofstra University, who are fans of the New York Jets and enjoy reading content about vacations to Mexico) might actually be someone different than who you were going for (e.g., women 25-44, living in 06117, whose kids attend Hofstra, but are actually hecklers of NY Jet fans and prefer day trips to go leaf peeping in the Poconos).
But Google felt your ads should show for them as well because when they look at all behavioral points (many we can't see or even comprehend), they realize there is another group close to your ideal audience that, if shown that one random Gmail ad spot, is more likely to convert than your ideal target.
Every creative director or graphic design major wants the final say on the creative layout, collaboration, and placement of their ad.
But guess what, the number of mental synapses to comprehend every scenario, for every consumer, would:
- Make one's brain explode.
- Unless you have a team of 1,000 robots working around the clock, will consume too much time to execute.
Once again, you the advertiser have the directional control, and more importantly, the kill switch. But you need to trust the platform to fine-tune it.
You can still get the details, but you will have to work harder to get it
You are not destitute to have no concept of where your ad triggers. (Disclaimer: I know what you're thinking – and I agree. GA4 really does make everything a living nightmare for data collection, courtesy of sampling thresholds.)
But simple scripts and parameters, if you're willing to give up something, can allow you to track additional data and placements within the URLs that can be passed back into your analytics or CRM system.
Will it be a pain in the butt to pull? Absolutely.
Is it necessary to deal with the necessary evil? Also yes.
Courtesy of the feds, the big 3 platforms aren't being secretive for fun (I hope), but because a lot of data has to be blinded. You just need to be willing to take the time to hunt for it all.
Yes, you have lost a lot of transparency on the minutia. You are more now than ever, at the mercy of the black box method.
But this has now put a renewed emphasis on A/B testing. Accept what you cannot control (placement, audiences, the loyalty you have to a specific Real Housewives franchise), and focus on what you can still control:
- Landing pages.
- Audience segments.
- And more.
Once you focus on those, do every possible test there is where you can have a true control element:
- Ad copy/images.
- Landing pages.
But the most important thing is to watch the performance. If it is not panning out, then just stop it. Pause it, regroup, replan, and go down your checklist for your next paid search game plan.
This will help to reduce the lack of transparency while still accepting the lack of transparency, helping you move on to the future of the next hidden gem to ruin your month as an advertiser.
Despite what you think, not knowing what % of your Bing search traffic is coming from Yahoo.com, just will not be worth the time to figure out.
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