Why Cutting Too Many Keywords Can Tank Your Performance

Think adding negative keywords always improves results? Think again. 

In this episode of the Google Ads Unleashed Podcast, host Jeremy explains why most advertisers are overdoing negatives—and how it can actually hurt performance. You’ll learn when to trust Google’s smart bidding, why data (not gut instinct) should guide exclusions, and how to run an A/B test to see if cutting back boosts results.

👉 Curious if your keyword lists are doing more harm than good? Tune in now and rethink your negative keyword strategy.

Get your free 30 minute strategy session with Jeremy here: https://www.younganddigital.marketing/

Scale your store with 1:1 coaching: https://www.younganddigital.marketing/1-2-1-coaching

Transcript

Hello and welcome back to Google ads and leash guys. Hope everyone is doing fabulously this Monday, as every week, and with this week, I'm gonna broach something that might ruffle some feathers, because I've seen it firsthand myself. I've actually seen a few posts the other day online, which inspired me to do this podcast, along with, for instance, my friend Vincent Bima, who is a very well known influencer, or Andrew loch, who is a very famous influencer who's sort of highlighted this topic. So maybe there's an interview following on from this, and that is a claim that I'm gonna put out there that most people are overdoing negative keywords. So I'll be honest, until very recently in this sounds nuts, right? Jeremy, what the hell are you talking about? Negative keywords is one of the core things that you need to do in order to, you know, cut wastage in an ad account to teach Google to search for the rights kind of search terms and in a in, in, you know, your prospecting campaigns, etc, etc, but you are wrong. Most people are overdoing it now, and until very recently, we've been doing negative keywords, quite frankly, the old fashioned way, but now we've changed our minds on it completely, and I'm here to change your minds on it as well and to educate you. So why do most people overdo negative keywords? I think something that I've really realized was last year. So last year, I had a client for maybe two or three months, and it didn't go well. Basically, it was very odd. I thought it was going to turn out to be an amazing client, large B to B, but basically and did loads of volume, loads of scale. But basically what had happened is that they didn't trust their own data, which sounds weird. Basically, what had happened was that they couldn't understand that volume of inquiries doesn't actually equal, you know, value of inquiries, right? So we've implemented offline conversion tracking. Had a lot fewer inquiries, generally speaking, but they were all a lot more profitable, a lot more valuable. I'm pretty sure that I've done a podcast episode on that. And it turns out their only goal was really to fill up the call centers, right? So they expanded the call center, and they just wanted to have them busy in the call center, and valued volume over value, which is really strange, but there we go. That's why we weren't the right turns out to not be the right agency with them, but one of the things that had happened was the following, that they didn't really have any negative keywords at the time, right? So they've had this huge account spending loads of money on various search campaigns, and they had no negative keywords. And one of the first things that we've done is we went through the ad account, we've created a big old negative keyword list, and we launched a campaign and added it to it, and it bombed. It did poorly. It did worse in the test than it did with the one without the negative keyword lists. So what has gone on there? So basically, what had gone on is that smart bidding has gotten so good at including the correct search terms and to bid correctly for the for the right search terms of which it will know that it'll generate conversions that we had a better result, not restricting the algorithm, right? This is really, really strange, because that used to not work this way, right? It used to not work this way because the smart bill bidding algorithm simply wasn't good enough, right? So we had to kind of give Google a bit more leeway. But how it now works is that when people when Google bids on a search term, it doesn't actually just bid you know based on the semantic value of what someone is searching and shows you know the ad based on where their keyword is triggered or a product or whatever, but includes things such as, what device is this person on right for instance, does a search term convert on mobile, but it doesn't convert on desktops whatsoever. You don't see this in the data, right? Or you know that smart bidding takes this into consideration, because that's what's shown in your ad account, but you don't see this when you go through the search term, and you can't, as a human, even make that differentiation. Right, it doesn't take into account gender. It doesn't take turn into account time of day. A search term might work on a certain time of day, but it might not right. It may sound crazy that this is a case, but this is de facto, while smart bidding works, and all the signals that smart bidding takes into consideration you don't know the previous history of something someone has searched, right? For instance, broad match operates that way. Broad match takes into consideration the previous searches of a searcher. So if someone has searched multiple let's say you sell desks, right? And someone has multiple times search for desks, he's been researching for months what the best desk is, and then suddenly they search for something like, just furniture, right? Something really generic. You might think, Oh, this is irrelevant, or this is too loosely relevant. Well, dude, the it turns out they've been, like, looking at the ad for months, and now is the time that they struck. So by excluding that, you're actually preventing the algorithm from doing its job, right? You also have no idea if a search term might be relevant for one keyword and less relevant for another. You can only see that, of course, if you can have the keyword in different ad groups, right, or if you look at the search term on a keyword basis, but it's the same within a campaign, right? You might actually have a catch all shopping campaign, and then you see an apparently irrelevant search term in there, but you don't know to which product that applied. It might be relevant for one, but it might not be relevant for another. If you exclude it, you're doing more damage than you're doing good. So we've changed this now, right? So we used to go in all of our accounts through all of the search terms on a weekly basis, and we still do right, especially the low volume ones, which are relatively easy to do, which very often attract a lot of shit, right? We just do that, do that. But we actually have now resorted to using, of course, scripts, right? Like, for instance, Niels royman has a fabulous script, Google it, if you want. You might get some traffic from this, where you can actually gather data and make data driven decisions to exclude search terms we use n grams. I've got a full episode on N gram analysis and how they work and or, of course, we still exclude when something is absolutely obviously nonsense. Okay, so what a lot of people now do wrong is that they basically add search to negative keywords on a feeling, right? They see a competitor's name and they say, Oh, this, this needs to be excluded. Well, should you exclude it, though? Because you don't know what this person is like. Maybe they are searching for a competitor because they've done 50 other searches beforehand, maybe not many. But you get the gist looking at all the various things in the market, right? And they are just at the moment, and for the 10th time your ad is shown, and that is maybe the final time where they actually pull the trigger, or a lot of people don't use, you know, don't rely on statistical data enough, right? So you should be doing some analysis, and whether you should like the N gram analysis, right, whether you should be actually excluding those search terms properly as negative keywords. And, yeah, so that's usually the mistakes that people do, and what you should be doing now is like I said, Do the n grams use? You know, scripts like from Niels royman, obvious nonsense can always go and what you should be doing is on low volume accounts, or when you have just recently launched new ad account, then the smart bidding algorithm just does not have enough data in order to make an informed decision. So that is where you need to jump in and sort of steer it in the right direction. With new ad accounts, for instance, we do negative keywords on a daily basis until it's stabilized itself, and then sort of start progressing to our normal SOPs, which have now rewritten. Or we do this with, let's say we got a few B to B clients with slow, manual bidding, right? Because it's so niche and so low search volume that there's no, no smart bidding involved. Basically, that is where we sort of supplement there on a regular basis as well. What I'd like you to go away, and what I encourage you to do now is super, super interesting, if you're running out of AB testing ideas, is to do the following, set up an experiment. Go into my podcast about how to run experiments on Google ads and what types of experiments to run to see how to do this. But yeah, run an experiment and AB test the following. What would happen if. I remove the negative keyword list completely, does it actually improve performance, or does it maybe even worsen performance? And the result may surprise you. And if you want more tips like this, then please give the podcast a follow. You can also reach out to me personally, at Jeremy, at Young at Digital dot marketing, or on LinkedIn Jeremy and Google ads, I would love to have a chat with you, see if I can help you, and especially e commerce founders, we got some spots available still for q4 coming up. Now. We're expanding the agency and recruiting at the moment, so we've got some more slots to fill to help you get your best q4 ever. This has been Jeremy young with Google Ads unleashed your personal Google Ads expert, and I wish you a happy and productive week ahead.