Using Analytics to Drive Optimization and Personalization
So I'm going to talk to you about a subject that is near and dear to my heart. It's using Analytics to drive optimization and personalization.
So a little bit about me. I am Krista Seiden, the Analytics advocate as well as a product manager at Google, Product Manager for Google Analytics. Cool dual role because I get to be out in market and speak and teach and train and hear feedback. What's going right? What's going wrong? And then bring that back to the team and actually use that to build a better product.
I've been in the industry for about a decade at companies like Adobe, the Apollo group and now Google for the past five and a half years and I'm really involved in the Digital Analytics Association and especially Women in Analytics. I'm, I actually have a bunch of Women in Analytics stickers with me, laptop stickers. If you want one find me after I would love to hand these out. If you want to call me on my cell phone, please don't, but you can get a hold of me on Twitter @kristaseiden or my blog kristaseiden.com.
So let's dig right in.
Factors that Contribute to Low Conversion Rates
To begin, to really to set the stage, there are a few factors that contribute to low conversion rates on our websites. So the first is user experience. Is your site slow loading? We've heard a lot about that today or just really hard to navigate. Do you have site content that is personal? Does it speak to your users? Are they finding what they want on your site?
Do you have actionable web analytics? So this is obviously very important to me being in the analytic space in particular and making sure that you're collecting the right data to be able to understand what's going on with your websites and optimize is critically important.
And finally, do you have the development resources that you need to actually do all of this? Often times I hear and I've been the person saying this, I don't have enough resources this quarter to do the analytics project I want to do and run all of these tests and I have to decide this really tough decision between doing all of it.
Thankfully, we have a lot of tools these days that help us to reduce the impact of that – tag manager, optimize various Google tools, as well as a lot of other third-party tools that help to relieve that burden, but it's still a critical cost to our businesses.
So today, we're going to focus mainly on numbers two and three, so site content and personalization and actionable web analytics.
And personalization is really important. 94% of marketers agree that you need to be doing personalization to be able to succeed and those who are doing it are actually seeing a 19% uplift in sales. It's so important that the Harvard Business Review actually has this to say – they say, “Not to know your customers, not to understand their unmet needs, and not to invest resources based on those needs is to seed one of your most important assets to potential challengers.”
4-Step Optimization Framework
So I have a four-step optimization framework to share with you today.
The first step is analysis. We're going to spend a good amount of time talking about this. This is understanding what you're saying, is telling you and finding those little nuggets to actually optimize on your site.
Next is hypothesis. So what might we do to improve this?
Then of course, we have testing, running those tests and seeing how they're actually doing out in the real world.
And finally, personalization, which is what we're going to spend the second half of this talk talking about.
Using Analysis to Drive Optimization
So let's dig in with analysis. I actually have five tips for you guys of how you can use analysis to really drive optimization.
So the first step is using campaign tagging to distinguish your variations. Now we have a lot of traffic coming to our websites from a lot of different places, right? We're getting people coming from email and social media and ads and all sorts of places and we want to know exactly where they're coming from, from which campaigns they're coming from so that we can understand which of those are driving the highest conversions for our businesses.
So the first one is email testing. Now, if you're doing email testing, if you're testing different headlines in your email or different calls to action on your buttons, you want to be sure that each of those buttons leading to your website is tagged individually with the variation that you are testing.
Now in Google analytics we call this UTM tagging if you've ever heard that term. In some other tools, it's called campaign tagging or many other names, but it's all the same idea.
Next we have ad testing. So if using AdWords or DoubleClick, hopefully you're using auto tagging. However, there's actually really cool feature in Google Analytics that let you override auto-tagging or let you add to that auto tagging and I like to use the UTM content field to actually add to my ad testing and what I do is I add the headline of that add to the UTM content field and that way when I'm doing my analysis in Google Analytics, I actually can see the exact headline of the ad in my data that is driving those conversions. It makes it much more tangible as an analyst and much more readable to your senior execs to actually see that headline copy right there in your analytics data.
And finally is social media. So if you're doing social media, which I'm sure many people in this room are, it is critically important to tag each of your posts individually. Now, this means that you actually need to create a different short URL for every post you’re putting out. Sounds a little bit painful, I know, but it's worth it and here's why.
So this is an example from my personal blog. And I am sending people from Twitter and from Google Plus to my blog and in the content field, I've actually put the headline of the blog post that I am driving traffic to. And once I start to collect a lot of data I can do some analysis and what I see here is that when I blog about Google Tag Manager, I'm actually getting much more traffic coming from Google+. When I blog about pretty much anything else, most of my traffic is coming from Twitter.
Now after seeing this and digging a little bit more, I realized there's actually a very active Google Tag Manager community still on Google+. Who knew anybody was still on Google+? Don't tell my colleagues I said that but it was really good for me to know because now I know when I actually post about Google Tag Manager, I should make sure that that post is on Google Plus, which is a platform I might have stopped posting on otherwise.
You can get much more granular with this actually and you can put in that same UTM content field. You can put a date, a time stamp. Then you can start to understand exactly what time on which platform is driving the highest conversions for your business so it can be really, really powerful.
Implementation to Track Conversion
Okay, our next tip is implementation to track conversion. And this comes with a data layer, so raise your hand if you're familiar with Google Tag Manager. Wonderful. If you're not, really quickly, the data layers and extra code on your page where you can put things in there and use that to send custom dimensions or various things to Google analytics. Very, very powerful.
So I'm home in San Francisco and I am getting hungry. I'm really craving some Italian food delivery. So I Google Postmates Italian food SF. Postmates is a food delivery company and I look at that first one, Tomaso's restaurant in San Francisco. Sounds good. Click on it. And I land on Postmates’ page and I scroll through here, looks pretty good, that lasagna looks excellent. I decide I'm going to order from here.
Now if we switch gears and we pretend that we are the analysts for Postmates and our boss, she asks us, what are the top types of foods in our top cities? Because we want to make sure that our marketing and our outbound efforts are really speaking to those top favorite foods in each of those target cities, right? Pretty simple, pretty straightforward.
I actually have a hard time answering this, though. If we focus on that URL, it says ordered at Postmates.com slash Tomaso's Restaurant, San Francisco. Now if I look at a few other pages, I would actually notice that it's not common or not on every different page that the city is in that URL. It doesn't tell me the type of food it is. I'm not able to dig into a pages report by page category. There's no sub directories in there.
So this is where the data layer comes in I can actually add information to what I'm collecting about this page view. So I can add something for city is San Francisco and page subcategory is Italian and then I'm going to take those data layer variable and I'm going to send them to Google Analytics as custom dimensions. And now, I can add those custom dimensions as a secondary dimension or as a segment to my reports and actually answer that question that my boss wanted to know about the top types of food in our top cities. So a really, really powerful way to add to the data to make sure that you have all of the information that you need to answer really specific questions like that.
Understanding Site Engagement
So our next tip is understanding site engagement. I'm going to stick with Google Tag Manager here for a bit. So last fall, I think it was October, Google Tag Manager released two new triggers that I was overly excited about. Actually I was in Paris, I think, when these launched and I kept pinging my product manager saying, is this live yet? Is this live yet? It was like 2:00 in the morning in Paris, but I really wanted to push out my blog posts on these because I was super excited about it.
So the first is scroll tracking. Now this isn't new, this is something you could have done for a very long time in Google Analytics. However, it required following complicated very long blog post with instructions, a lot of code on your page. I see you nodding. I know it was painful and now with this simple trigger in Google Tag Manager, I actually set this up on my own blog in five minutes. And it could have been less but I was stopping to take screenshots along the way.
Yeah. So the next one is the element visibility trigger. Now this one is probably even cooler. I'm using this one on my blog to send an event to Google Analytics to let me know when my newsletter subscribe box has been visible for 2 seconds or more. And I want to know that because then I can calculate my newsletter subscriber rate. How many people who actually had the opportunity saw the box to sign up actually signed up. Pretty cool way to be able to actually know exactly when certain elements were visible on your site.
Understanding Form Engagement
Okay, our next tip is to understand form engagement and we’re going to do this with event tracking. So you're probably somewhat familiar with event tracking in Google Analytics.
Now event tracking is great. It's very free form in Google Analytics. That's one of the things I love most about it. It's also one of the things I hate most about it is that anybody can put anything they want into any of these fields. This is a recommended strategy for how you might want to use event tracking in Google Analytics.
So we have our category. This is the location of where an event might be taking place, the home page, the about us page, the contact form, something like that. The action is the action that the user actually took. So if they watched a video, the action would be video. If they downloaded a white paper, it would be download white paper. And then the label is just additional details about that action. So it would be the video name if the action was watch video and so forth.
So let's look at a real example of this. I implemented the Google Analytics marketing website. So analytics.google.com. or google.com with dot slash analytics with event tracking a little while back. And I use this formula for how we would do this, this hierarchy.
So you can see here the category. These are the locations of where these events are taking place. We have our homepage, our tag manager overview page. If we focus in on line number 8, the contact form, if we click this we get to the next level of our hierarchy and that's the event action.
So here I've used the event action to collect which product somebody was interested in when they clicked on that contact form. And finally if I click into one of those let's click into analytics 360 Suite, I get to my event label. And here I've used on interaction events to send this event when somebody is actually clicked into a field and started entering text. So as soon as they start typing into the first name field, we send an event set last name field, etc.
Now normally tables in Google Analytics are sorted by the first column. So in this case, it would have been sorted by total events. I resorted this, you can tell by the downward facing arrow on the unique events column by unique events. And what we actually see is a nice little funnel. This is how people are actually going through this form. First name, last name, job level, email. There's one outlier down at the bottom, that's region. It's because it was an optional form field.
But there's something else that's very interesting here. Anybody see something that catches their eye? No? Speak up.
It does decrease that's how they go through the form. So people are dropping off as they had to fill out more forms right there.
Man from audience:
The total events on the…
Yeah, so our total events, if we focus on line number 9, submit, is significantly higher than our unique events there. And I was scratching my head. I was like, huh? Clearly people are trying to submit this form, but they're not able to submit it. Why might that be?
So I go to the form and I actually look and it turns out that the little asterisks that indicate that a form field is required were really small. They weren't very apparent. And so people were trying to fill out this form. They weren't able to submit it when they're hitting submit. It was probably leading to a lot of user frustration. But I let our UX and design teams know that and they went in, they made that fix because of what we saw in this form field analysis.
So a really great way to understand at a very granular level how your forms, how your website is performing in areas that you might want to improve.
Okay, number five. Your site can tell you what's important.
So in Google Analytics and in a lot of analytics tools, we have something called site search. Now this is useful if you have a search box on your website. You enable it, it's a simple toggle in the admin settings along with picking up a query parameter, but what you get from this is super valuable.
You are getting the actual search terms that somebody is entering into the search box on your website. That's the actual voice of your customer telling you exactly what they want from your website and maybe they can't find it. It's not very apparent to them or it's hidden or there's no content on what they're looking for.
So again, we're looking at an example from my personal blog and I can see that people are often searching for Google Tag Manager. They want more content about Google Tag Manager. That's what they're coming to my blog to find. So perhaps I should consider blogging a little bit more about Google Tag Manager.
And finally heat maps. So these are great tools to pinpoint areas that you need to optimize. Now Google Analytics does not actually have a heat mapping tool. We used to have something called in-page analytics. It sucked so we got rid of it. But there are a lot of other third-party tools that are great at this.
So one of my favorites is actually Crazy Egg, which is this one and what I like so much about this is the report we see on our screen. This is the confetti report and it's actually one colorful dot everywhere somebody has clicked on the site regardless of if it's a clickable link or not.
And what is really cool about this is you can actually change what those colorful dots mean. So right now, they’re referrers. But you can actually change it to any one of these other dimensions here.
And so one of my favorites is actually time to click. And if you change it to time to click you can start to separate out clicks that happened in under a second versus clicks that happened in over a minute and you can start to understand the type of actions on your website that take very little thought or easy to interact with versus things that maybe take a little bit more time to interact with or people are just a little bit less sure of. Pretty cool.
Now a great use of this tool that I've had. Before I came to Google, I worked at a higher education company. And we spent a lot of time developing a new career portal for our students and it had three main calls to action. It was things like help with a resume and a job search things like that. And we launched this page after months of development. We were really excited about it and we saw in Google Analytics that we got a ton of traffic coming to this page. B ut hardly anybody who was actually clicking our links and going to these resources that we had spent so much time building for them and we couldn't figure it out. We were scratching our heads. The page was working fine. The links were working fine, what was going on?
So we put Crazy Egg on top of that and what we saw was that people actually were trying to click. They were trying to click on the images associated with each of the main calls to action. But in our development process we had overlooked making those images clickable. I know, it was terrible! But it was a quick fix. The only thing that had been there was a tiny little text link, clearly people weren't finding that but Crazy Egg let us know that people were interested. Once we made that quick switch, it was a big win for the company.
So let's go through an example of analysis now that we've talked about a lot of different analysis techniques. So for this we are going to use the Google Merchandise Store. Now this is a online e-commerce store where you can buy Google branded gear. It's a real store. You can buy Google t-shirts, YouTube mugs, that kind of thing.
What's really cool about this store though is actually it's the Google Analytics demo account. Anybody familiar with the demo account? Yeah? So this is a free open data site anybody can play with. It’s real data from a real e-commerce store. And it's a great way to get an introduction to Google Analytics. So you can actually do the same analysis along with me if you really wanted.
So for this analysis, we are going to look at the Google Merchandise Store and we specifically want to know the performance of our paid media. So we're looking at our AdWords report. We're looking at the keyword report for that. Specifically at two key words. So we have Google Merchandise and Google Merchandise Store. And what we notice is that on mobile devices, Google Merchandise has almost twice the conversion rate as Google Merchandise Store and we're scratching our heads a little bit because these are very similar search terms. Why do they have such a drastic difference in an e-commerce conversion rate? What's causing it?
So that's where we come into the hypothesis stage of our optimization framework. We want to come up with a few ideas about why Google merchandise might have a higher conversion rate than Google Merchandise Store and figure out ways that we can actually test that.
So one of the things that I was thinking was that okay, if people are on their mobile devices and they are searching for Google Merchandise Store, they might be looking for an actual physical store location. Now, I happen to know there are only two of these in the world. There is one at our headquarters in Mountain View, California. And the other one is at our European headquarters in Dublin, Ireland. So if you're searching for Google Merchandise Store on your phone anywhere else in the world, you might be pretty disappointed if you go to this online store and figure out there's no physical locations near you. So that could be one of the reasons why we have a lower conversion rate for that term.
One way that we could possibly help with this is by maybe on that landing page highlighting our online only gear and fast shipping and other things that we offer online that we don't offer in store and targeting that towards these particular keywords on mobile.
Which is where we come to our testing phase. So now we want to test these hypotheses. So we're going to create a few variations of this potential test looking at different ways that we can incentivize users who are coming from Google Merchandise Store to stay and purchase stuff from our store.
We're going to select our audience. So our goals first and that's going to be increasing that e-commerce conversion rate. Next. We're going to target it to our audience that's mobile devices specifically for those two keywords, and I'm going to run this test, analyze the results and hopefully we're going to find a winner and improve that e-commerce conversion rate.
Personalization Maturity Framework
Now this brings us to the last part of this presentation, which is personalization. So this is the idea that we want to know what personalized experiences should you provide to your customers and what's the expected benefit of doing that.
Now if you're anything like me, you hear the term personalization and you roll your eyes and you say I don't want to buy a really expensive tool set that does all these fancy things because then I'm going to also have to hire really smart people and data scientists to run them and I don't have the development resources and I don't have the budget this quarter and I'm here to tell you don't need to do that. You can actually personalize with what you already have, with the data that you already have.
So I'm going to walk through another framework, something I call my personalization maturity framework to show you how you can do that.
So the first step is mass scale and this is the idea that you can personalize just by understanding who the majority of your customers are.
Next we have specialization. And this is the idea that you can take a certain type of people like a certain locale and target them with location specific content.
Then we have referral source. So targeting People based on where they are coming from.
And finally, we have actions. So targeting them based on what they are already doing on your website.
So let's start with mass scale.
So when I first came to Google, I built and ran an analytics and optimization program for the Google Apps for business group. What is now the G Suite Group by Google. So if you have Google Docs or Drive or Gmail for your work, that's this group.
And this was one of the marketing websites that I owned and that I wanted to personalize. And we had this headline that said ‘Discover a better way of working. Now, marketing loved this headline. They thought it spoke to this holistic message of the suite of tools that we are selling -- Docs and Drive and Hangouts. All of these things were going to help you to work better together and discover a better way of working.
And Engineering said no, that's a load of crap. We know that 80% of the people that come to your website only ever use Gmail so we should change this headline to talk to those users. And we were like that's never gonna work. We have this holistic message. But okay, we’ll test it.
So we ran this headline test and we tested “Discover a better way of working” against four headlines that -- let me stress -- our engineering team wrote, “Get Gmail for work 99.9% uptime and 24/7 support” or marketing’s favorite, “Get email for your domain.”
Yeah, we were pretty confident that “Discover a better way of working” was going to prevail here. So we fast-forward two weeks, this test runs, I'm actually over in Australia for another conference. It's bedtime. It's like midnight. I'm getting ready to go to bed and I start getting frantic pings from my VP of marketing.
And she is saying don't you dare launch that headline. I'm thinking. Oh no, what's going on? So I log into the results really quickly and what I see is that “Get email for your domain”, one. Not only did it win, it won by 20%, which was huge for this business. And we're thinking, oh man, this is the worst possible headline that we could have on our site.
So at the same time I'm getting these frantic pings from my VP of marketing, I'm also getting frantic pings from my VP of engineering. And he is saying, you know that server-side testing tool that we built you so that you could run these tests? Yeah. If you don't launch our headline, we're definitely going to cut support for that.
So I was in a little bit of a tough spot. I had to mediate between these two groups. And what we ended up doing was we went ahead and launched this horrible headline while I stayed up for the next few hours and created a couple of test plans to come up with some new tests that we would launch the very next day to try to get us back to a more marketing friendly headline.
Fast forward a couple of weeks, we actually came out with a headline that said, “Get Gmail for work” with a subtext of and Docs and Drive and Hangouts and everything else that helps you work better together. So everybody was pieced in the end, but there are two main lessons to take away here.
First is that we actually personalized this website. We spoke to the 80% of our users who were only ever going to buy Gmail and by doing that, we increased our signups by 20% which was really, really big.
And the second lesson is if you are going to do testing, make sure that you have sign off from all the people in your organization on those variations before you launch them to avoid losing a lot of sleep at night.
Okay, so specialization. Same website. This is our US English version of this website. Now Google, we have this funny way of working where we tend to have one global template of our website and we translate that for all of the different locales that we are in.
So this is our US English website. And this is our Japanese website, one more time, US English, maybe yep, and then Japanese. Now who here has ever seen the Japanese website? Does it look like this? No! If you have not seen the Japanese website, please humor yourself and go find one. But also it is incredibly busy. There are widgets and pieces of information all over the place. I personally see it and I'm a little bit overwhelmed but that is what Japanese people and that market expect and that is what they are used to and how they use the internet.
So a plain vanilla website like this was just not speaking to that audience. Specifically our team really wanted to get rid of one particular button and go back to the US English one for just a second. In the upper right hand corner, you see that grey button that says contact sales? So my team in Japan was very clear. There is nobody that is going to click on that button and fill out a form to buy enterprise software online. That is not how you do business in Japan. In Japan, they want to have a formal meeting. They want printouts. They want to be able to take that back to their companies, give those printouts to their bosses and have a much more hands-on approach to this.
So our team came up with an idea of how we can maybe bring some of that hands-on approach to the web. So what we did was we actually replaced that button with something that is roughly translated to flow of available. But actually it was a guided flow and it was a pop-up box you clicked on that you got this pop-up box, and it was this whole guide walking you through the 30-day trial process of this business and what you had to do throughout this trial process in terms of verifying your domain and setting everything up and what you needed to do to convert to a paying customer. And it was printable.
So what we saw from this test was we actually saw almost a 700% increase in clicks on that button and it led to an 8% increase in people actually starting the trial process. So this was really, really huge for this market and it was mostly like an i-told-you-so from them.
But you know, we took this a few steps further. So we actually mocked up what a Japanese version of our website could look like, a lot of widgets, a lot more information on there. We even had a nice Tokyo skyline that we wanted to try out in the header and a year later, after some more testing, that's the website that we actually launched. And it was the first time in our marketing team’s history that we broke with the global template and launched a locale-specific website. And this website did significantly better for the business.
So that is specialization.
Next we have referral source.
So I'm sure many of you in this room have spent a lot of time creating ads in your career and you've spent a lot of time finding the perfect ad copy and the perfect imagery and making that perfectly interesting ad for users to click on. And when they do click on it, it's actually a really good signal of user intent. They are clearly interested in what you're offering. They have clicked on that ad, they want to see more.
So we want to make sure that we have the right landing pages for them when they actually go ahead and click on those ads.
So let's walk through an example. Let's consider an example for a shoe company who is targeting a summer sale to an audience of 18 to 24 year olds. And you have that perfect ad that you've created, they click it. They're interested and they land on your landing page.
So here we have an example of a landing page. You can see there's some sunglasses and some shoes, some young people having fun in the sun, talks about our sale. So this could be a very good landing page, but maybe it's not the best. How do we know if this is the right landing page for those ads? So we want to test this maybe we can test it against a landing page that is all emojis because maybe emojis speak better to a millennial audience.
I don't know about you, I can't read that so it doesn't speak to me, but maybe it works for that audience. Who knows? But the important thing here is that you are testing from that referral source to your website to make sure that you're capturing as much of that interested audience as you can.
This brings us to our last piece of this framework and this is actions.
So here we have a website. This is Light in the Box. It's an online retailer. You can buy whole lot of things kind of like in Amazon and I have come to the website and I am really interested in digital cameras. So I've done a lot of searching. I've maybe even added one to my cart, but I haven't checked out yet.
So if again we flip and we consider ourselves to be the analysts for this business. Now, we're in Google Analytics and this is actually a report from Google Analytics 360, our paid version. This is the custom funnels report. Anybody familiar with this report? Yeah. Those of you who aren’t, it's a pretty great report.
So I've created a funnel here for people who have come to the website and then people who have looked at a digital camera, added to their cart and then checked out.
Now what makes this report so great is that any of those little red arrows you can actually click on and it's going to pop up a box, in this case from add to cart that says hey, there are 2346 people who added to their cart, but they didn't check out. Do you want to create a segment and remark it to them? Well, of course we do because we're all good marketers here. So let's go ahead and do that. We're going to create this segment of users who have added to their cart. They're interested in digital cameras. They haven't checked out.
Now as we're creating this remarketing ad we probably want to offer them something enticing to get them to come back to the site. Maybe we offer them free shipping on their next order when they come back through this ad. And then when they get to our site, we remember that we want to bring that information through to the site and have a really personal touch when they land on the site.
So we're going to bring that same messaging to the landing page and it's going to look something like this, probably you'll spend a few more minutes on designing this than I did but it's going to say welcome back, get free shipping on your next order.
I think that's a really great way of bringing kind of all of these concepts together and really personalizing that user experience and bringing them back and getting them to actually convert.
So that closes out our personalization maturity framework. Which was part of our optimization framework. And that is all I have for you today. So I want to say thank you very much for having me.
Any questions for Krista while she’s here?
Man from audience:
So uhm, thank you first of all. Second of all, with the breaking up the global rule for the Japanese market and the success that that had, did that makes sure that Google changed it for other countries too? Or why not if it's so profitable.
Yeah, that's a great question. I think that's a really individual question to every business. So in our in our org we had enough development resources to do that for a specific number of locations. What we found from a lot of our research was that most locales did pretty well with the global template with a few minor tweaks. We actually did a lot of CTA testing across different locales. And we used to have the same CTA on every single website. And then after we did that round of testing we ended up with four or five different CTAs that ended up working best in different locales, but something like the whole site thing, we did that for Japan specifically.
Man from audience:
And that was the only time you guys did that?
When I was on the team, yes.
Man from audience:
For that last panel that you presented. So that's only available in in the paid version of Analytics? Okay, so if you don't have the paid version, are you able to take an audience that didn't convert based on a specific segment and we market to them with one click or you have to do it the long way?
So it wouldn't exactly be one click, but you can do a lot of the same analysis through a customer report using goals and then creating an audience from that data.
Man from audience:
Man from audience:
I have a similar question with the returning user, if they actually were to have something in the cart, would you have it so that they could just basically return to the cart pick up exactly where they left off, hit pay?
Yeah, I mean, I think that's up to the business for sure. Depends on what your goals are if you want to get them to add other things to your cart or cross-promote, but I think it's probably a pretty good user experience to drop them right back in to where they left off if your goal is just to get them to complete that purchase.
Man from audience:
Woman from audience:
Hey, this may be a bit of a can of worms question. But thinking about the GDP, our deadline looming at the end of May and thinking about what tracking and analyzing and testing looks like in that world. What are some kind of higher level recommendations for how to think about potentially having users who totally opted out or may not have consented in and how to think about approaching these tactics with those regulations in place.
Yeah, absolutely. So GDPR is a big can of worms and it's not something that I am authorized to talk on in terms of or on behalf of Google. What I can tell you is that our teams have been very hard at work to make sure that we are 100% compliant. We're starting to put out comms this week in terms of what that means for the market and for our products and downward for our clients. So pay attention to those comms that are coming out. But we as a product will definitely be GDPR compliant.
Fantastic, in which case, that hits perfectly on time. Let's give a big thank you to Krista Seiden.