 All right, hey everybody, glad y'all can make it and excited to talk to y'all, let's get started. So yeah, my name is Scott Massey. I'm the head of product and marketing for Morft. I joined the team in January prior to that. I was with Pantheon for about a decade from the very beginning, ran support, customer success, Docs, Deverell, everything sort of after the signature for a while and then I moved to before the signature and helped marketing and sales and my last role was running our global go-to-market. And yeah, so I'm super excited to be on the Morft team working on cool stuff and personalization and some of the sessions you might have seen earlier about AI and that sort of thing. Naveen is also a colleague and a coworker, just due to logistics, he couldn't make it but he's an awesome developer on our team that builds awesome websites and as far as Drupal goes, really knows his onions as the British people say. Maybe Kiwi say that too, I don't know. I know Americans don't say it. But this American, he said something pretty famous like years ago, this guy's John Wanamaker and he's sort of the forefather of marketing and he had this quote that like, half the money I spend on advertising is wasted, the trouble is I don't know which half and that's why he has his hand to his temple and he's glaring it because he's trying to figure this out and even though technology has advanced rapidly and there's all sorts of new tech out there, modern site owners and marketers still struggle with these questions that are kind of phrased in a different way about like kind of like, at a high level, like is this site working or is my marketing working? What messaging is resonating? Like where are we losing users and customers and where could we improve our marketing? And I think it's difficult, it's made more difficult in today's world because of this like this Omnichannel, marketing structure that we work in where there's not just a website but maybe a portfolio of websites and there's devices and then there's iOS apps and Android apps and all these sort of things that kind of each have their own like visibility into the customer journey. So measuring them is hard. It's almost like sometimes there's a, almost too much to measure and it's hard to tell like what's really correlated with like customer happiness or for a value. And I think like segmentation and things like that, there's all these cool tools and there's this whole MarTech industry that's out there but a lot of this requires the resources that really only put, really are only accessible to enterprise level companies and that's kind of why I wanted to talk about this because I think there's a couple of things with a couple of advancements in Google's products that sort of allow us to make use of analytics in a different way and actually sort of get the same benefit that maybe a company that has millions of dollars of budget to spend on data science and advanced kind of MarTech stuff. So yeah, I'm here to kind of talk about analytics and I know what the troublemakers in the back are thinking you're thinking like analytics are boring and stupid and I didn't even want to come to this one. I meant to come to the other room so why do I have to listen to analytics the last session of the last day like why are you subjecting me to this? And so I at least want to posit to you that these three things are true that Looker and GA4 together, whether you like them or not can help you understand your users better that you can kind of think of it in a different way now you can actually do your marketing and your customer journeys and your domain knowledge first and plug that into analytics and be able to not just look at a bunch of canned reports but you can sort of analyze like real sort of action and actual sort of work that you're doing and seeing what the results are and paired with like face-to-face and customer development and like developing hypotheses you can kind of test them and create this virtual lifecycle where they feed into each other and the data you get gives you more hypotheses to test and you can test them on real people and sort of just gradually get better and hopefully fight above your weight class and compete with the big shots. And I think this is kind of happening like this is happening with Google not necessarily because like having worked with GCP and Google for a long time like I have the I always have the sense that like they don't build anything for you they build it for themselves and if it works for you that's kind of like a happy coincidence for them. And so I think that like this is something that matters to them like they need to kind of understand like where users are for their own sort of survival. And so in this one sense we're kind of aligned on this like website owners and agencies that are involved and marketing teams and Sundar Pichai from Google are all kind of looking at the same kind of things and have the same goals here for this talk. So I wanna talk about GA4 looker and how to put it all together with Drupal. I don't have anything to say about AI I just wanted to be cool so I put it up there. So yeah, so GA4 like as I was preparing this it resonated like how much GA4 is kind of like the Drupal 7 to 8 conversion. It's a pretty big C change for Google. And so if you've logged into your analytics you've probably seen this countdown to when they're gonna lift and shift it for you if you don't do it yourself just because the kind of the under the architecture of what was called well universal analytics or I think officially it was like called like web plus app or something like that is changing significantly enough to where it's a lift and shift and they'll do it for you but if you have custom stuff in there it may break. So I wanted to just kind of describe kind of the meaningful like what I think are kind of the important and relevant things that are changing here. One is the tracking like Google Analytics originally had this concept of users that had sessions these sessions were a bunch of page views and that's kind of the main sort of nouns that were involved here in this kind of process. And like since then obviously the aforementioned channels and applications and multiple sites and devices and things like that. And plus that content itself is much more dynamic and personalized and a lot of stuff going on. So like they had to kind of update the antiquated model for how stuff works. So now they have this concept of data streams and a data stream is sort of a place to ship data to Google Analytics or wherever you needed to go but you set up these data streams and then you can aggregate them and analyze them. And so they also have this concept of mixed reporting identity where they're attempting to be able to track one user across those different data streams across devices and things like that. So you can measure like a single user as they switch between mobile and desktop or whatnot. A lot of this stuff is still like in the works and sort of black boxy at this point but that's some of what's going on that I think is relevant to us. The user interface is also changed significantly which I'll get into because whereas GA used to kind of come with a bunch of standard kind of canned reports. It's much more looker-ish like you don't get those same reports but you're sort of encouraged at this point to use the kind of looker user interface to build your own reports and it's pretty powerful but it's not necessarily what people who've used GA before are used to. It's kind of a new paradigm for reporting but it does give a non-technical marketer a lot more power to do all sorts of stuff to have these events, you know, things that happen on the website that they want to track that's relevant to their domain or whatever and be able to ship them to Google Analytics or Looker or kind of wherever you want to. And yeah, so the data model is kind of built up of these events. Like a page load is an event now whereas it used to be kind of like a page load so now it's one of many events like one of hundreds of events that exist now in GA4 but can also be kind of you can create your own and have them in there. So it's very much open-ended to be a cool tracking tool. And then finally, like Google is committed by 2024 to not use cookies anymore. So that's also kind of a black box as to what's gonna happen but it seems to be centering around this consent mode kind of thing. So like there were updates kind of required to move forward with that initiative. So the reception across the board has been pretty solid and stable. This poor woman I think is hyperventilating. I don't think she's throwing up but I think it's taking her breath away and hopefully she'll adapt and be okay there. I think like in the words of kind of tech evangelist Kelsey Hightower, like tech is less about learning a specific tool as it is like the willingness to learn a different tool when that time comes. And I think if especially we Drupal people if we embrace end of life as a way of life like these sort of changes we can kind of look at the benefits of them. And I think like in general moving forward with GA4 is a good thing for us. That being said, like it's a means to an end. I have no particular affinity for Google Analytics and I'm not like a SEM guy or anything like that. But on the other hand, Looker I do like. I've been a fan of Looker forever. I've used it a lot over the last 10 years. My prior company, hundreds of users depended on it. Hundreds of my fellow coworkers depended on it from accounting to operations to support to sales and marketing. It's a super tool. It existed on its own for a while then it was acquired by Google and has been through some branding iterations. But now it's kind of pretty well integrated into their marketing suite. And it's kind of like built on, it's almost like an MVC sort of format but it has these models and views and it's very kind of DevOps friendly in the large scheme of things. You can, it's versionable and you can commit files and create the initial database views. And from there, you can expose those to the Looker users and they can build all kinds of reports and visualizations and stuff. I also think like dashboards are kind of good for agencies and for stakeholders. Like I recommend dashboards. Like I think you'll always make more money if the guy signing the check or the woman signing the check has a dashboard as opposed to a CLI. So I think like dashboards are a positive thing and Looker makes them very customizable and editable Yeah, and so like it does BI really well and it scales. You can aggregate across multiple sites, across a portfolio. It hooks into a big query if you want to do data warehousing or data warehousing or machine learning kind of stuff. But that being said, like it's not an ETL, like you kind of have to get it into a format that Looker can read stuff either in a MySQL database or one of their projects. And I think like it probably requires someone with at least like the willingness to be like mid-level, like MySQL competent. Like it uses its own Look ML kind of modeling language and you kind of have to learn that. But again, I think Drupal people are kind of used to like the abstracting of the database connection in code. So it's not super complicated. So I think that's a good point. I think that's a good point. So it's not super complicated. You get all these connectors with it. You can hook into Google Analytics and that kind of stuff off the bat. So you don't really need to have a whole lot of code to do that. That's pretty much out of the box. And then there's partner connectors and custom stuff that you can build. And the reports obviously, it has all the stuff that you kind of expect. But again, since you're kind of embedded with Google on this, you get access to their intelligence around building maps and time series and stuff like that. So it's out of the box, it's pretty powerful. The Drupal side of things is pretty straightforward. Like you can, Google Tag Manager may get you 100% of the way there. If you have that module installed, you can install the container ID and start creating events and placing events and start your own kind of analytics. But you can take it further. You could pipe data through the data layer, which is kind of a good way to take stuff out of Chrome and ship it wherever you want. And we use it for personalization. We use a lot of these tools for personalization because we can sort of grab what's going on behavioral data and create actions based on that and show people content that we think that they want to see. So there's some other links there that might be kind of helpful for for you to look at and get a feel for it. So let's put it all together and I'll show you a couple examples that I have here. So like marketers have these funnels and funnels are ways of gauging user interest and at the top of the funnel you have people that may be interested in a product but have no idea what it really does and at the bottom of the funnel there may be comparing pricing of one your competitor to you. So where people are in the funnel matters to marketers and creating content that's relevant to each stage in the funnel is important and like one of the cool things you can do is you can sort of expose content taxonomies as meta tags put that in the page and then when someone opens that page or opens something related to that stage in the funnel you can track that and you can ship that to Looker and then you can run reports that can show like in this case obviously we're mainly maybe a little bit too focused on the top of the funnel like in Discovery and maybe we want to get more in deep because there's users that are possibly falling off because they need to see you know more comparative stuff or more in the consideration phase before moving on there and this is something that like once it's set up it's pretty easy to build you can add all sorts of dimensions to see like by region and that sort of stuff as well same thing with whatever taxonomies you can expose like if it's if you're selling merchandise or something like that or if your content has its own kind of taxonomies you can expose those as well and be able to see see where users are at or what their interests are in or what writers are more effective than others and that sort of thing and so like I also wanted to sort of explain demonstrate the simplicity of this like it is something that really a marketer just has to kind of maybe have it explained to them in one sort of slow session but after that it's pretty easy to pick up on and I wanted to sort of pretend like have a thought experiment that we are selling bicycles and we're running the sale on bikes and so we're going to create this marketing email and we're going to send it to some customers and then we want them to click on a landing page it's going to show them a bunch of bikes they're going to pick one, they're going to add it to the cart and they're going to buy it so it's pretty simple the marketer prepares the email they get ready to send it and now they want to create a funnel report to be able to track that so what they can do is they can set up the page the way they want it to like focus on the right side where they're actually doing the navigation on the left side there's some debugging stuff that's showing the tags and the events that are happening and we'll get into what to do with that in a second but so they're going to navigate and duplicate the desired ideal customer journey there they're going to go to this bike and maybe it's on sale and they're going to click on it they're going to load the product detail page and they're going to say yeah I want to buy that stupid bike and then they're going to go to check out they're going to add it to the cart that's the purchase journey they're going to do the sign in and go through the transaction there and you'll see on the left what we end up with is we end up with those four URLs and those four URLs duplicate they replicate exactly what we went through they'll view the page, view the bike, give me the bike pay for the bike so now all the marketer really has to do is to figure out how to codify those into reporting and you can do that by debugging you can do this by kind of common sense you're looking at the URL but if we look at each stage we see there's a page view event because that's what GA4 is built on and it has those words in there it's that URL so we remember that we go to the next one and then there's another event that's a page view for the product detail page with that item in there so maybe we grab that item instead and we save that and then we have two more events that are basically the add to cart and the checkout page so we take that information that we've copied and pasted somewhere and we build a report we build a funnel exploration report and we go down here to the steps and we just basically recreate those in steps in Looker or in Google Analytics one being the landing page you'll see on the right hand side that we go from 100% like all users we're using one of Google's demo sites because there's real traffic being simulated on there so we'll add that first page in there and we'll see that it begins to get filtered we add the second step in there that shows the product detail page and we see that it gets further filtered down as it's actually running the report on you know existing data we add the rest of it in there we're adding the shopping cart and then we go to the checkout which is the final stage there hopefully at this point we've replicated the entire desired transaction that we go through here we'll plug in that last filter there to the sign in .html page and that's basically it we apply that and we end up with this funnel report that shows like where people fell off at each stage in the step where carts were abandoned to begin to optimize and like literally it's like a 5 or 10 minute report that can be run for every campaign to kind of see what's going on beyond that you can add dimensions if you want to I'll add campaign here to see like maybe what happened organically to filter out that to see how effective really the marketing really was so I can go in there and I can see like what's direct versus organic versus referral if I'm using a particular URL I could filter like that according to that as well I can also add in these segments like customer segmentation is hard it's an art slash science but this uses some common models of like recency and frequency or you can create your own based on purchase amount or existing users versus new users and that's built in and so you can do that kind of segmentation and modeling and see how your campaigns perform against in those situations as well so in summary you see now he's elated because we've actually made a dent in the omnichannel complexities the marketing, the measurement obstacles and the audiences and segmentations and he's happy now and you're happy now too because you realized you were wrong I'm a stupid idiot talking about analytics analytics or fire like the American kids say I don't know if you say that but Americans do so yeah getting back to the point the hypotheses that I made I think that they can help GA is at a very usable point for teams at this point you just have to kind of get a sense of how Google does things and sort of what their mindset is how they build their products and once it clicks it makes sense I think that like the tracking and marketing like if you're doing it in real life it's now much easier to bring that into your analytics and get reports the way that marketers want them that are effective and usable and I think for those teams that are fighting for budget that are trying to do more with less or trying to build out their web portfolio or their marketing portfolio the more successful they'll be and this is I think a good way to do it that's it folks thank you I I love using Looker I've used it for years and the stuff that I did for this was pretty simple but you can do a lot of cool stuff and back in the day at one point Pantheon was using it's a way to get out of Excel like the millions of Excel files and views like we used all of those and we sort of had a corporate initiative to unify on Looker and when everyone's using the same tool then it's great like there's no you know data issues and like you become almost like organizationally committed to data validity and keeping it up I think that they like I think with with Google with acquisitions that I know about you stand a fair chance of it going nowhere and I think that in this case they've really adopted it so I think it's going to be how visualization works for Google like it's going to be their default status and you know I think I don't know if I would expect it to change like it hasn't changed a whole lot for me like it's basically look the way it was when it was its own thing but I think running custom reporting and things like that like you know they're pretty agnostic about customers you know like they're sort of like we're going to do it our way it's going to be more difficult but we think it's better and I sort of agree with them in this case it's not like learning a new mark you know mark up line or modeling language is tricky and some won't be able to do it but I think like if you at least get it started there's a lot you can do with Looker is that a double negative? I'm thumbs down on the sucks I think you know like I said analytics are like the way that it's done is getting better I think there's increasing value I think for me and it's like okay now that it's not just looking at page views and it's been that way for a while but has it's gotten easier to sort of like wrap my head around like oh these key value pairs can be kind of placed anywhere now and now I can just you know I don't need to look at any other values other than the ones that matter for me and just start tracking that like I think the average marketing department can get pretty far with this yeah I've been terrible repeating the question so do I see any big gaps in being able to visualize and analyze what I want to I would say it's pretty fair to say there's not much that I think I couldn't get into there at this point like if you go if you look at my deck and you look at the links to all of the existing dimensions that are in there it's hundreds like it's like looking at api.drupal.org like there's a ton of stuff in there and then being able to do custom stuff I think it's still like my concern is that it's still like kind of not super user friendly and you do need someone that's going to like marketing teams are like a moving train and to stop and say I'm going to learn a modeling language so I can do analytics like a lot of times it's like eh page user fine and get back to business but I do think that like with all this like there are companies that are very successful with this there's actually oddly enough there was one one of the inspirations for me getting interested in this at all was Tableau Tableau has a web team that's running all sorts of like tests on their landing pages and stuff like that and they do it all with Drupal like they have some great Drupal case studies and they were just marketers that dedicated the time to learning this and so I think it probably existed before like the ability to report on whatever you want but it required a lot of hand coding now I would say it's close to zero what you could do it's just the initial time to put into it