1 00:00:15,000 --> 00:00:18,420 Dara: On today's episode, Dan and I talk about the very recent news that 2 00:00:18,420 --> 00:00:21,720 both Google Analytics and Google Ads will be getting rid of some of 3 00:00:21,720 --> 00:00:24,030 the rules-based attribution models. 4 00:00:24,630 --> 00:00:28,740 Leaving only data driven and last non-direct click. 5 00:00:28,890 --> 00:00:32,490 Daniel: So rest in peace, linear, time decay, position based and everything else. 6 00:00:32,800 --> 00:00:35,160 But if you want to learn a bit more around Google Analytics, what it has to 7 00:00:35,160 --> 00:00:39,008 offer we are currently underway with our first Google Analytics 4 training cohort. 8 00:00:39,008 --> 00:00:42,644 This kicked off a couple of weeks ago, it's going amazingly six week program and 9 00:00:42,974 --> 00:00:44,744 we've got our next cohort runs in June. 10 00:00:44,744 --> 00:00:48,524 So if you want to join the last opportunity to jump on boards and learn 11 00:00:48,524 --> 00:00:52,379 Google Analytics 4 with a bunch of peers and professionals and experts, then 12 00:00:52,379 --> 00:00:56,181 that's your opportunity before the big day that Google Analytics gets switched off. 13 00:00:56,451 --> 00:00:59,881 So check out measurelab.co.uk/training for more information. 14 00:01:00,061 --> 00:01:02,401 And we're currently running a early bird promotion that gives 15 00:01:02,406 --> 00:01:04,891 you 25% off while stocks last. 16 00:01:05,101 --> 00:01:05,761 Dara: Enjoy the show. 17 00:01:06,301 --> 00:01:09,401 Hello and welcome back to The Measure Pod, a podcast for 18 00:01:09,421 --> 00:01:11,221 analytics and data enthusiasts. 19 00:01:11,551 --> 00:01:13,801 I'm Dara, I'm CEO at Measurelab. 20 00:01:13,921 --> 00:01:17,291 Daniel: And I'm Dan, I'm an analytics consultant and trainer also at Meaurelab. 21 00:01:17,626 --> 00:01:19,246 Dara: So it's just you and I again today, Dan. 22 00:01:19,246 --> 00:01:22,096 It feels like it's been a little while, so we have to come up with 23 00:01:22,096 --> 00:01:23,716 something interesting to talk about. 24 00:01:23,716 --> 00:01:26,684 So I'm going to turn to you, put the pressure on you to come up 25 00:01:26,684 --> 00:01:31,034 with a interesting and engaging and challenging topic for us to cover today. 26 00:01:31,274 --> 00:01:34,814 Daniel: Well, luckily Google seems to change something every five minutes, so 27 00:01:34,814 --> 00:01:37,724 I don't have to come up with anything when Google have announced a bunch of 28 00:01:37,724 --> 00:01:41,084 changes that I think are going to be the topic of conversation for today. 29 00:01:41,284 --> 00:01:44,624 The main one and the biggest one and possibly the biggest change they've made 30 00:01:44,624 --> 00:01:48,929 since ,dare I say, biggest change they've made since they've decided to deprecate 31 00:01:48,929 --> 00:01:52,379 Universal Analytics is that they're removing a number of the attribution 32 00:01:52,379 --> 00:01:54,509 models that are available in GA4. 33 00:01:54,899 --> 00:01:57,209 And now these are attribution models that have been around 34 00:01:57,389 --> 00:01:59,009 since the beginning, in a sense. 35 00:01:59,009 --> 00:02:00,809 They've been around since Universal Analytics. 36 00:02:00,814 --> 00:02:03,329 Even in Universal Analytics, we can even create custom attribution models. 37 00:02:03,329 --> 00:02:06,839 But finally, they've come to the end of their life cycle in Google's eyes, 38 00:02:06,839 --> 00:02:10,169 and they're going to be switching off a few of the rules based attribution 39 00:02:10,169 --> 00:02:14,039 models specifically, they're going to be removing first click, linear, time 40 00:02:14,039 --> 00:02:19,019 decay, and position based attribution, leaving only last click, last ads 41 00:02:19,019 --> 00:02:22,349 preferred click and data-driven attribution to be used by us all. 42 00:02:22,569 --> 00:02:23,849 Dara: Good old trusty last click. 43 00:02:23,899 --> 00:02:27,819 I could just imagine all of the people who would be truly devastated 44 00:02:27,819 --> 00:02:30,359 if they announced the news that they were getting rid of last click. 45 00:02:30,359 --> 00:02:34,569 But at the risk of jumping ahead slightly, we'll park this but one of the 46 00:02:34,939 --> 00:02:38,569 questions, one of the first questions that comes to my mind is, who cares? 47 00:02:39,704 --> 00:02:41,894 Daniel: This is going to be another classic Dan versus 48 00:02:41,894 --> 00:02:43,244 Dara conversation I think. 49 00:02:43,424 --> 00:02:47,774 I think I care, I care not just because I cut my teeth in this industry in 50 00:02:47,774 --> 00:02:51,554 an attribution platform, or I've been using attribution or the variety 51 00:02:51,554 --> 00:02:55,117 of attribution models, should I say since I've ever been using Google 52 00:02:55,117 --> 00:02:56,727 Analytics and even beforehand. 53 00:02:56,827 --> 00:02:59,977 The way I've always explained attribution when I'm talking to clients 54 00:02:59,977 --> 00:03:03,877 or colleagues or people I'm training around Google Analytics, especially 55 00:03:03,877 --> 00:03:07,487 when it comes to marketing analytics is around having access to data. 56 00:03:07,487 --> 00:03:11,177 And I think an attribution model is just a lens you can apply on your data to get a 57 00:03:11,182 --> 00:03:13,127 different perspective of what's going on. 58 00:03:13,137 --> 00:03:16,407 For me, what I'm feeling like at least right now initially is that they're 59 00:03:16,407 --> 00:03:19,767 removing these lenses, which means I'm removing the ability to build a 60 00:03:19,767 --> 00:03:23,642 better narrative around what's going on with my customers around my marketing. 61 00:03:23,782 --> 00:03:26,902 Dara: To be clear, and let's maybe, this is probably a good point to maybe 62 00:03:26,902 --> 00:03:31,072 zoom out a little bit and go through the detail of what's been announced. 63 00:03:31,142 --> 00:03:34,682 I have to admit, I'm a little hazy on whether I know you think this is 64 00:03:34,682 --> 00:03:38,252 being removed even from the model comparison tool, and I was a bit 65 00:03:38,252 --> 00:03:42,602 less sure about that because I know it's being removed from the, the 66 00:03:42,602 --> 00:03:43,952 kind of active way you can use it. 67 00:03:43,952 --> 00:03:47,262 So from the default selection in GA4, so you're not going to 68 00:03:47,262 --> 00:03:48,282 be able to change the default. 69 00:03:48,282 --> 00:03:51,222 It's going to be data driven and if you don't meet the thresholds 70 00:03:51,222 --> 00:03:55,117 for data driven it will default back to last non-direct click. 71 00:03:55,177 --> 00:03:58,207 But it was always previously quite useful and this is leading me back 72 00:03:58,207 --> 00:03:59,857 to my kind of who cares question. 73 00:04:00,127 --> 00:04:04,087 And I know obviously some people really do care, but I think by and large, the 74 00:04:04,087 --> 00:04:10,462 majority of businesses using GA are probably still relying on last click, 75 00:04:10,552 --> 00:04:14,152 and at most if they've even done this, they might have gone into the model 76 00:04:14,152 --> 00:04:17,692 comparison tool just to have a look and compare the different models, 77 00:04:17,842 --> 00:04:21,262 which wasn't always that useful anyway because of the fact that you'd always 78 00:04:21,267 --> 00:04:25,592 get that weird thing where direct would look great in every other model. 79 00:04:25,598 --> 00:04:28,628 So it often shows you some slightly weird numbers anyway, but beyond that 80 00:04:28,628 --> 00:04:32,138 I'd love to know, and I mean, it'd be amazing to know this, how many GA 81 00:04:32,138 --> 00:04:36,956 accounts actually have a default model that isn't data driven or last click. 82 00:04:37,076 --> 00:04:39,536 If anyone listening knows that, please let us know. 83 00:04:40,316 --> 00:04:43,026 Daniel: Yeah, I'm inclined to agree and maybe we'll put a LinkedIn 84 00:04:43,031 --> 00:04:44,216 poll up or something like that. 85 00:04:44,216 --> 00:04:46,916 But I think the thing for me is that you're quite right that there's 86 00:04:46,916 --> 00:04:47,756 two ways of thinking about this. 87 00:04:47,761 --> 00:04:50,636 First of all, most people don't change the defaults, whatever they are, and 88 00:04:50,636 --> 00:04:54,836 Google Analytics has defaulted the default attribution model to data-driven 89 00:04:54,841 --> 00:04:59,546 attribution, which in itself, if you don't hit a certain threshold of data volumes 90 00:04:59,546 --> 00:05:01,116 defaults to last click attribution. 91 00:05:01,142 --> 00:05:05,182 So in a sense, we're using last click and data driven out of the box already. 92 00:05:05,232 --> 00:05:08,112 And this doesn't, from what the announcement reads, this is not going 93 00:05:08,112 --> 00:05:11,462 to affect the acquisition reports in the reports workspace in GA4. 94 00:05:11,462 --> 00:05:14,310 So your traffic acquisition and user acquisition reports, which are in a 95 00:05:14,315 --> 00:05:17,250 sense first click attribution and last click attribution, they're going to stay. 96 00:05:17,310 --> 00:05:21,266 What they are removing is the ability to use them as a default model for GA. 97 00:05:21,356 --> 00:05:23,681 So you can only use data driven, I assume they're just going to 98 00:05:23,681 --> 00:05:25,806 make it data driven and not even give you the choice of last click. 99 00:05:25,996 --> 00:05:29,656 And they're removing the opportunity to use it in the advertising workspace or 100 00:05:29,656 --> 00:05:30,946 the attribution reports they call it. 101 00:05:31,216 --> 00:05:35,086 So that is the model comparison reports and the other ones from what it says here. 102 00:05:35,086 --> 00:05:37,156 So it does read from the announcement, which we'll 103 00:05:37,156 --> 00:05:38,236 share a link in the show notes. 104 00:05:38,416 --> 00:05:41,266 It does read that they're removing the ability to use anything but 105 00:05:41,316 --> 00:05:46,236 last click or data driven in all reporting spaces in Google Analytics 4. 106 00:05:46,256 --> 00:05:50,366 The only way I can conceive of being able to get like a linear or a first 107 00:05:50,366 --> 00:05:54,476 click back or a time decay model is to be building it ourselves in a sense. 108 00:05:54,479 --> 00:05:57,009 Maybe there's some open source libraries or packages we can use, 109 00:05:57,009 --> 00:05:59,849 but using the BigQuery export and doing it ourselves in SQL. 110 00:06:00,069 --> 00:06:04,659 Dara: I guess I always think whenever I hear that, like that is obviously useful, 111 00:06:04,689 --> 00:06:08,329 but not being able to change it within GA is going to mean that you're going to 112 00:06:08,374 --> 00:06:10,834 have that issue, which exists, I guess already anyway, where you maybe have 113 00:06:10,834 --> 00:06:15,574 certain people within a business who are using modelled data in BigQuery and then 114 00:06:15,634 --> 00:06:17,284 maybe are feeding that into a dashboard. 115 00:06:17,284 --> 00:06:19,534 But you're probably still going to have some people who are going into 116 00:06:19,534 --> 00:06:23,074 the GA interface, and then you've got that age old question comes up of 117 00:06:23,074 --> 00:06:24,574 why did these numbers not match up? 118 00:06:24,678 --> 00:06:29,098 So just maybe this is, well, it is going to limit the amount you can 119 00:06:29,098 --> 00:06:33,268 do if you do create your own models outside of GA, unless you just 120 00:06:33,268 --> 00:06:36,288 bypass the interface completely, which I guess some people will. 121 00:06:36,378 --> 00:06:39,378 Daniel: Well yeah, but then I mean this poses the question of then, 122 00:06:39,378 --> 00:06:41,058 well, why use GA in the first place? 123 00:06:41,368 --> 00:06:45,893 I think this is another sort of like nail in the coffin in a sense of Google 124 00:06:45,893 --> 00:06:48,923 Analytics being a data product and it's more of an advertising product now. 125 00:06:48,953 --> 00:06:51,563 So Universal Analytics, I've always explained Universal Analytics 126 00:06:51,563 --> 00:06:54,893 was a data tool with a bolt on advertising marketing module, right? 127 00:06:54,893 --> 00:06:58,373 GA4 is a marketing product with a bolt on data module and I think the 128 00:06:58,373 --> 00:07:01,773 way that this is going is kind of reinforcing that idea for me at least. 129 00:07:02,037 --> 00:07:04,881 You know, that this is going to just focus on data-driven attribution. 130 00:07:04,881 --> 00:07:10,221 The only reason I can perceive of that is to better credit digital 131 00:07:10,281 --> 00:07:13,581 marketing channels, such as, for example, Google Ads, Display Video 132 00:07:13,581 --> 00:07:15,261 360 and Search Ads 360, right? 133 00:07:15,261 --> 00:07:16,341 So the Google marketing suite. 134 00:07:16,401 --> 00:07:18,691 There's no other reason, other than computation, it's not 135 00:07:18,696 --> 00:07:21,481 costing Google anything by having these models available, right? 136 00:07:21,481 --> 00:07:23,791 It's just a, you know, as a dropdown in a model comparison 137 00:07:23,791 --> 00:07:24,991 report, it's not doing any harm. 138 00:07:25,261 --> 00:07:30,416 So the removal of them is to, in a sense, forcibly encourage people not to use them 139 00:07:30,516 --> 00:07:34,356 to look at the data from a perspective of Google and the data-driven attribution 140 00:07:34,356 --> 00:07:36,996 model they don't share the source code, so we don't know exactly how it works. 141 00:07:36,996 --> 00:07:39,486 But we do know that it upweights Google advertising slightly more 142 00:07:39,486 --> 00:07:40,536 than other advertising, right? 143 00:07:40,596 --> 00:07:43,986 So in a sense, they're forcing us to use a model that upweights the Google 144 00:07:44,046 --> 00:07:47,841 Stack marketing or the Google marketing platform products, which I understand, 145 00:07:47,841 --> 00:07:49,401 I'm not bitter about it, I get it. 146 00:07:49,401 --> 00:07:52,221 If I was Google, I'd be doing the same thing with a lack of visibility 147 00:07:52,221 --> 00:07:55,641 across data with things like conversion modeling, data-driven attribution, machine 148 00:07:55,641 --> 00:07:59,181 learning modeling, behavioral modeling, you know, where consent's not provided 149 00:07:59,181 --> 00:08:02,301 through things like consent mode, it's all going into this big black box. 150 00:08:02,631 --> 00:08:05,653 And then machine learning happens, and a variety of different 151 00:08:05,653 --> 00:08:07,543 levels then out spits a result. 152 00:08:07,573 --> 00:08:10,123 And I think that's the nature and the reality of things nowadays. 153 00:08:10,123 --> 00:08:13,521 And I think this is just another one of those inevitable changes where, 154 00:08:13,581 --> 00:08:16,663 you know, acquisition reporting and attribution modeling is going to 155 00:08:16,663 --> 00:08:18,578 be all machine learning modeling. 156 00:08:18,638 --> 00:08:21,668 Again, you know, it's another layer of machine learning modeling that's 157 00:08:21,668 --> 00:08:23,588 obscuring the underlying data. 158 00:08:23,588 --> 00:08:25,718 And yes, we get access to the underlying data. 159 00:08:25,718 --> 00:08:26,808 Yes, we can use BigQuery. 160 00:08:26,808 --> 00:08:28,868 It's almost like a lazy answer to everything nowadays. 161 00:08:29,168 --> 00:08:31,538 Just use BigQuery, do it yourself, most people won't, most people don't know how 162 00:08:31,538 --> 00:08:36,098 to, but yeah, that's still not going to help because you know, Google does so much 163 00:08:36,428 --> 00:08:39,458 to embellish the data that we are never going to be able to replicate this stuff 164 00:08:39,458 --> 00:08:40,958 in BigQuery no matter how hard we try. 165 00:08:40,958 --> 00:08:43,928 There's lots of people trying really hard just to replicate standard reports 166 00:08:44,258 --> 00:08:48,428 in GA, and yet they're still struggling because the output, the data that Google 167 00:08:48,428 --> 00:08:52,118 outputs, the raw data to BigQuery is fundamentally different and misses a lot 168 00:08:52,118 --> 00:08:55,178 of the sort of nuance that Google does on top of the data, which they don't share. 169 00:08:55,328 --> 00:08:59,378 Dara: Yeah, and again, I would go out on a limb and suggest that the vast 170 00:08:59,378 --> 00:09:05,033 majority of users aren't going to be too effective by this, by this change. 171 00:09:05,033 --> 00:09:08,753 You do have your very advanced users who like to use all of the features 172 00:09:08,753 --> 00:09:12,683 and will have gone in and changed their default attribution model, created 173 00:09:12,683 --> 00:09:14,693 their own rule-based models, et cetera. 174 00:09:14,783 --> 00:09:17,423 But I think most people, that's not going to be the case. 175 00:09:17,534 --> 00:09:21,704 If I'm to be uncharacteristically less cynical, non cynical, you 176 00:09:21,704 --> 00:09:22,814 know, Google do have that data. 177 00:09:22,814 --> 00:09:26,919 So maybe, maybe they have looked and seen that, you know, not too many people 178 00:09:26,919 --> 00:09:28,629 are using, are changing the defaults. 179 00:09:28,629 --> 00:09:30,569 So they figure, well, let's get rid of that option. 180 00:09:30,659 --> 00:09:33,524 And the thing with the data driven is, I always think, it's the 181 00:09:33,524 --> 00:09:35,084 unknown that gets people, isn't it? 182 00:09:35,084 --> 00:09:39,704 Because in theory, at least, using the data-driven attribution modeling is 183 00:09:39,704 --> 00:09:44,559 going to be better than any abitary kind of rules based or semi data 184 00:09:44,559 --> 00:09:48,309 based kind of rules that you come up with based on your own understanding. 185 00:09:48,339 --> 00:09:50,239 It's always going to be limited to some extent. 186 00:09:50,489 --> 00:09:54,409 You're not going to be able to make it as customised, as kind of optimised and 187 00:09:54,414 --> 00:09:58,515 fine tuned as you could get with machine learning based attribution modeling. 188 00:09:58,605 --> 00:10:01,485 The problem is you don't know how it's working and you don't know if there is 189 00:10:01,485 --> 00:10:05,475 some intentional bias in that, where it is upweighting Google's channels. 190 00:10:05,539 --> 00:10:09,079 So it's probably the unknown that's going to get a lot of people, but something 191 00:10:09,079 --> 00:10:12,469 you said there as well, you know, it's like when something changes, people go 192 00:10:12,469 --> 00:10:16,489 up in arms and it's usually the, probably the minority people like us who will 193 00:10:16,489 --> 00:10:20,199 complain, but before long, to most people this will be the only way it's ever been. 194 00:10:20,259 --> 00:10:23,769 They'll take for granted that the numbers that Google Analytics says are their 195 00:10:23,769 --> 00:10:28,689 conversions will be correct, or to some, you know, correct in inverted comas, 196 00:10:28,779 --> 00:10:31,989 and they'll then maybe have a vague understanding that there's some machine 197 00:10:31,989 --> 00:10:33,909 learning going on behind the scenes. 198 00:10:33,909 --> 00:10:37,209 But you're right, this isn't something that really, if you want to continue 199 00:10:37,209 --> 00:10:40,399 to use the Google stack, then you don't really have a choice, you've 200 00:10:40,399 --> 00:10:44,229 just got to accept this, you either stick with good old trusty last click 201 00:10:44,229 --> 00:10:47,559 attribution, or you put your faith in Google's hands and think, okay, we'll 202 00:10:47,559 --> 00:10:51,309 use the data driven and hope that it's being done in a reasonably good way. 203 00:10:51,399 --> 00:10:54,514 Daniel: Well, for sure, I completely agree and the reason I think I feel strongly 204 00:10:54,514 --> 00:10:57,794 about this is because I've used them and I'm one of maybe the minority that 205 00:10:57,794 --> 00:11:01,001 have used these specifically mourning the parsing of linear attribution, 206 00:11:01,001 --> 00:11:04,031 which was my favorite because it's the my go-to attribution model. 207 00:11:04,036 --> 00:11:07,901 It's easy to, or the easiest one to explain outside of last click, and 208 00:11:07,901 --> 00:11:11,481 it's something that I used often to kind of demonstrate the inherent 209 00:11:11,486 --> 00:11:14,881 biases or the, the limits of last click attribution or first click attribution. 210 00:11:14,881 --> 00:11:18,091 So for me, I'm going to miss linear, I'm going to miss those other things. 211 00:11:18,091 --> 00:11:19,981 It doesn't mean I'm not still going to be able to do some 212 00:11:19,981 --> 00:11:21,151 of the stuff I can already do. 213 00:11:21,271 --> 00:11:23,871 And yes, of course we can go down the path of building ourselves. 214 00:11:23,905 --> 00:11:27,025 I think for me it's more like, and I think this is where people might 215 00:11:27,030 --> 00:11:30,906 feel a bit of an over inflated sense of outrage around this is like, even 216 00:11:30,906 --> 00:11:34,046 though I can't go to your party, I still want to be invited, right? 217 00:11:34,146 --> 00:11:37,589 I want to have the option of this thing there, knowing it's there, and 218 00:11:37,589 --> 00:11:41,009 then me deciding not to use it is different for it to be removed entirely. 219 00:11:41,069 --> 00:11:44,689 So I think that's the justification they've given for removing this, if I 220 00:11:44,689 --> 00:11:48,409 just read it out verbatim because it just doesn't sound satisfying enough basically. 221 00:11:48,679 --> 00:11:51,769 It says, these models don't provide the flexibility needed to adapt 222 00:11:51,769 --> 00:11:53,089 to evolving customer journeys. 223 00:11:53,149 --> 00:11:56,899 Data-driven attribution uses advanced AI to understand the impact each 224 00:11:56,904 --> 00:11:58,219 touchpoint has on our conversion. 225 00:11:58,309 --> 00:12:00,979 That's why we made data-driven attribution, the default attribution 226 00:12:00,984 --> 00:12:02,659 model in GA4 and Google Ads. 227 00:12:02,779 --> 00:12:05,569 For these reasons, first click, linear, time decay, and position 228 00:12:05,574 --> 00:12:08,568 based attribution across Google Analytics 4 will be going away. 229 00:12:08,598 --> 00:12:12,708 So in a sense, it's just saying times are changing data, driven's better, 230 00:12:12,738 --> 00:12:13,878 we're going to remove the old stuff. 231 00:12:13,933 --> 00:12:16,813 It doesn't feel like it's explained it, and I don't know if we've done 232 00:12:16,818 --> 00:12:19,748 a good enough job at the beginning of this conversation, at least Dara, 233 00:12:19,748 --> 00:12:22,548 just to say that this is a Google Ads thing as well as Google Analytics 4. 234 00:12:22,878 --> 00:12:26,658 This is not just a reporting thing, this is going to affect how 235 00:12:26,658 --> 00:12:30,138 people manage and optimise their kind of Google Ads campaigns too. 236 00:12:30,328 --> 00:12:34,798 It is a broader thing than Google are stopping to support these models, not 237 00:12:34,798 --> 00:12:38,128 just in Google Analytics 4, but across the kind of advertising ecosystem as a whole. 238 00:12:38,338 --> 00:12:39,118 Dara: I kind of like it. 239 00:12:39,448 --> 00:12:43,368 Okay it's a classically, kind of short, to the point, Google 240 00:12:43,398 --> 00:12:45,378 kind of help article explanation. 241 00:12:45,378 --> 00:12:48,378 You know, it doesn't go into, it's almost like the less detail you give, 242 00:12:48,378 --> 00:12:51,408 the less opportunity you have to catch your, you know, get yourself caught 243 00:12:51,413 --> 00:12:55,398 out or overexplain and give away some information you didn't want to. 244 00:12:55,668 --> 00:12:57,438 But this is kind of what I was saying, isn't it? 245 00:12:57,438 --> 00:13:02,273 It's like if in theory, using machine learning is going to come up with a 246 00:13:02,273 --> 00:13:06,353 better end result than you just going in yourself and thinking, I'll just 247 00:13:06,353 --> 00:13:10,883 create some relatively generic kind of rule-based attribution models. 248 00:13:10,913 --> 00:13:13,673 The problem again is that you can't tinker with that in any way. 249 00:13:13,673 --> 00:13:18,113 You get very limited visibility on that, but then that's probably the case with, 250 00:13:18,173 --> 00:13:19,853 that's going to be the case more and more. 251 00:13:19,853 --> 00:13:23,093 And you kind of hinted at this earlier, more and more of the day, and we've talked 252 00:13:23,093 --> 00:13:26,543 about this on this podcast before as well, more and more of the data is becoming 253 00:13:26,543 --> 00:13:28,968 modelled, that's not going to stop. 254 00:13:28,968 --> 00:13:30,588 Well, we haven't reached the end of that journey. 255 00:13:30,618 --> 00:13:33,618 The observed data has got gaps left, right, and center. 256 00:13:33,618 --> 00:13:35,658 So those gaps have to be plugged in some way. 257 00:13:35,663 --> 00:13:38,718 So maybe this is just another example of that and the fact that 258 00:13:38,723 --> 00:13:40,398 you can't see into that black box. 259 00:13:40,398 --> 00:13:42,138 I mean, what would you do if you could see into it? 260 00:13:43,218 --> 00:13:44,338 Daniel: Probably not a lot. 261 00:13:44,358 --> 00:13:44,958 Dara: Not a lot. 262 00:13:45,078 --> 00:13:46,638 Just look at it and say, yeah, that looks okay. 263 00:13:48,418 --> 00:13:50,058 Daniel: Yeah, we'll talk about it on this podcast, and that'll 264 00:13:50,058 --> 00:13:50,868 be an episode done right? 265 00:13:50,868 --> 00:13:53,173 They've opened the black box, they've shut it again. 266 00:13:53,623 --> 00:13:56,104 No, but I think just on back on something you just mentioned 267 00:13:56,104 --> 00:13:59,449 around this idea that these kind of algorithmic models are always going 268 00:13:59,449 --> 00:14:00,989 to be better than rules-based models. 269 00:14:00,989 --> 00:14:05,194 I've always found that a bit divisive because yes, a machine is going to 270 00:14:05,194 --> 00:14:08,999 be better at assigning rules with an unbiased, assuming there's no bias 271 00:14:08,999 --> 00:14:10,499 built in, but an unbiased nature. 272 00:14:10,499 --> 00:14:14,819 Like if you ask a paid search marketer what model to use versus a social marketer 273 00:14:14,819 --> 00:14:16,999 versus a display marketer, they're all going to pick a different one that 274 00:14:16,999 --> 00:14:18,324 makes their numbers look higher right? 275 00:14:18,324 --> 00:14:20,194 And so you kind of remove that aspect of it. 276 00:14:20,244 --> 00:14:22,524 The thing that I like about rules based models, and I think this 277 00:14:22,524 --> 00:14:25,434 is the thing that often gets overlooked, is that they are fixed. 278 00:14:25,434 --> 00:14:28,844 They are rules based, they are static in a sense that the rule that how 279 00:14:28,844 --> 00:14:31,274 we attribute email, for example, isn't going to change next year. 280 00:14:31,274 --> 00:14:35,324 So if I'm doing month or month or year on year reporting, the model of which 281 00:14:35,384 --> 00:14:37,184 I'm assessing value has not changed. 282 00:14:37,184 --> 00:14:39,524 And so I'm comparing like for like, apples to apple. 283 00:14:39,784 --> 00:14:43,324 The thing about an algorithmic attribution model is the value in the way that 284 00:14:43,324 --> 00:14:46,594 I'm rewarding email today is different to next week, which is different to 285 00:14:46,594 --> 00:14:47,914 next month and different to next year. 286 00:14:47,914 --> 00:14:51,664 So although we're still looking at a year on year or month to month report, 287 00:14:51,724 --> 00:14:55,614 the whole methodology of what we're looking at is kind of you know, apples 288 00:14:55,614 --> 00:14:58,854 to oranges, you know, in a sense it's an evolving thing and it's never static. 289 00:14:58,864 --> 00:15:01,284 I don't know the frequency of which they update their rules. 290 00:15:01,284 --> 00:15:05,019 Maybe weekly or at least it used to be in GA 360 and Universal Analytics. 291 00:15:05,469 --> 00:15:08,079 So I think this is the thing is there's no consistency anymore like 292 00:15:08,109 --> 00:15:11,067 in a sense, everything's always in flux, in change, evolving. 293 00:15:11,157 --> 00:15:13,977 And so data-driven attribution is another one of that where 294 00:15:14,272 --> 00:15:17,392 you might see that 50% of your conversions go to email this month. 295 00:15:17,482 --> 00:15:20,122 Next month it might be 10% of conversions go to email, but you've done 296 00:15:20,122 --> 00:15:21,592 exactly the same marketing activity. 297 00:15:21,592 --> 00:15:25,252 But the modeling behind the scenes has changed and used a different 298 00:15:25,322 --> 00:15:28,682 approach to email marketing, and it does beg another question of like, 299 00:15:28,682 --> 00:15:30,332 how quickly can it react to change? 300 00:15:30,337 --> 00:15:34,022 If I introduce a new marketing channel in today, how quickly before that 301 00:15:34,022 --> 00:15:37,592 starts to get attributed value, or how long until the machine learns to 302 00:15:37,662 --> 00:15:41,502 recognise it and understand the uplift or down lift that this channel makes. 303 00:15:41,682 --> 00:15:44,142 Same as if I remove a marketing channel from my marketing mix, like 304 00:15:44,142 --> 00:15:48,062 how long before it keeps trying to, you know, like reserve credit for a 305 00:15:48,072 --> 00:15:49,212 channel that doesn't exist anymore. 306 00:15:49,212 --> 00:15:52,542 And I think this is the stuff you never have to consider in a rules-based model. 307 00:15:52,602 --> 00:15:56,922 It's just a fixed way of approaching things that will never have these 308 00:15:56,922 --> 00:15:58,827 things thrown into consideration. 309 00:15:58,957 --> 00:16:02,322 Dara: Yeah, I was just thinking as I was listening to you, if it's totally 310 00:16:02,322 --> 00:16:04,702 opening the black box, it's not going to be useful because you're not 311 00:16:04,702 --> 00:16:05,772 going to know what you're looking at. 312 00:16:05,862 --> 00:16:09,912 But if it could combine it with some kind of like weekly or monthly insights 313 00:16:09,917 --> 00:16:14,587 report where it would tell you the reason why email is now getting half 314 00:16:14,587 --> 00:16:18,547 the attributed credit that it was before is because of X, Y, or Z. 315 00:16:18,787 --> 00:16:22,267 And then have a kind of percentage contribution of those different factors. 316 00:16:22,267 --> 00:16:26,557 So it might say, you know, you introduced a new channel and we think the likelihood 317 00:16:26,557 --> 00:16:31,587 that affected email's role in the conversion journey was 70% and then 318 00:16:31,587 --> 00:16:35,757 20% of it was because you changed the landing pages for your emails or whatever. 319 00:16:35,877 --> 00:16:37,137 That would be useful, wouldn't it? 320 00:16:37,137 --> 00:16:39,267 Because with what you're saying, I get it. 321 00:16:39,367 --> 00:16:43,147 It's like you won't necessarily know what it was that you, you could have 322 00:16:43,147 --> 00:16:46,372 changed so many different things because you would think the model wouldn't 323 00:16:46,807 --> 00:16:51,487 change how it's crediting a channel unless something does genuinely change. 324 00:16:51,487 --> 00:16:54,337 So even if your email activity is exactly the same, something 325 00:16:54,337 --> 00:16:55,447 else must have changed. 326 00:16:55,447 --> 00:16:58,307 Or the model, when it gets updated, it won't change how it's 327 00:16:58,307 --> 00:17:01,787 crediting email unless something else in that chain has changed. 328 00:17:01,967 --> 00:17:05,207 But you don't know what that is, so if it could kind of also give you 329 00:17:05,207 --> 00:17:09,707 a little report each time that the model updates and say, this is why 330 00:17:09,707 --> 00:17:12,742 you're seeing this, or this is the likelihood of why you're seeing this, 331 00:17:13,072 --> 00:17:15,697 then that would be useful, and then maybe people would be okay about it. 332 00:17:15,847 --> 00:17:18,277 But it's not going to tell you anything it's just going to give 333 00:17:18,277 --> 00:17:20,497 you, it's just going to output the numbers, and you're just going to 334 00:17:20,497 --> 00:17:22,657 have to trust that they're reliable. 335 00:17:22,897 --> 00:17:25,297 Daniel: Well yeah, trust is a huge part of this anyway, and I think if you're 336 00:17:25,297 --> 00:17:28,807 using, if you don't trust Google, then you won't be using Google Analytics anyway. 337 00:17:28,807 --> 00:17:33,182 So I think everyone, whether you like it or not, has an inherent trust placed in 338 00:17:33,182 --> 00:17:37,172 Google to track all your data and do all this modeling and sync it to Google Ads, 339 00:17:37,172 --> 00:17:38,492 or you're advertising through Google Ads. 340 00:17:38,492 --> 00:17:42,212 I bet someone's using some automated system there whether it's automatic 341 00:17:42,542 --> 00:17:46,412 bid optimisations, budget adjustments, performance max campaigns, you know, 342 00:17:46,412 --> 00:17:50,002 even looking at some of the audiences, the audience expansions or the kind of 343 00:17:50,002 --> 00:17:51,472 the new version of lookalike audiences. 344 00:17:51,472 --> 00:17:54,562 Like you've got to have a level of trust that Google knows what they're doing and 345 00:17:54,562 --> 00:17:55,762 they're targeting the right people right? 346 00:17:55,792 --> 00:17:57,082 So I think it's just another one of those. 347 00:17:57,082 --> 00:17:59,667 But just on the subject of Google Ads though. 348 00:17:59,667 --> 00:18:02,127 So we talked a lot about Google Analytics 4 and how, I suppose I've 349 00:18:02,127 --> 00:18:03,867 used it or we've used it in the past. 350 00:18:03,941 --> 00:18:07,181 Specifically thinking about the Google Ads side, and I think this is something 351 00:18:07,225 --> 00:18:10,304 that was announced about a couple of weeks ago and they say in the coming weeks, 352 00:18:10,304 --> 00:18:12,409 so I don't know exactly when this is. 353 00:18:12,469 --> 00:18:16,089 But Google Ads, anyone with a Google Ads account might've got this email recently, 354 00:18:16,089 --> 00:18:20,229 but they're now moving over to using the Google Analytics 4 attribution model. 355 00:18:20,289 --> 00:18:23,349 So, at the moment, if you select something like data-driven attribution 356 00:18:23,349 --> 00:18:27,789 or last click attribution in Google Ads, it's only going to do that attribution 357 00:18:27,789 --> 00:18:31,869 model on top of the data that Google Ads has, which is only Google Ads 358 00:18:31,989 --> 00:18:33,339 click and impression data, right? 359 00:18:33,399 --> 00:18:36,399 What this update says is that they're moving over to using the 360 00:18:36,399 --> 00:18:40,914 data-driven attribution conversion values from Google Analytics Four. 361 00:18:41,124 --> 00:18:44,171 So this is in a sense, it's not, I don't think just a coincidental 362 00:18:44,171 --> 00:18:45,461 timing of two features releasing. 363 00:18:45,461 --> 00:18:48,011 I think it's all moving towards one thing, which is we are going to make 364 00:18:48,011 --> 00:18:49,481 data-driven attribution the default. 365 00:18:50,051 --> 00:18:53,306 And when you use your conversions from Google Analytics 4 in Google 366 00:18:53,326 --> 00:18:58,186 Ads, we are going to output or export for Google Ads to use the data-driven 367 00:18:58,191 --> 00:19:00,136 attribution credit for Google Ads. 368 00:19:00,196 --> 00:19:03,646 So it's no longer just going to rely on, you know, the last click data from 369 00:19:03,651 --> 00:19:06,616 Google Analytics, which a lot of people didn't use anyway in Google Ads because 370 00:19:06,616 --> 00:19:08,146 it was never as good as the pixel. 371 00:19:08,206 --> 00:19:11,266 I think if we look at all of these things kind of in total, holistically, 372 00:19:11,416 --> 00:19:14,926 Google Ads is moving over to using Google Analytics for conversions. 373 00:19:14,931 --> 00:19:18,086 I think all signs are pointing towards that, I would even put money that 374 00:19:18,146 --> 00:19:21,127 on in the next year to two years that we are going to see a switch 375 00:19:21,127 --> 00:19:25,207 off of the Google Ads pixels and the floodlight tags as well, just because 376 00:19:25,207 --> 00:19:28,597 like everything's moving over to GA4 and the modeling that it's doing 377 00:19:28,597 --> 00:19:30,757 there, the data-driven distribution modeling that it's doing there. 378 00:19:30,817 --> 00:19:33,007 So if everything's moving that way, if we take that on faith 379 00:19:33,007 --> 00:19:34,267 that, you know, that might happen. 380 00:19:34,737 --> 00:19:37,047 If they're moving over to using data-driven attribution as the 381 00:19:37,047 --> 00:19:40,407 default model in Google Analytics 4 and data-driven attribution credit 382 00:19:40,407 --> 00:19:43,317 is going to be exported to Google Ads for optimisation, there's a lot of 383 00:19:43,317 --> 00:19:44,487 things happening at the same time. 384 00:19:44,492 --> 00:19:47,157 But from a marketing perspective, what it means is that you're going 385 00:19:47,157 --> 00:19:51,327 to be using different data to credit your campaigns and which feeds the 386 00:19:51,327 --> 00:19:54,327 model over there, the optimisation and the, you know, the advertising and 387 00:19:54,332 --> 00:19:55,707 the, the bid adjustment models too. 388 00:19:55,707 --> 00:19:59,917 So the kind of ROI, the CPA, you know, all of that kind of lovely data's 389 00:19:59,917 --> 00:20:03,397 going to be changing because you're going to be using different underlying 390 00:20:03,547 --> 00:20:05,497 data alongside all this other changes. 391 00:20:05,547 --> 00:20:08,847 It's all data driven, of course, but you're now using the data from Google Ads. 392 00:20:09,057 --> 00:20:10,887 All of this is happening in May, by the way. 393 00:20:11,127 --> 00:20:13,437 So at the time we're recording, it's the beginning of April. 394 00:20:13,497 --> 00:20:16,107 And so by May this is going to be for all new accounts set 395 00:20:16,112 --> 00:20:17,157 up, it's going to be this way. 396 00:20:17,457 --> 00:20:19,807 And then eventually they're going to switch off these old attribution 397 00:20:19,807 --> 00:20:21,607 models in both platforms by September. 398 00:20:21,637 --> 00:20:24,794 So in terms of the timeline, it's pretty quick, I don't think that they are. 399 00:20:24,805 --> 00:20:28,155 Well, I think that they are very heavily related features, you know? 400 00:20:28,375 --> 00:20:32,034 Dara: I think you're right, and that's a theory you've, I can't remember if 401 00:20:32,034 --> 00:20:34,854 you've mentioned it on the podcast before, but it's certainly something 402 00:20:34,974 --> 00:20:36,834 you've said to me in real life. 403 00:20:36,924 --> 00:20:40,527 And I wouldn't argue with you, it seems like they're simplifying and they are 404 00:20:40,687 --> 00:20:45,707 kind of converging the different products and the focus will be on GA4 which does 405 00:20:45,707 --> 00:20:49,667 make, it does make complete sense and goes back again to the point about if 406 00:20:49,667 --> 00:20:51,857 you are tied in, shouldn't say tied in. 407 00:20:51,857 --> 00:20:55,277 If you're using the Google stack, then you're going to accept this, you're 408 00:20:55,277 --> 00:20:58,877 going to understand that, you know, Google is going to make these decisions 409 00:20:58,877 --> 00:21:02,337 that are you know, at times potentially favour them, but why wouldn't they? 410 00:21:02,337 --> 00:21:04,197 It's their tech at the end of the day. 411 00:21:04,267 --> 00:21:07,292 You mentioned people need to have a bit of trust, I think they don't 412 00:21:07,322 --> 00:21:09,242 even need to have that much trust. 413 00:21:09,242 --> 00:21:11,552 They need to have some trust, obviously, but you don't have a 414 00:21:11,552 --> 00:21:15,437 choice if you're using Google, and that's what you've always used then 415 00:21:15,437 --> 00:21:19,217 the cost to move away, not just financial costs, but of retraining 416 00:21:19,217 --> 00:21:23,957 people and learning new systems is so high that people will just accept it. 417 00:21:23,957 --> 00:21:26,187 It's a new change, they'll accept it and they'll move on. 418 00:21:26,377 --> 00:21:28,862 Daniel: This kind of harks back to what we were saying about how you can just 419 00:21:28,862 --> 00:21:31,802 use the lazy excuse of just saying, yeah, just do it yourself in BigQuery, right? 420 00:21:31,892 --> 00:21:33,962 Another option there is just use a different product. 421 00:21:34,502 --> 00:21:37,802 Like when they announced the switch off of Universal Analytics and every 422 00:21:37,802 --> 00:21:41,522 analytics vendor came out of the woodwork saying, you know, Google Analytics isn't 423 00:21:41,522 --> 00:21:44,522 dead over here, we've replicated the dashboards, we've got the thing over here. 424 00:21:44,622 --> 00:21:47,442 I think this is going to be another, a bit of catnip for those other analytics 425 00:21:47,447 --> 00:21:50,202 vendors that are going to be like, we've still got all the attribution 426 00:21:50,202 --> 00:21:51,742 models, we're still doing attribution. 427 00:21:51,755 --> 00:21:55,049 Which I, again, I don't think any of these things are real deterrence for the 428 00:21:55,049 --> 00:21:58,819 customers that use Google because it's all going to be in the service of the Google 429 00:21:58,884 --> 00:22:00,594 ecosystem, which is really huge, right? 430 00:22:00,594 --> 00:22:01,634 I mean, that's the reality. 431 00:22:01,634 --> 00:22:04,494 And obviously it's free, but nothing's truly free, like what 432 00:22:04,494 --> 00:22:07,034 you were saying is there's going to be compromise or cost in some way. 433 00:22:07,034 --> 00:22:11,154 Either you pay for the product and it is, you know, in a sense non Google and 434 00:22:11,154 --> 00:22:14,334 it's yours or you get a free product like Google Analytics, but you pay for 435 00:22:14,334 --> 00:22:17,719 it in a sense of, you know, a different way in terms of data or using their 436 00:22:17,959 --> 00:22:21,469 tailored models that maybe Google weighted and, and things like this. 437 00:22:21,469 --> 00:22:23,649 I mean, you've always got choice and I would always say 438 00:22:23,728 --> 00:22:24,958 assess all options all the time. 439 00:22:25,018 --> 00:22:26,488 You know, especially if you're making a change. 440 00:22:26,488 --> 00:22:29,458 You know, assess all options, including not Google Analytics but I think the 441 00:22:29,463 --> 00:22:33,698 other reality, and I think this is just another fact of the digital analytics 442 00:22:33,728 --> 00:22:36,443 marketing life is that everyone will end up using Google Analytics in some 443 00:22:36,443 --> 00:22:40,553 way anyway, because they'll run ads through Google or the Google ecosystem. 444 00:22:40,603 --> 00:22:42,973 Just another bit on the whole convergence of all these products 445 00:22:42,973 --> 00:22:45,523 is that, you know, remember recently they moved everything over to the 446 00:22:45,523 --> 00:22:47,233 Google tag, the one Google tag. 447 00:22:47,293 --> 00:22:50,753 And so now you put one tag on your website that does all floodlights, 448 00:22:50,773 --> 00:22:54,193 Google Analytics, Google Ads tags, and so you don't need to have three 449 00:22:54,193 --> 00:22:56,743 separate pieces of code anymore, it's one tag that does it all. 450 00:22:56,743 --> 00:23:01,123 So in a sense, now when you implement Google Analytics or Google Ads, you 451 00:23:01,123 --> 00:23:03,693 are implementing the same products, you're implementing the Google Tag. 452 00:23:04,123 --> 00:23:06,823 And actually all of that technology is built on Tag Manager anyways. 453 00:23:06,933 --> 00:23:09,650 The point is, is that there is no difference in the implementation 454 00:23:09,655 --> 00:23:12,020 now between Google Ads, Google Analytics, and floodlights. 455 00:23:12,260 --> 00:23:15,080 And I think what they're going to do now that everyone's implementing the same 456 00:23:15,400 --> 00:23:17,290 literal code on their website or app. 457 00:23:17,340 --> 00:23:19,090 They're just going to move the behind the scenes stuff and 458 00:23:19,090 --> 00:23:20,170 just move it over to one place. 459 00:23:20,170 --> 00:23:22,480 So I think, you know, with all this stuff, I think, you know, we know what 460 00:23:22,480 --> 00:23:25,090 we're getting into, we know who we're buying from here in a sense, right? 461 00:23:25,090 --> 00:23:28,330 We know it's Google, we know Google have got a vested interest in making 462 00:23:28,330 --> 00:23:31,330 money, and that's through advertising so, you know, you can't blame them. 463 00:23:31,490 --> 00:23:35,420 Dara: Again, I'm feeling very charitable today, far less cynical than usual. 464 00:23:35,450 --> 00:23:38,930 But you know, there's some benefits I guess, if they tighten up the integrations 465 00:23:38,980 --> 00:23:43,330 between the two products, between Google Ads and GA4, then it, you know, at 466 00:23:43,330 --> 00:23:46,780 least in theory, it should reduce some of the discrepancies that you see when 467 00:23:46,780 --> 00:23:50,220 you're comparing numbers between the two, it's less implementation as well. 468 00:23:50,309 --> 00:23:53,429 So there are some benefits you know, it's cleaner, simpler implementation. 469 00:23:53,609 --> 00:23:56,669 This should be tighter matching of numbers between the two. 470 00:23:56,669 --> 00:23:58,649 Obviously you'll still have differences between, you know, 471 00:23:59,129 --> 00:24:03,059 expected differences between metrics like clicks versus sessions. 472 00:24:03,249 --> 00:24:07,919 There are some benefits to the, to this as well and yeah, nothing comes free does it? 473 00:24:07,919 --> 00:24:12,199 So you have to accept as a result of that, that it could become a little 474 00:24:12,199 --> 00:24:15,369 bit more opaque or you'll have to accept that some of the numbers may 475 00:24:15,369 --> 00:24:18,349 have some bias in them potentially, but when has that ever not been the case. 476 00:24:18,929 --> 00:24:21,389 Daniel: To be more cynical if you are not going to be Dara. 477 00:24:21,449 --> 00:24:26,062 I would say that this is one step closer to the idea that you spend your 478 00:24:26,062 --> 00:24:27,982 money through Google to advertise. 479 00:24:28,072 --> 00:24:32,047 It's got an audience, which you can't see, it then tells you how well it's doing 480 00:24:32,047 --> 00:24:35,677 through models that we can't access and reports on its own performance in a way 481 00:24:35,677 --> 00:24:37,207 that we can't tell if it's correct or not. 482 00:24:37,267 --> 00:24:40,927 So I think this is a, a classic sense, if I use the term of just 483 00:24:40,927 --> 00:24:42,037 them marking their own homework. 484 00:24:42,127 --> 00:24:45,361 So I think this is, you know, to be the cynic here, it's one step closer 485 00:24:45,361 --> 00:24:48,331 towards being unable to validate. 486 00:24:48,726 --> 00:24:52,146 Not only, you know, things that are already removed, like keywords from SEO 487 00:24:52,356 --> 00:24:56,286 or you know, actual audience, people in audiences and things like third 488 00:24:56,286 --> 00:24:59,676 party cookies going away and audience matching and lookalikes all disappearing. 489 00:24:59,706 --> 00:25:03,336 And then you've got machine learning trusting in Google's idea of machine 490 00:25:03,336 --> 00:25:06,816 learning and use of it that they are spending your money, that people are 491 00:25:06,816 --> 00:25:09,126 really clicking on it, that they're reaching the people that they say they're 492 00:25:09,126 --> 00:25:12,716 reaching, and then we are saying, and then tell me how well they went, how much 493 00:25:12,716 --> 00:25:14,216 money did we make from that campaign? 494 00:25:14,286 --> 00:25:17,628 All from a perspective of the company that's serving the ad is doing the 495 00:25:17,628 --> 00:25:22,893 full thing, which, you know, as a independent analytics agency that we 496 00:25:22,893 --> 00:25:26,013 are at Dara, you know, that's one of our kind of bread and butter statements 497 00:25:26,013 --> 00:25:29,643 is when we go work with clients, it's like, well, your agency might do that. 498 00:25:29,643 --> 00:25:32,433 Your agency might be running your campaigns and then marking their own 499 00:25:32,433 --> 00:25:36,483 homework by using sort of attribution models or data points or ways of viewing 500 00:25:36,488 --> 00:25:40,073 data that make it look more profitable or valuable than maybe we could provide 501 00:25:40,073 --> 00:25:42,113 as an unbiased, agnostic perspective. 502 00:25:42,113 --> 00:25:45,212 But in a sense Google's closing the full loop, they're closing that loop 503 00:25:45,212 --> 00:25:48,962 so that you know it's going to be very hard for other people to be able to 504 00:25:48,962 --> 00:25:53,017 be sort of a, an objective or kind of third party in that whole system right? 505 00:25:53,257 --> 00:25:55,701 Dara: Yeah, and you're bringing my cynicism back out again, 506 00:25:55,776 --> 00:25:56,796 which I'm quite happy about. 507 00:25:57,094 --> 00:26:00,624 In a way it's amazing how long we got away with being able to look 508 00:26:00,624 --> 00:26:06,264 at a cleaner picture of attribution within GA because one of the benefits, 509 00:26:06,454 --> 00:26:07,439 you've probably done this too. 510 00:26:07,439 --> 00:26:12,079 You would push somebody to use GA data rather than Google Ads data because of 511 00:26:12,079 --> 00:26:15,529 the fact that GA included all the other channels and treated them equally with 512 00:26:15,529 --> 00:26:16,969 the exception of direct, obviously. 513 00:26:17,059 --> 00:26:20,509 So you could say, well, look, in GA you get to have a, a clear view and 514 00:26:20,509 --> 00:26:24,204 whether you like last click attribution or not, all channels are being treated 515 00:26:24,204 --> 00:26:25,884 the same with the exception of direct. 516 00:26:26,064 --> 00:26:30,204 So it was kind of surprising in a way that you could do that in a platform that was 517 00:26:30,204 --> 00:26:35,304 basically provided mostly for free as a result of people spending on Google Ads. 518 00:26:35,454 --> 00:26:39,144 And now, finally, now that they have the opportunity, I guess, to do it with maybe, 519 00:26:39,414 --> 00:26:42,424 maybe they've calculated this is the point in time where it's going to cost the 520 00:26:42,444 --> 00:26:45,714 least amount of controversy to do this. 521 00:26:45,714 --> 00:26:47,754 And to actually say, look, we're getting rid of these. 522 00:26:47,874 --> 00:26:50,314 You're going to have data driven, last click, that's it. 523 00:26:50,314 --> 00:26:53,754 Daniel: It's going to be interesting to see where the product of Google 524 00:26:53,814 --> 00:26:58,134 Analytics goes, sort of post this and what the next change like this will be. 525 00:26:58,134 --> 00:27:01,464 What's the next step towards this kind of unity of all of their product 526 00:27:01,464 --> 00:27:03,264 suite or their advertising suite here? 527 00:27:03,384 --> 00:27:04,104 Dara: One interface. 528 00:27:04,224 --> 00:27:05,704 Daniel: One interface, yeah god. 529 00:27:05,704 --> 00:27:09,714 The thing that we call keep talking about on this is just that they really 530 00:27:09,739 --> 00:27:11,329 want to rock the boat of Google Ads. 531 00:27:11,359 --> 00:27:14,059 Because Google Ads is like 90% of alphabet's revenue. 532 00:27:14,059 --> 00:27:16,819 It's like over a hundred million or billion a year or some, 533 00:27:16,909 --> 00:27:18,349 something like crazy like that. 534 00:27:18,429 --> 00:27:22,165 So it's like, do they risk even rocking the boat gently and some 535 00:27:22,165 --> 00:27:23,395 of that spilling over the edge? 536 00:27:23,395 --> 00:27:26,877 And I think maybe, maybe they do, but I think it's you know, when you read 537 00:27:26,877 --> 00:27:30,717 stories like, Facebook, Meta lost 10 billion last year because of a change 538 00:27:30,717 --> 00:27:34,442 iOS rolled out and it's like, well, if Google had that, it's going to be a lot 539 00:27:34,467 --> 00:27:38,842 bigger than 10 billion because they're a bigger advertising company and so all of a 540 00:27:38,842 --> 00:27:42,332 sudden it's like, if that's the evolution of marketing is kind of restriction 541 00:27:42,332 --> 00:27:45,272 and lockdown of like tracking and third party pixels and things like that. 542 00:27:45,332 --> 00:27:47,972 If Google's going all in on machine learning and AI to solve these 543 00:27:47,977 --> 00:27:50,582 gaps, then why the hell would they not move everything over to it? 544 00:27:50,582 --> 00:27:53,872 Because some of their sort of material collateral is, they've quote, unquote, 545 00:27:53,872 --> 00:27:58,307 solved the privacy issue in Google Analytics 4, so why would they not 546 00:27:58,307 --> 00:28:02,576 use that solved version of the truth in all their advertising stack, and 547 00:28:02,876 --> 00:28:04,136 they get a bit more control there. 548 00:28:04,216 --> 00:28:07,936 Dara: Is there an exploration technique you can use to look at 549 00:28:07,936 --> 00:28:11,296 different touchpoints in a conversion journey, or is that information just 550 00:28:11,301 --> 00:28:12,856 not going to be available anymore? 551 00:28:12,856 --> 00:28:15,391 Unless you use the raw export to BigQuery. 552 00:28:15,631 --> 00:28:17,581 Daniel: So there's no exploration technique as such. 553 00:28:17,631 --> 00:28:20,721 I suppose you could look at funnels and things like that, or you could 554 00:28:20,721 --> 00:28:23,361 look at the user explorer, which gives you a bit of a, like a CRM 555 00:28:23,721 --> 00:28:25,564 record which is a bit crap to be fair. 556 00:28:25,564 --> 00:28:28,506 But there is the conversion partners report in the advertising workspace 557 00:28:28,511 --> 00:28:31,926 that does give you, that gives you every interaction bar direct, of course, 558 00:28:31,926 --> 00:28:33,306 leading up to a point of conversion. 559 00:28:33,396 --> 00:28:36,961 So I think you've still got that data, you've still got the visibility into 560 00:28:36,961 --> 00:28:39,758 that, but it's still not going to be, it depends what you want to do with it, 561 00:28:39,758 --> 00:28:41,451 because it's quite a micro level report. 562 00:28:41,451 --> 00:28:44,151 Like you're going to look at every unique journey and quantify them. 563 00:28:44,151 --> 00:28:47,031 It's not going to, I don't know if it's going to give you what you need. 564 00:28:47,031 --> 00:28:49,851 Dara: Well that's the one actually, that's what I was thinking of. 565 00:28:50,241 --> 00:28:55,761 In the past at least that was useful if you wanted to see how often a channel 566 00:28:55,761 --> 00:28:59,431 played a part if you didn't really care too much about what the part was. 567 00:28:59,731 --> 00:29:02,891 How many conversions has this channel contributed in some way too. 568 00:29:03,071 --> 00:29:06,751 That was useful, I assumed wrongly, thankfully, that 569 00:29:06,751 --> 00:29:08,011 was getting removed as well. 570 00:29:08,161 --> 00:29:10,861 Daniel: No, so all of those reports will stay, but the attribution 571 00:29:10,861 --> 00:29:13,981 model that it's using is going to be removed except for data driven. 572 00:29:13,986 --> 00:29:14,996 And I think this is the key part. 573 00:29:15,026 --> 00:29:17,426 The one thing I would say on that, and I know exactly what you mean 574 00:29:17,616 --> 00:29:20,076 because we are looking at the contribution attribution, right? 575 00:29:20,076 --> 00:29:23,406 It's the contributed value of like total journeys it's kind of existed in. 576 00:29:23,496 --> 00:29:25,926 The one thing that GA4 doesn't allow you to do, in which Universal 577 00:29:25,931 --> 00:29:28,986 did, which I find really annoying, is that you can't do a filter. 578 00:29:28,986 --> 00:29:32,946 You can't search that to just show you all journeys with a specific channel, and 579 00:29:32,976 --> 00:29:35,136 you can't just search and get the number. 580 00:29:35,166 --> 00:29:36,306 It's so annoying, it's so frustrating. 581 00:29:36,466 --> 00:29:38,386 Dara: Back to my first question. 582 00:29:38,456 --> 00:29:39,686 How big of a deal is this? 583 00:29:39,746 --> 00:29:41,246 How much should people care? 584 00:29:41,246 --> 00:29:42,536 And how much do people care? 585 00:29:42,536 --> 00:29:44,356 Have you, are you in the thick of it? 586 00:29:44,356 --> 00:29:48,496 Are you part of any groups of rebels complaining online about this? 587 00:29:48,496 --> 00:29:52,476 Or are you just thinking, you know what, this is fine, it'll happen not a big deal. 588 00:29:52,636 --> 00:29:54,826 Daniel: I'm in the camp that it is just going to, it's inevitable. 589 00:29:54,826 --> 00:29:56,806 We don't have a say, this is not a democracy right. 590 00:29:56,806 --> 00:29:59,356 Google has decided and told us that something's happening. 591 00:29:59,356 --> 00:30:02,015 I'm in the camp of like, whatever, sure, another change. 592 00:30:02,028 --> 00:30:04,688 I have to update some training content and collateral that I've made, you know, 593 00:30:04,693 --> 00:30:08,018 that kind of stuff that I'm a little bit peeved about, but it's just inevitable 594 00:30:08,085 --> 00:30:09,136 that things like that will happen. 595 00:30:09,374 --> 00:30:12,284 I think it's been a bit of an equal split, some people welcome the change, some 596 00:30:12,284 --> 00:30:15,804 people are against the change, but a lot of the circles that I'm reading through 597 00:30:15,804 --> 00:30:17,439 or listening on is analytic circles. 598 00:30:17,439 --> 00:30:20,449 And I think what's going to be interesting for me is going to be the marketing 599 00:30:20,449 --> 00:30:23,449 circles and wondering the impact that this is going to have alongside that other 600 00:30:23,449 --> 00:30:26,359 change of using the GA data in Google Ads. 601 00:30:26,359 --> 00:30:30,329 So I think it's a bit of a split bag at the moment, you know, there's I would 602 00:30:30,329 --> 00:30:32,639 say 50/50 just to be, you know, safe. 603 00:30:32,639 --> 00:30:34,829 I'm just going to say it's a 50/50 split in terms of sentiment. 604 00:30:34,879 --> 00:30:39,094 I will miss it, but I also appreciate, you know, or understand it a little bit. 605 00:30:39,094 --> 00:30:41,914 You know, just say like, you know, it is what it is, it's happening regardless. 606 00:30:42,034 --> 00:30:44,944 But like I said, I'm going to be interested to see where the marketers, 607 00:30:44,944 --> 00:30:48,477 the search marketers, the display market, anyone using Google ads, I'd 608 00:30:48,477 --> 00:30:49,557 love to get their opinion on this. 609 00:30:49,717 --> 00:30:52,517 Dara: That's a really fair point actually because I guess mainly why 610 00:30:52,517 --> 00:30:55,847 I'm kind of asking that question in a bit of a flippant way because 611 00:30:55,967 --> 00:30:59,987 it's maybe something that I don't always need to worry too much about. 612 00:31:00,047 --> 00:31:05,327 And often the data coming from GA has been the standard last 613 00:31:05,357 --> 00:31:07,127 non click attribution model. 614 00:31:07,577 --> 00:31:12,922 But the change affecting ads is bigger because what's going to happen 615 00:31:12,922 --> 00:31:16,882 to your performance if you, so you either make the change in advance, 616 00:31:16,972 --> 00:31:18,162 you accept that it's going to change. 617 00:31:18,162 --> 00:31:18,832 So what did you say? 618 00:31:18,832 --> 00:31:22,042 You said from May I think new properties, and I guess that 619 00:31:22,042 --> 00:31:25,852 would apply to new ads accounts, wouldn't let you change the default. 620 00:31:25,942 --> 00:31:29,347 Then from September, all accounts will lose the option to select 621 00:31:29,347 --> 00:31:30,417 any of those other models. 622 00:31:30,417 --> 00:31:33,897 So if you've been bidding based on an attribution model that's going 623 00:31:33,897 --> 00:31:37,287 to get taken away, that could mess up your whole account temporarily 624 00:31:37,287 --> 00:31:38,907 while it all readjusts itself. 625 00:31:39,207 --> 00:31:40,497 That would be a reason to care? 626 00:31:40,647 --> 00:31:43,737 Daniel: Well yeah, if you are currently got a marketing or bidding strategy based 627 00:31:43,737 --> 00:31:47,487 on something that's not data driven, then I would, I would be kind of, you 628 00:31:47,487 --> 00:31:51,387 know, there'll be a sweat drop running down my head and just thinking about how 629 00:31:51,387 --> 00:31:55,107 much work have I got to do to rebalance this portfolio of ads, and especially 630 00:31:55,157 --> 00:31:56,667 on some of these bigger ads accounts. 631 00:31:56,706 --> 00:31:59,586 So I would be moving down that way, but like I say, combining this with 632 00:31:59,586 --> 00:32:02,791 the other change I think maybe you could do two birds, one stone and not 633 00:32:02,791 --> 00:32:04,071 have to redo the whole thing twice. 634 00:32:04,071 --> 00:32:07,821 But the move over to using the kind of data-driven export from Google 635 00:32:07,821 --> 00:32:11,011 Analytics, not the data-driven attribution from Google Ads, I think 636 00:32:11,011 --> 00:32:11,871 that's going to be the next step. 637 00:32:11,926 --> 00:32:14,636 I will tell anyone right now, it doesn't matter who you are, if you are not 638 00:32:14,636 --> 00:32:18,146 exporting your GA4 conversions into Google Ads, you need to start right now. 639 00:32:18,236 --> 00:32:21,326 Even if they're secondary conversions purely to run in parallel 640 00:32:21,686 --> 00:32:24,536 alongside your Google Ads pixel, you know, while it's still there 641 00:32:24,541 --> 00:32:25,526 you know, all that kind of stuff. 642 00:32:25,526 --> 00:32:27,415 But I would always in a sense, be doubling up. 643 00:32:27,459 --> 00:32:30,899 I'll be doing the Google Analytics export to Google Ads for conversions, 644 00:32:30,899 --> 00:32:34,279 and I might even be using that already, but I'd at least have it ready to 645 00:32:34,279 --> 00:32:37,279 go so that it's not a shock or a surprise to you, so you're not kind 646 00:32:37,279 --> 00:32:38,868 of like worried about that change. 647 00:32:39,248 --> 00:32:43,513 So yeah, maybe do two at once but I would be, I mean, for me, it removes a tool 648 00:32:43,513 --> 00:32:48,363 in my tool belt when I'm doing marketing analytics reporting, budget adjustment, 649 00:32:48,393 --> 00:32:52,023 you know, kind of end of campaign analysis, reporting, things like that. 650 00:32:52,023 --> 00:32:54,903 I'm not spending tens or hundreds of thousands of dollars a month on 651 00:32:54,903 --> 00:32:57,843 someone's account that's now about to potentially have to change. 652 00:32:57,848 --> 00:33:00,690 You know, I don't have that level of dependency on these right now. 653 00:33:00,760 --> 00:33:04,270 Dara: And there's a, a, an open call to our listeners, if anybody out there 654 00:33:04,330 --> 00:33:09,790 is working in the kind of Google Ad space and is seeing repercussions of 655 00:33:09,790 --> 00:33:13,450 this that maybe we're not paying enough attention to, then let us know and 656 00:33:13,540 --> 00:33:16,630 if you want to come on and chat to us about it, then you can do that as well. 657 00:33:16,720 --> 00:33:20,080 Okay, I think that's all we can really say about this for now, Dan. 658 00:33:20,150 --> 00:33:23,585 We've given our take on it and given people some things to think about, but 659 00:33:23,585 --> 00:33:27,603 I guess we just kind of like with a lot of these changes we continue to kind of 660 00:33:27,603 --> 00:33:29,553 watch and see what the fallout is like. 661 00:33:29,733 --> 00:33:31,803 So it's been a while, I think I mentioned this at the start. 662 00:33:31,803 --> 00:33:34,473 It's been a while since it's been just you and I so we've been off the 663 00:33:34,473 --> 00:33:37,113 hook for our wind downs for a while. 664 00:33:37,113 --> 00:33:40,593 So that should have given you plenty of time to do something interesting. 665 00:33:40,713 --> 00:33:45,003 So what have you been up to, to kind of switch off from work lately? 666 00:33:45,413 --> 00:33:48,991 Daniel: Well, I tell you what Dara, I bought a steam deck not too long ago. 667 00:33:49,081 --> 00:33:51,571 So this is a handheld gaming pc. 668 00:33:51,571 --> 00:33:53,641 And I think I may have mentioned this before, but I've been playing 669 00:33:53,641 --> 00:33:57,391 this lovely game called Sable, and it's made by two people in London, 670 00:33:57,412 --> 00:33:59,122 so it's just a two person outfit. 671 00:33:59,722 --> 00:34:04,100 And they made this beautiful sort of narrative adventure puzzle game, 672 00:34:04,350 --> 00:34:05,680 and I can't recommend it enough. 673 00:34:05,680 --> 00:34:10,000 It's a non-combat game, and it's about solving puzzles and exploring 674 00:34:10,000 --> 00:34:13,630 and uncovering this narrative around why you're there and collecting 675 00:34:13,630 --> 00:34:14,830 things and doing other things. 676 00:34:14,835 --> 00:34:19,069 It's very stylized, very artistic, and I give it a double thumbs 677 00:34:19,069 --> 00:34:20,532 up, Dan's double thumbs up. 678 00:34:20,542 --> 00:34:23,324 So if anyone's out there thinking of something to play next, something that's 679 00:34:23,369 --> 00:34:28,350 chill, relaxing, and maybe 10 to 20 hours long, which is a real big positive, rather 680 00:34:28,355 --> 00:34:31,787 than these 100 to 200 hour epics that I like to play to then check out Sable. 681 00:34:32,057 --> 00:34:35,237 Dara: My mind, even though you told me about the steam deck when you said it, 682 00:34:35,237 --> 00:34:40,417 I thought about like some kind of steam based contraption, like for pressing 683 00:34:40,437 --> 00:34:42,807 your trousers or something like that. 684 00:34:43,167 --> 00:34:46,567 Like an old school iron type, you know, are the ones they have in dry cleaners. 685 00:34:46,567 --> 00:34:47,457 Daniel: Like a trouser press. 686 00:34:47,517 --> 00:34:49,347 Dara: Yeah exactly, a trouser press yeah. 687 00:34:49,347 --> 00:34:53,457 So it just goes to kind of, well reiterate how little I know about gaming. 688 00:34:53,637 --> 00:34:54,297 Daniel: What about you, Dara? 689 00:34:54,297 --> 00:34:55,347 What have you been doing to wind down? 690 00:34:55,557 --> 00:34:58,371 Dara: So, I've been on holiday at least at the time of recording. 691 00:34:58,371 --> 00:35:02,311 So I'm going to just back I got back two nights ago, we went to 692 00:35:02,311 --> 00:35:04,791 Tenerif for a bit of sunshine. 693 00:35:04,876 --> 00:35:08,704 Lots of outdoor activity, lots of hiking, time in the sea, it was really nice. 694 00:35:09,064 --> 00:35:09,514 Daniel: Oh, beautiful. 695 00:35:09,574 --> 00:35:13,324 Dara: And very little time on the screen, which is always great. 696 00:35:13,324 --> 00:35:16,474 A true wind down. 697 00:35:16,784 --> 00:35:20,104 Daniel: Yeah, a true wind down, an escape, amazing. 698 00:35:20,454 --> 00:35:24,994 Dara: That's it for this week, to hear more from me and Dan on GA4 and other 699 00:35:24,994 --> 00:35:29,104 analytics related topics, all our previous episodes are available in our 700 00:35:29,104 --> 00:35:34,264 archive at measurelab.co.uk/podcast, or you can simply use whatever app you're 701 00:35:34,264 --> 00:35:38,704 using right now to listen to this, to go back and listen to previous episodes. 702 00:35:38,984 --> 00:35:41,764 Daniel: And if you want to suggest a topic for something me and Dara should be 703 00:35:41,914 --> 00:35:45,734 talking about, or if you want to suggest a guest who we should be talking to, 704 00:35:45,734 --> 00:35:49,029 there's a Google form in the show notes that you can fill out and leave us a note. 705 00:35:49,479 --> 00:35:53,289 Or alternatively, you can just email us at podcast@measurelab.co.uk to 706 00:35:53,294 --> 00:35:54,939 get in touch with us both directly. 707 00:35:55,419 --> 00:35:58,299 Dara: Our theme is from Confidential, you can find a link 708 00:35:58,299 --> 00:36:00,099 to their music in the show notes. 709 00:36:00,669 --> 00:36:03,819 So on behalf of Dan and I, thanks for listening, see you next time.