Inside Google Ads podcast: Episode 94 - Is Google Ads Stupid?

Have you ever opened up your Google Ads account, looked at what was going on and thought, “Oh my gosh, Google, you're so stupid”?

We've all been there, but as someone who used to work at Google and now sees hundreds of different Google Ads accounts a year, I wanted to bring you some perspective that might help you get over the frustration and into action. Because the things that might seem stupid to you may, okay, be stupid, but may actually have good reasoning behind them that just don't happen to be aligned with what you want.

This episode is inspired by two great LinkedIn conversations I got into recently. So I'm excited to share those here with you today.

I'm your host, Jyll Saskin Gales. I spent six years working for big brands at Google, and now I work for you.

This is Inside Google Ads: Episode 94, Is Google stupid?

Spoiler alert: the answer is probably not.

First, I had this great conversation with Martin Sandgaard on LinkedIn. They shared a post that said, “This is why you don't enable AI Max on your Google Ads Search campaigns.” And in the screenshot, you can see they searched “what converts” and they see an ad for TikTok ads. Specifically, the ad headline says, “the platform for quality leads - gather high quality leads.”

Now, first things first, I get where Martin's coming from. This seems like a stupid thing that AI Max or Broad match or PMax did when he's searching for a very specific brand that is not at all what TikTok ads does.

But I actually had a bit of a different perspective on this. You see, WhatConverts is a platform for lead generation businesses. And Martin is a Google Ads freelancer. His bio on LinkedIn says, “Make more money with Google Ads.” So by searching for “what converts” and taking everything else Google knows about Martin, he is a Google Ads practitioner who works with lead generation clients. 

You know who the target audience probably is for a TikTok ads ad about using TikTok for lead gen? People who use Google Ads and Meta ads for lead gen! It’s trying to convince them to use TikTok ads as well. So while this ad did not answer the Search for “what converts,” it absolutely was serving to the right audience. 

And this is what I think so many people forget about when we think of Search as keywords, query, match, and go. But it's so much more than that now.

Perhaps this wasn't the perfect search for that ad, but this was the perfect user for that ad. And in AI-powered campaigns, that's actually what matters. 

Now, this is one isolated example. In aggregate, I'm sure that “what converts” is not going to be a common search term in this campaign. But assuming Google's not stupid, always good to start with that frame of reference, it knew exactly what it was doing here to get the best possible results for the TikTok ads campaign specifically trying to get lead gen advertisers.

Also remember that because Martin didn't click on this ad, that impression was free. TikTok didn't pay for that. And because this was a branded search for what converts, even if you had clicked on that, it probably would have been pretty cheap. Not so cheap, right? Because it would have a low click-through rate, which means low quality score, which means higher CPCs. But when you think about the lifetime value of a customer to TikTok ads, we're talking in a worst case scenario, hundreds of dollars, best case scenario, six to seven figures a year - yeah, I bet that trade-off for that reach is absolutely worth it.

My takeaway for you from this example is when you see stuff pop up in your search terms that you're like, that looks really dumb, or on the user side like Martin, if you see an ad pop up and you're like, that was really dumb, just pause for a moment. Put yourself in Google’s shoes and think, hmm, assuming I'm smart, why would this be a good thing?

I'm not promising there's always going to be a good answer. AI makes mistakes, humans make mistakes. It's how we both learn. But with that mindset, it can really help you get better results for your business when you try to understand the system and work with it rather than fight against it.

So the next time you see someone posting some clickbait thing on LinkedIn or X, like, “Oh, look at my search terms report from AI Max from day one, it's so dumb!” Don't engage with that. 

Yes, obviously AI Max starts dumb. It's day one. It has no data yet.

Can you imagine if I gave birth to a baby and then the next day I was like, “oh, this baby is so dumb, it can't talk yet.” Like, yeah, it was just born yesterday. It has barely heard speech yet. It hasn't had a chance to develop its muscles, its brain or anything else it needs, but it's going to get there. It's going to learn. And thankfully AI Max learns a lot faster than babies do. 

Now, I'm not saying you should necessarily use AI Max, but again, try to understand the system. It'll make you a better practitioner and you will get better results for your business or your client's business by working with the system rather than fighting against it.

All right, this next example I'm going to share with you, trigger warning, it genuinely hurt my eyes when I first saw this. This is a client who I've had coaching calls with. They hired me to audit this other Google Ads account. When I opened up the account, I saw what I'm about to show you right now. Don't worry if you're listening, not watching. I'll describe it for you as well. And I literally had to close the account and come back to it an hour later.

Here we go. It is one Search campaign. And the longer you look, the more things look wrong. 

First, we can see from the icon that the Display is turned on in this Search campaign. You never want that. 

Next, I have a segment by conversion action turned on, so we can see that there's like eight or nine different conversion actions here. And some of those conversion actions that we're optimizing for in this campaign include user engagement, scroll, and page view, among others. 

So, this campaign has tens of thousands of conversions. In fact, it is showing a 536% conversion rate, which means every single person who visited this website from this campaign converted five times. 

That’s bananas, right? 

And if we look deeper, we see there's one conversion action that's not being imported from GA4. It's called contact form. And there are 5,424 contact form conversions. The problem? There's only 2,700 clicks. So again, whoever set up this contact form conversion definitely didn't set it to just trigger when a contact form is submitted. Also didn’t even just mess it up to trigger when a page view happens. This is happening on average twice per ad click.

And then if you even get beyond both of those things, you're going to see that the average CPC is $0.85, which in this day and age, even on your brand name, for the most part, you cannot expect $0.85 CPCs on 2,700 clicks. That's just… anyway, so I really wanted to understand for my client what went wrong here because my client is a business owner, a service provide,r and what I found out next made me feel even worse about what I see here.

This campaign was originally set up by a marketing agency, or someone calling themselves a marketing agency. So the legacy of having Display and Search Partners in here came from a marketing agency and the legacy of importing all these different conversion actions came from a marketing agency. Shame on them.

However, about two to three months before I did this audit, a Google account strategist with an XWF email address (I'll get to that in a moment) reached out to my client and started working with them. 

XWF means that they are a contractor working for a third party contracted by Google, not directly employed by Google. So usually, although these are lovely people, this is nothing personal against XWF, but you don't have to have any Google Ads experience to get that job. They're given pretty limited training. They just have a list of things they're supposed to recommend to everybody. They are not trained to actually use Google Ads or actually take your unique business goals into account.

But still, you would think that this person from Google would see what's going on here and say, hmm, maybe we should fix the conversion tracking? Sadly, no. This person decided to double down, add a bunch of different values to all these different conversion actions, and then advise my client to turn on Target ROAS bidding, which means now from this campaign's perspective, it thinks it's kicking butt, exceeding its Target ROAS, when actually all it's doing is driving a whole bunch of page views and no actual real life form submissions.

Now it's tempting to say, especially with the name of this episode, that Google rep is so stupid. So I will be very clear: I am not saying that.

I think the training and incentives provided to this person are very stupid, which is why this happened. Let me explain.

This person obviously does not have an actual business understanding of how Google Ads works beyond just what's a metric called and what's a conversion action called. The real business sense to understand this? Not training that person was given, because then even with all these GA4 conversion actions, even if for some reason you wanted to keep those all in, you would see this contact form tag-based conversion is just not configured correctly. And this rep didn't choose to take that out, and they didn't choose to fix it, they just left it there. So no matter what they did with the rest of the conversion actions, this was always going to be a mess. I put that down to training, not the individual.

But the other piece is incentives. What this rep did is get my client to leverage value-based bidding. And pushing aside the specific case, I think we can all agree that generally, in a lead gen account, it's a good idea to bring in a few different conversion actions along the funnel, assign different values to each of those, feed those values back to Google Ads and use Target ROAS. That is the value-based bidding narrative to a ‘T.’

And of course, I would imagine this rep is incentivized to get their clients using value-based bidding and using Target ROAS. They probably got a giant pat on the back from their manager after this. The problem, of course, is this client was not a good candidate for value-based bidding. Value-based bidding makes sense if you want to measure, for example, “interacted with the form,” “submitted the form,” “qualified lead,” and “customer.” It is not meant to be used for “visited the page,” “scrolled on the page,” “clicked on the page,” right? Those are not helpful things for value-based bidding at all. 

So again, these were not good things to do. Objectively, I know in Google Ads, so often we say, it depends. There's no “it depends” here. This is the wrong thing to do. But that's because the training and incentives given to this person advising my client were simply not right.

I have advised my client not to take calls from Google anymore, and instead, I'm going to help her clean this up, which probably means starting a whole new campaign rather than trying to wade through this mess.

The purpose of sharing these stories with you today is to show you that at first glance, it's really easy to say, Google's so stupid. Google just wants to steal your money. Google doesn't know what they're talking about. 

I don't find that mindset helpful. It may be true. It may not be true. But I encourage you, if you're trying to be effective at managing ads, to pause, put yourself in Google’s shoes, whether that's the Google algorithm literally (in example one) or a Google sales representative’s shoes (as in example two) and think: How could this be a good thing from their perspective? What are they trying to do? Is there a way for me to work with that rather than fighting against it to get results for my business? 

In this case, working with Broad Match or AI Max can absolutely drive great results for your business with sufficient budget, conversion data, and time. In this example, working with this Google rep no longer made sense for this business, but it could potentially make sense for yours if you get someone who can genuinely add value for your account.

Today's Insider Challenge is this. I want you to think back to the most recent time you were using Google Ads, saw something and thought, that's so stupid. Take a deep breath. It's going to be okay. We'll get through it together. My challenge for you is to try, really try and think, is there a way that from Google's perspective, this could actually be a useful thing and not a stupid thing? And then if you can keep going with the thought exercise, go a step further to think, is there a way this could potentially be a positive for me and not a negative? 

I'm not trying to get you to gaslight yourself, okay? I'm not saying Google is always right or Google always knows what it's doing. But I encourage you to practice, through the Insider Challenge, just being open to the possibility that sometimes a different perspective may be all you need.

The beauty of the Insider Challenge is, of course, there's no right or wrong answer. Just an opportunity to stretch your brain on real-life Google Ads problem solving.

Last Episode’s Challenge, Episode 93, was this. Let's say that you want to run a Demand Gen campaign from the middle of November to the middle of January. Would you choose an average daily budget or a campaign total budget? How would you make that decision? 

This is trickier than it seems at first glance because, of course, the middle of November means we're right dropping into Cyber Monday, Black Friday, holiday season, and then potentially a slump right afterwards. Unless you have a different seasonality, maybe if you work in the fitness space, like January is going to be your peak. Or maybe if you're in a B2B space, November might be busy, but then December will be totally dead. So of course it depends, but no matter what industry you're in, you're probably having some kind of seasonality happening through November, December, and January. 

If we use an average daily budget, it means that half of November might get a little wonky, December might be normal, and then the first half of January might get a little wonky. 

If we try to use a campaign total budget to even that out, the challenge could be if things are going really well in November, it could spend a lot of budget then and then just have nothing left by mid-December, nevermind January. Or on the flip side, it could go steady pacing through November, try to steadily pace through December, but then if demand totally drops off, you could get to the end and then it hasn't spent your budget and it's just going to blow it all in January. 

Anyway, I say all this to say, while there's no right or wrong answer here, of course, I would probably still be biased towards doing an average daily budget. That would at least let me have a monthly spending limit for the month of December. And then depending on the seasonality of the business I'm working with, late November and early January may each need higher budgets or lower budgets than what we need in December.

That's my thought process, but I'll be honest, I'm not sure I even love that answer. So what do you think? Would you do something similar or something different?

I'm Jyll Saskin Gales and I'll see you next time Inside Google Ads.