Drive Visitor Engagement with a Personalized Web Experience


Jason Beyer and Doug Feit explain how to optimize real time consumer interactions on your website.



Hi and welcome everyone. Thanks for joining us for Drive Visitor Engagement with a Personalized Web Experience webinar. My name is Jason Beyer, director of ElementOne Consulting at TARGUSinfo*. And I'm joined by Doug Feit, executive director of market development also from TARGUSinfo*. This is part four of our ongoing ElementOne one webinar series. We appreciate you joining us. As most of you may agree, your website is one of your most important marketing channels for your business. Today's webinar will focus on how better understanding the distinct audiences that are visiting your site. So you can create more personalized web experience that will improve site engagement and ultimately result in more conversions or sales. Before we begin our webinar, I'd first like to go over some house rules for you. Your phones are automatically muted. But please feel free to submit any questions you may have in your chat browser, and we'll answer them during the webinar. Also, if you have ideas for future topics or suggestions, please feel free to forward them on to us, and we'll follow up with you directly. We're expecting a decent turnout this afternoon. And I notice that about half of you that are joining today are current clients at TARGUSinfo* and familiar with us from having attended past webinars and conferences.

There are some of you who may not be as familiar with us and have requested a little bit more background information on TARGUSinfo* and our ElementOne product. So as a brief introduction, we're a $150 million information services and analytics division of Neustar, which is a $750 million public company. We were acquired by Neustar in November, 2011. We are headquartered here in McLean, Virginia, mainly focused on large US brands to help improve marketing insights and performance, whether it's over the phone, at the point of sale, or online. ElementOne is our industry-leading segmentation and marketing analytics solution. So some brief background on the company and what we do. Now, I'd like to begin our Drive Visitor Engagement with the Personalized Web Experience webinar. And I'll turn it over to Doug, who will take it from here.

Thank you very much, Jason. As Jason introduced us, we're going to take a look at one of our most recent ways that we're able to bring the ElementOne analytics segmentation platform to life and one of the direct applications that can help a marketer engage their customers and prospects better. So with that, we'll go ahead and dive into the webinar. So most of us have heard the old adage that advertising is content. Truly successful marketing campaigns are more than just messaging and product. They are living, breathing content experiences. Two great example of that we'll go through real quickly.

The first one is the Swagger Wagon Toyota campaign. I'll spare everybody the funny commercials and videos that go with it. If you haven't checked it out, I encourage you to search on YouTube for the videos. But they're very well done and engaging commercials that live and breath and feel like mini TV shows, if you will, or digital shorts. And going back a few more years, there's the famous Apple "I'm a Mac, I'm a PC" campaign. Again, folks tend to remember these for the content and the engaging experience as opposed to simply just the products. So it becomes clear that truly impactful marketing is more than just a product, a message. They are engaging experiences, engaging content experiences, content experiences, which frankly are targeted towards distinct audiences.

So in are two examples, that Swagger Wagon is really geared towards the suburban family, call it middle aged or young to middle aged family, probably two kids in the household, middle to upper income, probably live in the suburbs, but someone that's going to relate to a funny, here's how I deal with my kids in my van, and I'm the best cupcake chef in the neighborhood mentality that that campaign brought out. The other side of that is the, I'm a Mac, I'm a PC campaign, which is geared towards the college aged start-up, probably someone who's single and or married with no kids. Again, relative to their peers, probably a middle to upper income person. And again, in an urban or college setting. Both of these campaigns are incredibly successful. Toyota, I believe, is the number one selling vehicle in its class. If not, it's certainly in the top two or three. Macs, certainly in the early 2000s when that campaign came around, were increasingly popular on college campuses, tech startups.

And even today, while Macs have become more prevalent in the workplace, if you go to your average large company you're going to see a lot more PCs than Macs. The campaigns simply were effective because they reached the right audience. So what about on a marketer's owned and operated website? Are the same capabilities there to allow you to identify the distinct audiences and then tailor a message to them? The predominant solution to the market for website optimization today are what we'll call A/B multivariate and or behavioral testing solutions. These solutions are valuable. There's no doubt they add some value. But ultimately, what they're going to do is determine the best single message that should reside on your website.

They're not going to tailor that message to each of the distinctive discrete audiences that visit your site. And by virtue of offline and online channels not really connecting as well as, frankly, they can and should, oftentimes what you're doing on your website with one of these tools is siloed from your offline channels and the segmentation scheme that a marketer might spend millions of dollars on. That simply isn't surfaced on the website. There are also some third party data providers out there-- TARGUSinfo* being one of them-- who can help, although other players frankly aren't focused on the website optimization space. Most of them understandably are concentrating on the third party advertising or display world. They lack the scale and the authoritative offline data to really be impactful on a site. And again, there's an issue with not being synchronized with offline channels. So a couple of quick stories, especially about the A/B multivariate and behavioral targeting.

I was sitting with a client who told us that we don't need anything beyond behavioral targeting because we're trying to reach 18 to 24-year-old women for our product. While that is a factual statement that those behavioral targeting solutions can help you reach that age and gender, what you're missing there is, does this person live within a $200,000 a year income house or a $50,000 a year income home? What is our capability to buy our high end products or our lower end product? There's that additive bit of intelligence that data can add to the story. So what this results in is frankly a sub-optimal consumer website experience.

Websites are an incredibly effective communication channel. It can give you the ability to communicate on a one to one basis with your customers and prospects. Fortune 500 and Fortune 1,000 companies are investing millions upon millions of dollars in the channel from content to buyflow to the branding that prevalent on the site. And just the notion of digital marketing, which consists of search, display, and another things is a $32 billion industry. And most of that, frankly, is marketers sending traffic back to their site.

However, when somebody lands on the site, most website owners are under-investing in optimization and personalization. They're not creating what I'll loosely deem audience targeting solutions. They're not identifying the discrete audiences that visit their site, and they're not able to combine, again, that siloed data that sits in the offline channels doesn't come to life in the online channel. The result is millions and millions of, frankly, wasted consumer interaction. Online, if you're converting anywhere from 2% to 10% of the traffic that comes to your site, you're probably do a good job relative to your peers in other organizations. But that 90% to 98% of your interactions that are wasted and or sub-optimal-- shallow business, folks who come visit one site and leave, as opposed to viewing many products or learning more about the products and services that a marketer has to offer.

The most alarming thing, I think, is the fact that there is a high margin opportunity that's being left on the table. What we're saying is, you don't have to spend a bunch more money to send traffic to your website to convert more sales, which is often the way marketers try to increase sales. There are solutions in the market with the help of segmentation platforms like ElementOne that can help you make more of the visits you have today with nominal investment, which will result in more sales, again, for little to a nominal investment, and frankly, sound margin and revenue for your organization. So with all of that background, that's why we created our PageAdvisor solution.

The tagline for the product is, fueling website optimization. We are not a website optimization product. We integrate with and work with the solutions in the market that do that-- the Adobe slash Omniture's Test and Target, Optimizely. There are dozens of solutions in the market. Those are an important part of the ecosystem. And our product works with them, does not replace them. It provides the fuel, the data to make those services better. Underlying what we do with the PageAdvisor solution is our ElementOne platform.

That's the theme for all the webinars we're going through. And what we can do-- and we'll get into this in a little bit more detail as we go through-- is we can use actual conversion and customer data to analyze the types of customers that convert for different types of products and then use that information to help determine which offers and products an anonymous visitor should see on the website.

And the other upside to using a product like ElementOne is the ability to then synchronize that with other channels. If we know that people who have a certain set of characteristics are going to buy a certain product on a website, the ElementOne platform has the ability to be replicated, instantiated throughout other channels, be it your call center, direct mail, frankly, even potentially in store, if somehow happens to be operating brick and mortar stores.

All that is a really great extension of the TARGUSinfo* mission to optimize real time consumer interaction across all of those channels. So how does the ElementOne segmentation platform play in? So tend to start any engagements with PageAdvisor by doing an ElementOne segmentation, getting a set of customer data from a marketer. That is then loaded into our ElementOne segmentation platform, which has a view or segmentation that buckets every US household into one of 172 distinct elements.

The elements six demographic elements as their building blocks-- things like family income, age of the head of household, are there children in a home, do you own a home, urbanicity, things of that nature. And then based upon the data that a client brings to the table for PageAdvisor, we can then bucket those distinct elements into an appropriate number of tactical segments. In other words, this group of elements index high toward propensity to buying a high end laptop versus a value desktop, or a gaming PC versus a home slash media PC.

The level or group number of groupings or number of tactical segments is a custom number set, depending upon each client's needs. And the rules of thumb I like to use here is to look at the top line navigation one has on their website. Do you send folks down four to five different product sets? We can very easily help you figure out which of those top line navigations is the right path to send a given user. Again, the idea is increased page views, product views, page interaction, and then ultimately lead to more conversion on the site. So there are three main ways that we can put it into action on the site. We'll loosely call them use cases. The first is to personalize the content experience.

This works very well for folks in an interactive marketing world who are setting up websites, landing pages, what have you in an effort to generate leads for another party. So what you're actually seeing here is an example of what we've done for one of our clients in a financial services slash mortgage lead generation world. In this world, it's a bit of a unique world, because you don't want to discount any visit to your site. Every site is valuable. But what you do want to do is put your users and visitors into a smart set of buckets and deliver a relevant content experience around that. It may all be for the same product. You might be ultimately trying to get this person to enter their information or submit a mortgage lead. However, the wrapped that's around that can and should be customized to each distinct audience that visits the site.

Therefore, folks will be more likely to enter that lead. In this particular example, we got six months worth of converted leads for the mortgage lead generator and put them into the six buckets you see to the left there. The chart relatively shows you how each segment relatively indexed towards visiting the site. So to call out that leftmost example, we bucketed these folks into what we call the middle affluent families segment. So these tend to visit more than the other segments relative to their population within the US. And we can learn a lot from them by this group. The bottom image there is our audience details, if you will.

But what you have here is a middle aged head of household. They're likely to have children in the home, and they are relatively high income. So what we should do is show this person the family or dream home imagery, as opposed to the next segment, which is the far right. These folks tend to visit less relative to their population in the US. But they're still valuable. You still want to show them an optimized content experience. Here they tend to be of a retirement age. Age 71 or so is the average age. The index very low toward having children in the home. They are relatively low income on what we call fixed income. But they do tend to be homeowner. So here we might show somebody the retiree imagery and talk about a reverse mortgage as opposed to the first segment, where for this particular market, we might talk about refinancing their mortgage, or a new mortgage for your dream home.

Again, the idea is identify each of those distinct audiences. Don't under-serve any audience that comes to your site. You still want to show them the appropriate imagery. And then ultimately, increase that conversion. So the second use case beyond just rendering the content that is shown somewhere on a website is actually varying the product or the product path that someone goes down. So for this example, we work with one of the leading PC manufacturers. And what they were looking for is an optimal way to segment their users based upon the product or product series that they may purchase.

Oftentimes, we find that going down to the model level is just way too granular. However, bucketing somebody into looking for, again, a high end laptop, a budget PC, or a gaming machine can be really impactful on the site. For this particular example, we looked at 2 million existing customer conversions on the website, loaded them into the ElementOne platform. And we're able to see some very interesting discrimination amongst the products. So what you see in front of you now is an index bar chart of the five main audiences that come to this particular marketer's site, bucketed into young upstart families, middle single, affluent homeowners, middle income families, and again, there's golden year retirees.

The green bar are the high end laptops, blue is the value desktop, and white is a gaming machine. What we see, probably not surprisingly-- and this is certainly consistent with what the marketing group at this particular client was expecting-- but we see that young start families tend to be in their 30s to 40s make a little bit more money. They're more interested in that high end laptop. The power, the sexiness of the machine is important to them, and they're willing to spend a little bit more money. So they index quite highly towards buying those types of products. The more middle income, however single folks, that tends to be your gaming machine audience.

They may not make that much money, but they probably don't have kids. They're probably not married. And they're willing to spend a little bit of extra money on the high end gaming machine, because that's the use that they're looking for for the machine. Then we've got an affluent homeowner. As you might expect, they under-index towards that white bar or the gaming machine, but they are indexing relatively high in the high end laptop and the value desktop. The middle income families, again, not too surprisingly, there's not too much discrimination there. Not every audience you're going to see a high index for a given product. But again, that's fine.

These are the type of insights we're trying to glean. And then for that golden years, fixed income retiree, they're much more cost sensitive. They're probably not quite as tech savvy. And frankly, they just want a really reliable, usable machine, don't need all the bells and whistles. And they tend to buy that value so what we can do with this type of intelligence is create a segmentation scheme that when an anonymous user comes to marketer's site and they fall into one of the elements for young upstart families, we should show them the high end laptop. But when a golden years visitor shows up, show them the value desktop. Instead of showing them a Chinese food menu of products that you can buy, show them the one product that they index higher towards, increase engagement, increase conversion on the site.

And the last of the three use cases is it provides a much higher fidelity segmentation and analytics view to the types of visitors to your site. So what you're seeing here is standard web analytics from Quantcast for one of our clients. And what you see is the typical panel-based view of what web analytics metrics can provide. Specifically, relative to the household income, we can see that $50,000 to $100,000 a year families or visitors over-index towards coming to this site by the second row in the household income. But what you're lacking here is granularity into those segments. And frankly, this isn't actionable in any way.

Web analytics are great for figuring out the types of audience that visit your site. But the shortcoming of a panel-based solution is that you cannot put that in action on the website. We've already talked a little bit about how we put those things into action. But what's interesting to call out here is the ability to take a deeper, more granular look as to those segments. So the same website we capture data on for some period of time, loaded each segment into our ElementOne platform, and what you see in that red box is the set of segments that are $50,000 to $100,000 household income segments. Some of them, the bars that go above the x-axis index rather high towards visiting the site. However, some of them, frankly under-index towards coming to the site. So that ability to see that granular level of information shows you that somewhere around 32% of those segments would actually under-index for converting. That $50,000 to $100,000 year home over-indexed towards converting towards a lead or the product for this website. However, with this additional level of granularity, we can remove 32% of those folks out in that bucket and show them the type of product that they would be more likely to purchase.

In conclusion, again, the idea here is to bring data, authoritative demographic data, to a solution where it's not today. Website optimization is based largely on what we bucket as behavioral information-- which types of sites has somebody visited and or which types of products have people who have clicked on this page on our website before been likely to convert? What we're adding to that here is an ability to know authoritative demographic information about an anonymous visitor to your website and bring that to life as far as which product, which creative, and frankly, which segment somebody falls into in a much more granular level. The end goal is increasing customer interaction, clicks, pages viewed, products researched, and then ultimately, increasing conversion on that site via a lead generator, an e-commerce provider, or a marketer looking for any sort of a registration. So with that, we'll ask our moderator if we have some questions from the group. Yes, Doug, we do a couple questions here. And I encourage anybody out there listening that has additional questions, please submit those. And we'll get to those.

The first one here is, can you describe how the process works? Can you start from once a visitor lands on my website? Go from there. Sure. The short answer to that is-- not knowing everyone's technical understanding of how web or digital data is integrated and transferred across the ecosystem. The short answer is, it's integrated with a very simple standard call to the TARGUSinfo* domain. And how that works on the site is, a visitor lands on a given site. Today, in most cases, a website optimization service is present on that site. So new visitor comes to the site, the website calls to the website organization service and says, what creative, what offer, what should I show this person? When PageAdvisor and TARGUSinfo* are integrated there, that website optimization service makes a second call to the TARGUSinfo* domain and says, what information do you know about this person? We return our ElementOne segment code, which has all of that rich demographic and psychographic behavioral data behind it. That code then is entered into the optimization solution. And those optimization solutions can make rules-based decisions based upon those. We set an ElementOne analysis that says, segment 12 should see the high-end laptops. And that optimization service can make that decision based upon that and render that image as it is appropriate to the type of person that is visiting the site.

And what type of engagement or conversion improvements can website operators, marketers expect from utilizing this solution? This is great unanswerable question, for the most part. It's dependent upon what a given marketer might be doing today to optimize their experience. I've seen increases anywhere from 10% to 70% or 80% in conversion rates, depending, again, on what the quote unquote status quo is. In most environments, I can pretty confidently tell you, you can see 20% to 30% improvements in both interaction and close rates relative to a one size fits all approach. And again, one of the challenges here is creating the proper test bed to isolate these effects. It is definitely a challenge and something that needs to be worked through. But in general, you can see very high increases in interaction and conversion rates.

OK, great. Looks like those are the only questions we had for Doug here. So I'd like to thank everyone for joining us for this Drive Visitor Engagement with a Personalized Web Experience webinar. We hope you got some great ideas on how to make your website more powerful by delivering personalized website messages and creative. This can lead to increasing your visitor interactions, page views, stickiness, and ultimately, what we're all looking for is boosting revenue with higher sales and more conversions.

As I mentioned at the beginning of the webinar, we welcome your feedback, and if there's a topic that you're particularly interested in, we'd be happy to put something together for you. Don't hesitate to reach out to either Doug or myself at any time. And with that, that concludes our webinar. And I really appreciate everybody joining us. Thanks a lot.


*TARGUSinfo was acquired by Neustar in Nov. 2011.

Related Resources