 All right, we're going to get started here momentarily. We were informed the keynote went a little bit over. It's going to be a hard act to follow, but we are going to do our best here today. Welcome to Personalize Your Path to Conversion. We're so excited to be back at DrupalCon 2022 in Portland. We're here with Elevated Third, and we're going to be sharing with you a little bit on personalization. My name is Mindy Leigh. I am the UX director at Elevated Third, and I'm here with Govindra Antel. He's a technical architect, and we're going to be sharing with you today how you can make your personalization strategy more successful. Elevated Third is a global marketing agency specializing in Drupal for enterprise. And what that means is we build high-performing websites, and we leverage open source so we can better integrate with technology partners and marketing platforms to deliver results for our B2B clients. So let's see a show of hands if you are a marketer or if you have a marketing role at your organization. There you go. And let's see if you're a developer and you implement marketing technologies. Anyone raise your hand twice? You're a very lucky person. Thanks for coming. You're very busy. So if you raise your hand, chances you've asked the question or you've been asked, how can I increase web conversions on my website? Well, for today's conversation, we're going to define a web conversion as a point when a user on your website gives you information such as their personal data in exchange for something they consider valuable. So they either fill out a form. They're going to send a contact request or purchase a product, maybe download a white paper. So that's what we're going to consider a conversion is that organizational goal that gets met through an exchange of data and information. Well, there's a lot of ways to do that. So what really summarizes down into provide a more relevant web experience, personalization is the key to providing relevancy in their web experience. So we're going to spend some time talking about that. And if you're a marketer, it's probably already on your radar, personalization. But you're wondering, how can I do it better? How can I get more results from my personalization strategy? Or you're asking yourself, how can I make it more personalized? Because when you speak to everyone, you speak to no one. As a marketer, you generally understand that. And that really will help with the relevancy on your website when you start to target those message into a personalization strategy. A McKenzie business survey recently reported that most marketers interviewed see personalization as a prime driver of their marketing success in 2024. So that means they're going to be utilizing AI platforms. Smart content is one we're going to be talking about later today to help reduce the burden on their teams, their developers, and their marketing teams to build out all the scenarios. They're also needing to integrate more closely with campaigns, and so they can increase user engagement and provide meaningful context to those users. Finding that personalization requires targeted prospects. Sorry, I'm getting sign language from the mask. Need a mask? OK, thank you. Excuse me, why put on a mask? OK, let me know if you can't hear me. All right, thank you. Sorry about that. All right, so personalization as targeting your prospects so you can eventually convert more of your customers with that relevant messaging. So artificial intelligence is key to this, because most personalization platforms use some sort of artificial intelligence, and it's misunderstood by marketers sometimes, and this is a great quote of a market I worked with who said, the thing they didn't tell me is that the intelligence behind artificial intelligence was going to be me. So they really rolled out their first content strategy without that insight that there was a lot of legwork to be had. And so for creating that personalization strategy, you really need to go into it without understanding and build out those scenarios, because the system still needs to be told the logic and given the content and messaging for it to work really successfully. So we're going to start with the personalization spectrum because personalization can take place in minimal forms in very simplified ways and through very robust ways within your organization. All of these ways of personalization use dynamic content, which means that the content on your website will be replaced by this personalization mechanic, so that might mean that the user is actually given different images and different content. So on the bottom left, you'll see relationship-driven personalization. Drupal does this really great through taxonomy, so if you're looking at a blog article and you have a related article or a product and you might like these other products, it's on the page what you're looking at and you're getting dynamic content based on that view. Customization is the next level up. It's considered when a user manually makes a selection when they're looking at your site of a topic they're interested in. So if you're on a dog food or a pet store and you pick that you have a dog, it might change the homepage and all the content you're looking at for dog food and dog related items. So that's a user customization. Behavioral data is something driven, personalization driven by previous activity either on that current site or off your site. So things like using cookies or UTMs to help influence something you may be interested in. Demographic data is based on user qualities, such as your IP address, your location data, your title, other information about yourself. And then personal data is where we get into more an authenticated situation where someone may have logged in or opted into a personalization mechanic, such as like if you're on your bank, this is a very personal experience. You log in, you see your name, welcome Mindy, here's your bank accounts, here's your balances. That's an extremely personalized experience. And then the upper end of the spectrum is where we have predictive intelligence, where we're using personalization for anonymous and authenticated users. Some examples that I've heard of for this, that like a higher education institute once your college student logs in and they start to detect that they're not doing so great in their degree programs and they fine tune the messaging to help drive them to another program that they may be tested really well on in order to get student retention rates increased. So there's a wide array. Today we're gonna be focusing more in the middle of this spectrum, talking about relationships, customization, behavioral and demographic data. So as a user experience director, I'm really an evangelist for user experience because I know that it works. In combining that with marketing, we really see the success as you're building those personalization strategies. So how we're gonna do that is we're gonna plan the path of the user through the journey and we're gonna align the messaging of all the different channels and then we wanna use technology seamlessly so for ensuring that best user experience. So a traditional user journey and traditional user experience is just thought of as a user moving through a single system. So in this example, you have a user moving through a single website. They might visit a homepage, they're gonna browse some products, they might select a product to view and then eventually add it to their cart. They may not buy this product but this is a simple journey through a simple system. A multi-channel marketing user journey, this is a bit of a more robust journey where the user may travel across multiple properties. One might be an inbound or an outbound email that the user receives from your organization and there's an ad or there's a Facebook link that they go and go to Facebook, browse your company, look at some events, go to another channel such as Zoom and register for a webinar which then redirects them to your website. Well, what is that experience like? Was it consistent for them? Was the messaging aligned? Was your brand even consistent? So that kind of thought goes into the marketing strategies and the journeys that we create. So it's this thought we wanna take in to personalization. So what's in your content strategy for personalization? So in this example of a user journey, we have a healthcare tech worker and let's say they're browsing social media, they're on LinkedIn and then there's a targeted ad shown to them that you then put in place for technical users of healthcare technology. That user would click the ad, healthcare focused and then land on your website and let's say you sell software in all different industries. So you have one landing page. Sign up for a software demo, that's what they see. Well, that's not relevant to their journey that they've taken in the steps prior to arriving to your website and they lose interest and they move on. So therefore that user didn't convert. You may have lost an opportunity to drive them into your lead funnel or potentially convert them to a customer. In this example where we've aligned that content from their previous journey into the LinkedIn ad, to the targeted to your website. In this case, the UTM here is passed from their ad to your website indicating the behavior and the attribution where they were prior to your website. With that knowledge, the site then serves them up a different headline, not a generic demo, sign up for a demo, it says solutions that help provide better care. Get a demo of our healthcare software. That is much more compelling, much more aligned with my content. And this is something we have seen creating more engagement on the website. They may start to then explore the site further and they may also go ahead and exchange that valuable personal data that they have to sign up for that demo. Here's some examples, we did it at elevated third. On the left you'll see just our Drupal standard web development landing page. On the right we have actually detected that these people came from a targeted campaign for the SAS vertical. We've given them a new hero background image, new hero messaging, a new call to action to get them to register for a webinar of interest. We've also changed out some of the images on the homepage with the technologies that they probably are using. Here's another example where we did a stack specific. We know that these people on the left, these are anonymous users, we don't have any, it's a default view on the right. We do know that they were either at a campaign or came to our site through a channel that was more stack specific. So we know these audience members are looking at a specific marketing stack and we've put the commonly used market tools that go with that stack here on this page. So ways to get started with developing your personalization strategy, you wanna define your audience segments that you wanna target first. Where's the most results gonna be? What are your organizational objectives driving you at that moment? Then you wanna understand the people, patterns and problems within those segments and that way you can offer personalized content that better meets their needs. So personalization uses segments of user groups to target with scenarios. It's gonna be too tedious to go look at every single individual and create a new messaging scenario for them. So you wanna collapse them into things like user groups first. So you look at your user groups. These, in this particular example, you have customers and prospective customers. Maybe two very different groups, partners and students, looking for very different things on your website and job applicants. Those are some potential user groups we see a lot of people use. Then you can take that user group and then define it further into a segment such as their geographic data. If they're in Atlanta, Georgia or if they're in India they might be looking for something very different from you. If you can also look at demographic information such as their age range, their role, their organization. Firmographic is actually information about the company they work for. How big is the company, what is the revenue? Behavioral information is what are they doing on your site? Where have they gone before? What are they doing while they're there? And then psychographic information is what's really key also to the content strategy being successful. Psychographic information is more around what are their motivations and what are their goals and what are their problems so your solutions can better align to their needs and be more compelling and your site can be more relevant. One way we get down to the psychographic level of information is through personas. And personas are a characterization of the target users of your website. They're a little bit stereotypical at some times but often we arrive to these through research where we interview multiple people and then collapse them into similar stereotypes. Often you can just do a lean persona exercise within your organization with assumptions and things you already know about who your target users are. In this case we have two students and you could just say we have a user group of students make a new segment of prospective students but then you have the undergraduate and grad student and these two personas that have very different motivations and undergraduate students looking for a great culture fit in this particular example and then the graduate student is just looking for relationships and networks for their future career. So you would be giving these two potential prospective students different information to help drive them into registering for your school, finding out for more information and eventually applying and signing in. So that's how personas can help influence that psychographic information once you understand and build empathy for those people ultimately. We're seeing a shift in overall marketing strategies from topic-driven to persona-driven marketing because it's just finding it's more successful. So the focus on the individual psychology of those key buyers and decision makers, their needs, desires and interests. B2B businesses realize these more personalized strategies are actually winning business because they're making those human connections. It's generating more effective ads, more effective content, events and you're actually giving the people the resources they need to make better decisions and by providing that value, they're gonna come to you more frequently and give you that valuable data for their personal information as well as their business ultimately. So now we're gonna talk about the buyer's journey. This is another layer to your personalization strategy we talked about, your personas, the individuals that are coming to your site. Another layer we're gonna put on here is the buyer's journey. So if you're familiar with it, the awareness, consideration, comparison, decision phases, what most people go through when they make any decision from what shoes to buy to who to do business with or even if they're gonna decide to download a white paper. So if someone's in the awareness phase, they're likely coming to your site for the very first time, their first time visitor and you probably have some content you wanna drive to that first time visitor. So like such as about us or a white paper about what your solutions provide. So you can easily detect a first time visitor on your site and ensure that information is first and foremost. But as you go through the different levels, you might think about what content do you have that serve someone who's in consideration or a comparison phase. If someone's on your site and they're at that point where they actually leave your site, they don't convert. They might just be leaving to go do some comparison on competitors, see what other people have to offer before they are investing in your site giving you personal data to either be contacted to sign up for a demo. So but they might, cause it might come back. So the buyer's journey is not a straight line. They may come back to eventually give you their information, sign up for more information, learn about your company or purchase a product. In this case, they may come back several times and do that conversion many times. They might sign up to learn more. They might sign up for a white paper before they ultimately become a customer. So this is a great opportunity for you to remember your site visitors. When they're back on your site and they're ready to continue their journey with your organization, you can greet them with the content that may be more relevant to help drive them further into your sales funnel. So you can then hopefully give them the information that's gonna answer their questions and the resources to make that decision, that buying decision. But you have to be cautious. You've heard, we don't wanna greet them with their first name even if you might have it. We don't wanna let them know, hey, we've been watching you your back finally after 10 days, what took you so long to call? We're not doing that, but we wanna play it easy. You wanna lay back, hey, we get you, we got what you need. Just sort of relaxed. Unless this person has already opted in, they've given you consent to greet the personalization. They've authenticated on your site. In those cases, it's completely appropriate to use their name and that personal data. So what we wanna do, as I mentioned, is casually align your message. And here's how those couple of different tactics of personalization can be combined into an example of this dynamic message scenario. If you're doing a workshop on personalization in your organization, this is a good place to start with a personalization matrix. Just draw a big square on your whiteboard and identify your top personas, your top segments. So for this particular example, this is a healthcare technology company that we're targeting and we have two types of people. The HR manager who's needing the software to help bring their teams closer together and communicate better. And then the financial officer who's really concerned about risk and budget and cost. Even though they're at the same company, even though they're probably colleagues with each other, they still have a different motivation and a different need. So the top layer is their first-time visit, the layer underneath, and I believe that is their second-time visit. So in the middle, you'll see the default landing page, learn more, sign up for a demo, you'll see that over and over. But this person comes through a campaign that we know is through LinkedIn, for example, is a great example, because you can target those particular roles when they visit your site. They can bring the UTM over and you know who they are. And they're a particular persona. So these HR managers headline might say, bring remote teams together, sample our HR software that helps your team work smarter, their second visit, ready yet, we can't wait to get started, see how our software unites teams. So it's a little more intimate, a little more familiar without being overt and without using too much information about them. And the financial officer, same software, but same demo page, the dynamic content replaced with software that saves you money. See how your bottom line can grow and try a free demo. Ooh, free, they like all those sounds, right? So maybe that's gonna help that persona convert and potentially invest with their data and give you more information. The second time that financial officer visits software costs keeping you up at night, they're obviously still thinking about this, they've looked at this page twice, they still haven't given you their information. What are you waiting for? Sign up now. So that's one potential way you can use a content strategy within your personalization to just get started. So that's a very simple, straightforward scenario that we just layered in a couple of different layers there. But your website can be needing to meet a lot of different needs of your marketing organization. There's likely more than one driver, such as this LinkedIn ad that I've used from these examples, you probably have sales activity and emails going out, you could be using ABM strategies, you have SEO influence for marketing, just a plethora of things, driving visits to your site and your site would have a hard time doing that without the help of additional technologies and platforms to really stand up to all that traffic with relevancy and personalization for those visitors. So it really could be overwhelming if you try to tackle it all at once. So don't try and boil the ocean. Just start with a small pilot, like we showed. You wanna align your messaging on outbound emails, ads and landing page content so there's consistency, what is their original intent and make sure you carry that through to your website. Layer on with repeat visitors to help drive that funnel and bring people closer to conversion and then you wanna measure and refine what's working, what's not working, change the messaging, change the ad targeting until you start to see that pilot really successful and then once that's good, rinse and repeat. Try it again, try a different segment, try a different landing page content that can be revised. And so that's the user experience portion of that. One piece of this, like we mentioned early on is using technology seamlessly and we have Grunder here who's implemented several personalizations with our clients and he's gonna tell us about how that works. Yes, so before we get started with actually what technologies you have available to you, let's look at just a high level overview of how things work. First of all, of course, like Mindy said, you want to segment your users correctly, right? And what kind of data you use that, you have a lot of options, we'll cover that in a bit. Secondly, to begin with at least, you want to define all your content variations in the form of simple if then rules. That's how we do most of our implementation for personalization projects. And then you can kind of layer on more technologies and more sophisticated technologies to make it more intelligent, right? So now let's look at some of the options of how we can do this. Think back to that personalization spectrum. In the beginning of that, you kind of saw relationships, relationship-based and customization, right? So you can start with most sites already have some kind of content tagging in place. You can use that to kind of give users more personalized experiences. And for that, you don't even need any other technology besides Drupal, right? Drupal already has a very robust taxonomy system. You can build views for related content. So in this project, the screenshot you can see on the right, we actually built a little content explorer block for users that the content admins can place on a bunch of pages. And then we gave them the option to filter by topics. And once somebody selected the topic, it kind of followed them around the site. And when they are browsing for these other resources, white papers, blogs, we pre-filter some of that content. So that's already personalization without any kind of fancy AI, any technology leared on top of what Drupal already gives you. So that's something you could start right away. So moving on, talking about those if-then rules, some of the more behavioral demographic personalization that we saw on the spectrum, we use the smart content module that we actually built in-house. It's just a Drupal module that you can install and start using out of the box. This just uses if-then rules. So if a user is in a specific segment or if such and such conditions are met, show them this content variation. So it's conceptually really simple. There's a link to the module down there. And I'd actually like to spend a little bit of time talking about the little video that you see on the right. So that was for a project for a client that does fleet management software for trucks. And we decided that the segments made sense, that made sense where we would like to segment users based on a first-time visit or a return visit. We also decided to segment users on their fleet size where we had a little micro interaction where users could specify what fleet size they have, their company has, and then based on that we would show them different content, different products. And then we also decided to segment users based on their location. So if they're in the US, certain products were more popular. If they were in Mexico, certain other products were more popular. So what we recommended changed based on their location. So those were the segments. We helped them write out all the content for how to approach each of those segments. And then we implemented that using smart content. So moving on, the data that you can use to segment users, you have a lot of options. The browser list that you can see, like the language, the operating system, cookies, query parameters, those are available out of the box with smart content so that you can use right away. We have in the past implemented more data points such as just where user self-id so where they're viewing certain pages, you can keep track of that. If they're clicking certain CTAs, right? Like for example, if somebody clicks on the login button, you could assume that they're a returning visitor or they're an existing customer. So that can be your segment, right? Any form fills, any other interactions, like I mentioned in the last slide, we had like a micro, I guess, slider to indicate fleet size for users. So stuff like that, you could also easily implement. And then you also have a lot of options for third-party data, which we'll get into shortly. But smart content is super flexible. It's very extendable and it has, it actually has integrations already with a lot of third-party options where you can get like formographic data, for example. So if a user is visiting your site, you could actually get information about their company, their industry, what their company's employee count is, what their company's, like revenue ranges, a lot of options. So a couple of those examples, one of them is IP stack that we've used frequently. And there is also a similar platform called IP Info. Both of those have integration modules with smart content on Drupal.org already. You can find them on the smart content project page. They're also linked at the bottom of this deck. And you can look at that later. But IP stack and IP Info take the user's IP address and give you some regional information about them, what country, what state, what city. So if you want any kind of location-based personalization, these would be good options. Do keep in mind that this isn't always accurate. A lot of people use corporate VPNs. So the IP address might not give you correct information. Some other options, that give you the formographic data that I was talking about, a user's company information. Demand base is one of the options. There's another one called six cents. Both of those also already have integrations with smart content on Drupal.org. So it's just a matter of installing modules, configuring it with some API keys, and you should be good to go. And when these platforms will give you options like company, employee count, revenue size, all of that information. Now one thing that we must talk about is analytics. I'll be talking about them in context of smart content and how we did it, but regardless of your tool of choice, your technology of choice, this is very important because like Mindy said, you want to measure and refine. You want to see if your personalization is working or not. So analytics is something you should think of early on while you're developing your personalization strategy. So in this case, there is another companion module to smart content that pushes information into the data layer that can be picked up by a tag manager and then we push that to Google Analytics, right? And then once it's on Google Analytics, we can kind of track as to what the success was. So in this particular screenshot for this project, which was I think Comcast Technology Solutions, we defined conversions or success as users filling out forms. And when users fill out forms, the way smart content pushes data in, we have the option of filtering by segments. So like which user was in which segment and out of all of the users that converted, how many you can filter by segment. So like who was in, how many people were in each segment. So that kind of gives you insights on how your personalization is working. Beyond that, we also have filtering options for the company which you can see on this screenshot. And then also the specific campaigns that they used. So who was in which campaign? How did that perform? Now I'm not an analytics expert, but if you are interested to know more about how this was done, what kind of analytics setup they had, what was the implementation like and what kind of success they saw. The Digital Marketing Director of this line, Comcast Technology Solutions is actually joining Mindy for a fireside chat tomorrow morning at 8.30 in Room B116. So definitely attend that if you have any questions, if you're curious about that. You can go there and ask first hand. This one, not related to smart content, but this is at the far end of the spectrum where we said predictive technologies, right? This is a screenshot from Amazon Personalize. I haven't had a chance to use it, but if any of you out there are thinking about this, please let me know. I would love to use it. This one is a machine learning real time recommender engine. So you can actually use this to build something like Amazon's frequently bought together feature where users are like viewing certain types of content and then based on their choices, this would recommend other content that would be relevant to them. I think it just takes, you do have to provide a data. It won't just work on its own. You have to give it a pretty big data set either at the beginning or over time and then your recommendations would get more accurate. No session about personalization would be complete without talking about privacy and ethics, right? And a lot of marketers have questions all the time about like cookies, is this PII? What about GDPR? So it is important to know that personalization and privacy are not mutually exclusive. You can be respectful of both and still achieve success in your personalization strategies, right? There have been other sessions done where there are other resources. There's a lot of content out there already that has covered the privacy and ethical aspects of personalization. So we won't go into too much detail for that in this session but all of those basically just boiled down to make it user centered. Let your users be the owners of their own data. Let your users control their personalization journeys, right? And there is actually, I was reading a study that said 80% of all consumers are more likely to make a purchase if your brand offers more personalized experiences. And I've linked that study in the end of this tech as well. And they also interestingly talked about a paradox where users don't want to share their data because they don't trust the companies to use it ethically. But at the same time, they also want more personalized experiences, right? And I'm one of those. I'm on Instagram. And you know how Apple now asks you, like ask app not to track or allow it to track? I always say no because I don't trust companies. But then I'm on Instagram and I get those personalized ads, right? Toys for my cat, love it. And I'm buying off of these personalized ads. That's where all my shopping happens now. Yeah, and I get ads on Instagram for all these exotic snacks, which I love. So the targeting is good. It's working. And I do buy that. And the next strategy is for, this is the shirt that fat people used to cover up their bellies. So it's very targeted, but anyway. So it does work. And I'm one of those users that have that paradox. And a lot, like I said, the other resources out there who'll tell you to, earn the user's trust, be transparent. I think that does help a lot compared to just ask app not to track or allow them to track. But even without that, there is a lot you can still do with activated anonymized data. Like I said, the smart content module and the data options we saw, there is a lot you can already do with just anonymous data. You don't have to ask them for any information. Just like there's some query parameters, like UTM query parameters, that you're adding to your links in your marketing. Like all of that already helps you. There's a lot you can do with it already. So you don't even need to ask for PII for personally identifiable data a lot of the times. So quick recap before we get into questions. Like Mindy said, always consider the UX when you're thinking about personalization. Always have your strategy ready. Always have your content ready before you start thinking about the technology. Because as we've seen, the technologies are there. They're sophisticated, they work. What is missing is the strategy and the content. And then you can start using technology to kind of enhance it, to make it better, of course, ethically. And then also have a plan for your analytics for how to measure it, for how to refine it. There's the link to the smart content module, like I said, and also a link to the slides. The team behind smart content and the team that does the personalization strategy at Elevated Third are all here at DrupalCon. So you can come see us at our booth. We don't have a demo for any of this for smart content plan for this session because there wasn't enough time. But if you're interested, please do stop by our booth and let us know. Booth 119, you'll see us right as you enter Exhibition Hall B, we're right there. And the session that I mentioned for tomorrow, which was a case study for the implementation of smart content and our personalization strategy, there is a fireside chat tomorrow at 8.30 that you can join. And with that, are there any questions or ideas or opinions? We'd love to hear, oh yeah, of course. So for the data being stored, we use, most of the data is available either in the browser. It's all client-side still. So for example, like demand base or IP info, it's all happening in JavaScript on the client-side. For the decision-making, that is all also happening client-side. The only thing that happens server-side is the definition of those segments and of those if-then rules. But the rest is happening client-side. Go ahead. I've heard of the name, I don't think we've used it. Have we, Mindy? Okay, yeah, we have not used that. But I've heard the name though. There was another question over there. Anybody? How do we load in the content, you mean? Oh, do you want to answer that one? I think the question was how do you proof the content after it's been loaded in for the scenarios? It's actually within the CMS, so you can preview it within the interface of the Drupal admin experience. Oh, so you want to preview the content as different segments? Okay, so we actually didn't have a way to do that until recently, so we were just spoofed like having a cookie in the browser or having like the UTM parameter. I think there was recently a contributed Drupal module that lets you preview smart content segments by marking each of them as true. I don't remember the name of that module, Mike, do you? Smart content preview. So that's the module to use. Any other questions? Oh, in the back. Yeah, so most times what we do is once the personas have been developed, we'll think about more concrete ways to classify users into personas. Like the example I gave you, for example, if you want to segment users as new and returning customers, what we used was clicking on the login button. So if somebody clicked on the login button, we just assumed that they're a returning customer. So that's how we define it, and then once you have that information, like if somebody's clicked the CTA, or if somebody has viewed the specific page, those can be built as conditions in smart content, so that's how we build out the segments. But we do have to decide what the trigger is going to be that puts a user into one of these segments. Does that answer your question? Well, this module is one part of implementing a personalization strategy. So if your personalization strategy includes just showing different content variations based on some simple rules, just like if then kind of a logic, that's when you would use this module. But like I said, the strategy itself has that whole spectrum that Mindy showed, right? So starting from just like tagging and taxonomy all the way up to predictive. So this is just one part of the personalization strategy. Were there any other questions? Yes, we did sort it in the cookie, yeah. Any other questions? I think we're almost at time too. But were there any other questions? Okay, if you think of anything else afterwards, do come and visit us at our booth. Thank you everybody.