We Turned Off Brand Search in Google Ads and Saved €60,000

A client was spending €5,000 a month on branded search because a YouTube tutorial said to, so we ran a geo test to prove it either way. 

In this episode of the Google Ads Unleashed Podcast, host Jeremy walks through a full brand search incrementality test for a toothbrush brand, splitting Germany into two equal regions, running a six-week geo holdout, and tracking every metric.

The results were unambiguous: revenue in the test region was only 0.6% lower, organic clicks rose 22%, and paid search orders that disappeared were almost entirely replaced by organic, confirming that branded search demand was already theirs at zero cost.

Get your free discovery call with Jeremy here: https://www.younganddigital.marketing/services#GoogleAdsManagement

Scale your store with 1:1 founder advisory coaching: https://www.younganddigital.marketing/one-to-one-coaching

Transcript

Copy Transcript Copied

Probably a topic that I've done more podcasts on than any other is the topic of branded search. Should you run it? Should you not run it? What are its advantages? What are its disadvantages? Um, I've talked about incrementality and how to test it. If you run it, then what is actually the best practice of running it, but never have I ever actually broken it down like step for step what we've done with an ad account and how we've saved um in in a case study more than €60,000 a year on branded search for one of our clients. Welcome back to Google Ads Unleashed. Hope everyone is doing fabulously this Monday. And as I've said in the intro, um today I'm going to break down something that we have uh done for a client of ours and that is we actually run a fullblown brand incrementality test. The brand is a uh bamboo toothbrush brand um which we've been working for with I think since April or something like that. And one of the cool things that I ran when uh or in our audit when when I ran the audit is I spotted that they spent five grand plus uh in euros every single month on branded search and just because a YouTube tutorial told them to do so. So something that we've been very keen to test was does actually branded search bring any additional revenue because think about it what is branded search it has its value when let's say you've got a brand new ad account and you want to pop um some conversion data through the ad account you want to pop additional volume into the ad account so that it all the non-branded campaigns can learn. It's pretty good if you run a sale and you just need it as a uh as a sort of um post for what you're doing right now and brand communication. It's pretty good if you have a specific new customer offer page where you can send people to, it's pretty good if other people bid on it. But if no one does that and if the ad account can run by itself and all the non-branded volume, what's the point if no one else bids on it? And this was exactly the case here. Uh but the client was cautious. They was sort of wondering, hm, you know, what can we actually sort of um how do we sort of make sure that we really aren't losing any revenue. Okay. So, in comes a geo test, an incrementality test. What we've done, we've split uh Germany in half um and actually had a look at quite two equal regions of Germany and looked at the revenue there and compared um the revenue and the profit in basically the test region um where we let um the uh ads basically run and the control region where we turned the ads off. We did this quite easily in the brand search uh campaign where we just removed the locations which uh we wanted to control for and after 6 weeks I believe we ran it from uh the end of August no sorry the end of July to the end of August we saw the following first of all revenue in the test region was €327,000 versus uh €329 000. So the uh revenue in the test group was only6% uh lower which quite honestly is barely anything. AOV was identical in both groups. Orders versus control um was almost the same 4 uh% difference. So to be honest, this is exactly what one would expect plus minus. It couldn't have been completely identical, but quite frankly It made hell of none of a difference. Uh we had a bit of a fluctuation in the week. Sometimes there was one dip. For instance, in the second week we had a dip of 5%. But this s*** happens, right? It's just needs a couple of orders and boom, you got a bit of a dip in a in a different region and then you got uplifts in the others. Overall though, um there was no downward drift. If brand search were really driving sales, we would have seen that the gap would have widened quite dramatically and it simply didn't. Now, you can of course argue that um you know in the grand scheme of things with a brand that does over half a million a month that 5k in ad spend on brand search how much is it going to move the needle anyway and how much would we see a gap widen but the fact that we didn't see nothing just confirmed that it just did not really add additional revenue to the plate people were just seeing the demand gen ads people were just seeing the meta ads and searching for the brand and found a branded search which would have converted anyway and to test for this We actually looked at the organic search as well in um the control group which um um sorry I got it mixed around in the test group sorry which dramatically grew of course paid search revenue more or less h haveved right um uh as as it would because we took 50% of the budget away but in the test region surprisingly the organic revenue actually grew by exactly that amount to compensate for uh what we would have lost on paid search which is really funny. Uh so it just means we proved by that that the uh paid search demand literally just on the traffic moved over to the organic uh side of things. I'd highly recommend by the way um to have a look maybe at the YouTube video that that um is going to come out with this as well because I'm actually presenting a fullblown deck there where I'm showing all of the numbers. Okay. Um What we also did to back this up further is we actually had a look at uh search console because uh we wanted to have not only Google Analytics 4 as a sort of source of uh traffic where we could see organic traffic rising but we also saw this in uh search console as well and um that happened uh exactly as predicted. Clicks rose 22% against the baseline run rate and um it's just a separate sort of source that showed us that clearly a lot more clicks and a lot more views um happened because fewer people saw um because yeah more people saw of course the brand search. Overall the test regions came in 1,923 below control over 4 weeks against £327,000 in total revenue. This is basically a negligible gap. Uh this is a gap that sits inside the completely normal weekly variation and revenue held completely. Paid search revenue in the test regions was 15,600 against 23,600 in control. A shortfall of 7,940. And the paid orders uh paid search orders on branded search fell of course as expected by 34% as you would if we drop all of that. But organic search increased by 44% and the revenue uh was 33,000 versus 23,189 on a normal week, which is a gain of 10,000 more. than the paid search shortfall on organic row orders rose 45%. This is not uh a coincidence. This is um exactly what we would have um sort of seen. Uh a second source agree search console I said 22% it was up um in in total clicks that is independent of Google Analytics 4 confirming this as well and search overall grew 5% anyway in that time frame just to give you reference of how much um overall search traffic uh did. Okay. Um brand search was clearly largely non-incremental and the demand was already ours and organic captured it for free. So across um everything this now means that we have turned brand search off not only here but also Microsoft ads where we've seen a similar situation saving the client €60,000 in one year and probably paid for us um uh until the end of time uh of working with this client. Just an insight like that. Okay. Um what's also quite interesting is um uh maybe something to take away from this is that it doesn't really stop here. It means that we also have to keep an eye now on competitors. Are they suddenly bidding on our branded search? So as a weekly sort of to-do, we still need to check are other people bidding on us. What is something that um yeah, what do we do if others people other people bid on us. So, it's just something that we need to keep an eye on. But yeah, really good lesson on why you should always question what you do and maybe some of the best practices in PPC aren't always best practices. And if you want someone who uh thinks this creatively, get in touch, Jeremy.marketing. You can also head to the website.marketing. You can also find me on LinkedIn, Jeremy on Google Ads. Just head there, send me a message. Would love to connect with you and have a bit of a chat about PP. See um maybe some upcoming things. Um I will uh uh speak at a PPC roundt with profit metrics on the 17th of September. So keep an eye on the profit metrics website when this comes out and their socials. I'm going to be there with uh Ruben Roniboom from uh Task Force and with um uh uh I think Sam Khaled from Profit Metrics. And I'm also going to be with my good friend. One second. Uh I think it's Thomas Etchel. Yeah, that's it. Uh who has been on the podcast before and it would be awesome if you uh looked into that. I will also be speaking at PPC Live in London uh late October. I believe it is the 21st of October. So if you're around uh let me know. I would love to see you there. Um this has been Google Ads Unleashed with Jeremy Young, your personal ads expert and I wish you a happy and productive week ahead.