Webinar: Paul McConville and Jason Beyer introduce ways to increase online lead generation success in multiple channels.
Thanks a lot for everyone for joining us here today. Just want to level set on some expectations for this particular webinar. It really came about because we're going to be out at LeadsCon in a couple weeks talking about how to optimize and use analytics to optimize leads across all different marketing channels and understanding and recognizing that today, lead generation can take a lot of paths. And there's a lot of opportunities to drive inbound leads through advertising and communication efforts. So how can we as TARGUSinfo* help our clients better understand who their audiences are in terms of best in performance from a lead and wanting to go out and fine tune and target those leads and drive more of them in through various marketing channels? So the first thing here is media optimization strategies and tools for making those very difficult media buying decisions.
We're hoping to address a couple of key questions around cross channel convergence, the first which that starting point of how do I segment my leads into audience groups based on their performance, based on those audiences that are most likely to be your best leads. What will help me understand which channel is doing the best job at driving those leads in so you can invest your marketing dollars in those channels and deprioritize those that aren't working that well. And then once you understand that-- OK, where do I go out and find that best audience in that channel? Is there particular programming, or schedules, or sights I should use in my targeting efforts? And frankly, which locations or parts of the country am I drawing those best leads from? Those are all things that we'll explore in addition to once we know where to find them, what's the best, most relevant message that we can deliver to those leads to drive them into our channel and then the media plan necessary to go out and get them. So we're going to take steps through our element one analytics platform, which I'll reference in a use case in doing this.
And then we're going to wrap up with an actual client of ours, an EDU Lead Gen use case that's taken this platform, implemented within their business, and seen some great results. So let's go ahead and get started here. But first wanting to talk about generally speaking how do we use segmentation? And what is segmentation? And Paul noted that a lot of this industry is using scoring to prioritize leads. What segmentation does is allows you to put a face and attributes associated with those scores.
And so segmentation models are really designed to assign leads or cluster them based on common attributes that we can use to describe them. So we'll look for behaviors around their likelihood to be a lead, their value as a lead, and then certain demographic characteristics as well as a function for grouping leads into distinct segments. There's a couple of assumptions here when we talk about segmentation. The first of which is that all leads are not created equal. They're not all the same. There are differences there that we can measure. The second of which is that their behavior is not random.
Those that are more likely to enter their information on a lead form, that's something that's distinct. There's a behavior that-- the pattern to that among common segments. And we can use that as a measure. And then finally once you've identified leads into segments or audience groups, there are different marketing strategies that can be applied and are needed for different segments. There is certain messaging that's going to be relevant to some segments over others. And so those things all need to be taken into account. And those are what we go in with an open mind toward as we segment lead pool. And so finally I just want to introduce, once again, our segmentation marketing analytics platform is called ElementOne. So I'll be referring to that a lot.
We're actually going to go in and do some real analysis in the platform around this area. Before getting into that, I did want to just give you some background on our methodology for building these audience groups, these segments. And as Paul mentioned and discussed, we have a tremendous identification layer in our architecture that gives us a huge advantage in terms of looking at all US households at the individual level and understanding, at that level, common demographic and behavioral differences among those households.
So that's kind of the foundational layer that we use. From that we built 232 distinct elements. You can think of those as micro segments, or clusters, or groups of those households exhibiting common demographic, neighborhood, and behavioral characteristics. From those 232 distinct elements, we've rolled those up into 172 tactical segments.
This is a very granular view of the total US households that we use to apply to our client and allow them to create custom groups, or audience groups, based on these very granular segments, very tactical segments. These audience groups, which is at the top of this pyramid, are really based on client business drivers and your own lead data. So we're looking to see who are the leads coming in? How do they score against these distinct elements and eventually these tactical segments and grouping those into much smaller audience groups between five to 10, exhibiting a very particular behavior that's highly relevant to a particular industry or business.
So what does an audience group look like when we do roll that up to the highest level? Only remember, from all US households to five to 10 audience groups, there are a lot of steps in that process. But what we end up with is typically those five to 10 groups that are bucketed based on compound behaviors for Lead Gen. And in this industry, we'll use their likelihood to fall into a particular lead group and then also certain desirable characteristics about that lead.
So for example, if we're looking for leads in the auto industry, and for premium automobiles, we're looking for higher incomes potentially and disposable income in those leads to make them higher quality. In the education space, we're looking for that desire to go back to school and complete certain degrees. So those are characteristics that we're looking for in conjunction with their propensity to be a lead for a particular generator.
And so in this example, we've broken leads out into common groups based on their affluence, using income as descriptor for different audience groups. But we can certainly take that across a variety of demographic factors, as well as behavioral and attitudinal dimensions to create custom groups for our client. So with that background and understanding-- didn't want to spend too much time on that. I want to dive into our ElementOne platform and take you through an example of a segmentation project for understanding some leads. These are example leads. They're from an auto lead aggregator. So we're going to use that as an example. There are approximately 50,000 to 60,000 leads that we brought in through our platform.
And I'm going to take you through the process we would do to better understand who those leads are, create defined audience groups for those leads, and then better understand which channels are driving those leads in, and then how we describe those leads at a more profiling level. OK, so this is our ElementOne platform, which is all web based, no custom client installation needed. Our clients have direct access to this. And any data they upload goes into the cloud. And it's behind those firewalls that Paul described, the Fort Knox of our company. And so we keep that data highly secure at all times. Through the platform our clients have the ability to upload as much lead data as they choose. And that could be for various campaigns or everyone that they've seen coming through and want to analyze and understand what those leads look like.
This can be done just for the media and messaging optimization that we're talking about today or for providing added value to your advertisers that are looking to secure leads from you. They may want to have a better sense of what comprises those leads. Outside of that lead form data, there are 14,000 different attributes available in the ElementOne platform. And all of those can be brought to bear, in terms of profiling individual groups. So let's go ahead and jump in here. I've already preloaded this sample data set of auto leads into the platform. Like I mentioned about 60,000 of those. And I built custom profiles around that.
The next step then would be to go into our customer analyzer here to understand what those leads look like relative to this 172 segment universe that we've defined. And so what we're looking at here is essentially all those leads loaded in, relative to a representative geography of where they're from. So taking the total segments in that geography or those zip codes that the segments or the leads were pulled from to get a sense of the total opportunity of everybody that could have been a lead in that geography as kind of the benchmark. And then we compare that against who actually became a lead to get a sense of their propensity to be a lead. And so here we're going to start, just for example, segment one. In the represented geography that this generator was pulling from, there were 133,000 households, which represents about 1% of that geography.
There were 501 leads that came out of that group, which is about a 1% composition, a 0.38% penetration rate, and an index on penetration of right at average, so right around 100. So for this segment one, they're just as likely as average to be a lead as any other segment of the 172 that are here. So let's look at segment 88 here, which is exhibiting almost a three times propensity to be a lead. Let me scroll down here so you get a sense of segment 88.
They comprise approximately 80,000 households in the geography. And they're making up 902 of the leads, which is a penetration rate of almost three times the average. It's 1.14% and an index of 296. And so right off the bat we can look across all these leads and understand which ones are performing better than others and giving us a sense of who you would want to target with your marketing message and dollars.
You want to go after segments that are exhibiting higher propensities to be leads for you, rather than ones that are under index. And here you can see some examples. Segment 148 where there are zero leads. There would be no reason to go marketing to segment 148 since there is a very low likelihood that they're going to be an eventual lead. So when you're looking to generate leads, you would look to the segments exhibiting that higher propensity to become leads. And I can sort all of these here by that propensity. And as you can see, it's skews here on the left side towards the highest indexing segments.
And over here on the right are those that would be non-targets for a marketing message. And so scrolling back up to top here, looking at segment 88, again a three times likelihood of being a lead, what do they look like demographically? And this is just kind of a summary here. But average income of about 60,000. They're 46 years old. 75% of these households have children. 50% are homeowners. Cost of living is relatively inexpensive for these segments. And their urbanicity is a one. That's the highest urbanicity, meaning the population density in which that segment tends to live is very high. Urbanicity of one is the highest urbanicity with a five being the lowest urbanicity in most rural, kind of small town living population. So just by loading in that lead list, we immediately know the segments more likely to be leads and what those segments look like.
The next step in the process is to create those audience groups we talked about in the methodology portion, where we're looking at these 172. 172 is great. It's very granular though, and it's too granular to really develop an advertising campaign and certainly too granular to develop a media buy against. So what the platform allows you to do is group segments that are exhibiting a common behavior together. And so what I've done is group the segments with high propensity to be leads-- so high indexing segment here. Also remember that these are auto leads. So let's say this auto lead generator was looking to group leads or value them based on income level, and discretionary income, and ability, and buying power for them to purchase a new vehicle.
And so they set a threshold of having an average income above 50,000 annually. And so what I did is I created a few custom groups, a primary target, which was high propensity to be a lead and high income. And then the secondary target was high propensity to be a lead but lower income and then on down the line. So created those custom groups. And we can pull those groups up here in the platform. And what this does is show those segments grouped now by those characteristics. So I created this primary target here that has an average index of 182, so twice the propensity of average to be a lead.
Clearly a segment worth targeting from that perspective. You spend $1 on this segment in marketing for every $2 that you'd spend on average for other segments. And this non-targeted bucket down here, which is exhibiting a 50% less likelihood to be a lead will cost you $2.50 for that one that you'd spend on a primary target here, segment one. So the combination of that high propensity to be a lead and then also that average income is approximately $80,000 makes them an attractive target for those lead buyers in the auto industry. And so clearly a segment worth targeting here. The average age of the segment is 50. They're not as likely to have children, 38% having children, 62% homeowners. So a higher proportion of homeowners. And cost of living is pretty average, a lot of affordable housing. And urbanicity of 2.5. So somewhat urban, moving out more towards the suburban areas is where this target tends to live. And so really quickly here, we just took a data file of just leads. That's all we knew about them. In this case we had address information used that to match the segments, and then grouped those segments based on some common behaviors and demographics. And immediately we have an audience group worth going out and pursuing and tailoring marketing for.
So if we think about what we want to talk about today, one of the key things is defining that audience group. Instead of casting a very wide net, going after everybody with your media buying, isolating that towards a smaller proportion that's going to be much more likely to drive leads. And so in this case, we're looking at this primary target being about 8% of the total universe. If we combine the primary target and secondary target, that comprises about 15% of the total opportunity.
But it's driving almost 30% of the lead traffic. So you can get a lot more bang for your dollar, higher marketing ROI in investing and going after these two targets, or even the top three targets, then going after everybody equally through those channels. So let's talk about channels. And the next thing we wanted to look at was understanding which channel was driving the best audience. And so what I've also done and was included in the lead file was the source of acquisition.
And so it's important-- and understand that a lot of buys are done today, and leads are generated through the online channel through display and search. Those are certainly two channels we'll take a look at. But some alternative channels are starting to emerge. DRTV, although expensive, is being used by several lead generators. FSI inserts in newspapers and magazines is also driving leads, as well as even the DM channel is being used. So in this example, I've taken all those channels to see which of those are really driving more traffic for this provider. And so I'm going to start with the display leads here broken out by these same groups. What you see here is display. If we think about the proportion of these leads, it's representing 40% of these leads.
So it's a significant channel. It's not one that you'd necessarily want to drop. But it is one you could work to optimize a little bit better. And so what I'm looking at here is group one, your primary target for this example is actually less likely to come in through the display channel. So where that display and that retargeting is being done today is not as effective at driving this primary target to the site. And so we're seeing it's doing OK with the secondary target and hitting on the secondary target. And actually the secondary target is the one that is most likely to come in through this channel. But so definitely some opportunities here to better optimize the display advertising in pinpointing that target there. When we look at it overall though, we don't see high skews across these industries. So as a channel, display is driving a lot of volume.
And you're not seeing a lot of differentiation across these segments. What you certainly don't want to see is your display driving these non-targets because that's money wasted. The non-targets are 50% less likely to be leads. And so it's wasteful to be investing in advertising directed that's driving into those non-targets. When we look at a few other channels here, we can also see similar analysis. So looking at search, knowing that that's also a paid search being a key contributor, how is paid search doing?
So the blue is representing display leads. The green is those that are coming in through that search channel. And again, non-targets here are more likely to come in through that search channel than average. Not doing well with the primary two targets when it comes to search. And so again, key words there-- other ways of driving leads through the search channel and areas for optimization. If we're looking to optimize around those segments most attractive to the business, some areas of improvement are needed. So looking at alternative channels, those that are coming through to the site directly, we see where you get good lift from the two primary targets here just coming through the native site. And that's being driven in through some other channel.
So let's look and see how the other channels are doing relative to the segments we've identified, the groups we've identified. Let's start with DRTV. The DRTV as a channel we see is not doing well at driving our two primary targets to the site or to telesales. Their under indexing on the primary target and significantly under indexing on our secondary target here. So as a channel and then knowing how expensive that channel is, not one that's performing very well for us.
When we look at FSIs, again it's doing OK. It's doing a little bit better with our primary target but not very well with our secondary target, where FSI, they're really having an impact, as you can see here on this chart, is with the above average lead. So as a channel, this isn't a bad one because it is driving those segments that are showing a slightly higher propensity to be leads, and one to investigate there, and to understand how you can drive success more so in the primary target. And then finally, let's take a look at the DM as a channel.
So DM is the first one that we've looked at outside of our two primary online channels where there is success with the two primary targets here, group one and group two. And so what that tells you is there is some opportunity through that DM channel to really drive some highly qualified leads into the funnel. And so this is just example representative analysis.
But really one of the focal points of the discussion today is how you can effectively leverage this platform to really understand the channels that are working and those that aren't working. And then those that are working, how do you optimize against those? So the first step, particularly if you're looking at DM, is to understand the geography in which your leads are coming from.
And so in the platform, we have a module called Analyze Markets, or the Market Analyzer, where you can go in for any custom group that you've built, project that onto any geography. And so I'm going to go in here and grab the primary group that we created. And it's going to default to total US against the state. I'm going to go ahead and grab media markets from that.
And you can see this is processing very quickly. This is all real time. Looking at the media markets in which the highest concentrations of my primary target fall. So for online, you can do some very sophisticated geotargeting. Once you understand the zip code where your best leads are coming from and optimize that way. For local search, you can leverage geography for that.
And then the channel that we isolated through our channel analysis with DM. And so we immediately know the markets where DM drops are going to be more successful, where there's higher concentration of those key segments. New York City, for example-- we're looking at almost 8 million households in the New York market. 1.6 million of those households, or roughly 21.6% penetration rate, falls into our primary target group. Remember they only make up 7% of the total US.
So finding a market where you can get to 20% penetration is very good. And that's actually four times the size of this segment and one we're targeting. And you can take this down all the way to the block root level in the Market Analyzer. And for those that are more visually oriented, you can plot this on a map and see where these high concentrations, or hot spots, are relative to core groups. And so quickly here, we see the areas of the country highlighted in the darker orange, which are the top deciles for their concentrations of our primary target.
And you can see over here in the Northeast, New York City being that dark orange. And then there's the light kind of yellow ares of the country where you would certainly want to avoid any geotargeting and any direct mail effort as you're not going to get cost efficiencies in those areas. And I'm kind of racing through here. I did want to show you kind of where to find them. And now quickly I want to go into better understanding those targets. I mentioned we identified some channels to explore. There are better ways to optimize those channels, though, through programming and also where you're placing those by.
So the first thing I wanted to show in this describe targets module, we take all 14,000 of those attributes, and we associate those where certain segments are highest indexing and ranked at the top to really get a good sense of the profile of those target segments. And so for this primary target group that we established, let's look at the demographics. We already know that their average age is around 55. What we can also learn is that there's a higher skew of Asians in this population. They're in the finance and insurance industry. They have very expensive homes. They were showing here a home value of 1 million. I wouldn't put a lot in that. They're high indexing for that. But it's only about 1% of this segment is in $1 million homes.
But overall they do skew high towards home value and higher income. And part of that is based on their occupation. This module shows you where they're highest ranking and where they're lowest rankings so you can get a sense of what they are and what they're not. And they're not a segment that doesn't recycle. They do recycle. They also are not a low income segment. You can see that as well. We can dive into certain categories as well when we look at the automotive space. So now we're looking within this segment to understand what's relevant to them and who we want to sell these leads to. And for this segment, they're over indexing on Honda, as well as Nissan, and the Accord specifically. And so we know that this is a segment that favors foreign Japanese automakers over domestic with the exception of the sport utility category where they do prefer domestic.
So depending on vehicle type and make, it would orient where you potentially could route these leads. And certainly of value when profiling a lead coming through the door and creating those audience groups as a baseline, you could take particular components of the category to be factors of influence in building those groups. So taking that intelligence into how you're profiling, grouping segments together. I do want to showcase the media side of this as well. So for internet and to inform internet buying, when we look at-- what are some of the sites that this segment is more likely to go to relative to the other segments, or groups, or audience groups that you created.
This one in particular we saw the high association with the New York geography. So it's not surprising that they're going to the New York State site, also New York Times. They're going to hotels.com, FOX sports, nba.com, dell.com. This is the social segment. They're on Twitter. We can see much of this just looking at their online behavior. We can look at where they're not. They're not on a lot of the job websites. When we look into-- since we're exploring direct mail, what are some of the magazines that they're doing just to give you a sense of copy and content that will be relevant to this segment. Again "The New Yorker" pops for this segment, "Vanity Fair," "Town and Country." I'll highlight "Playboy" here only because the last time I did a webinar, this segment also read "Playboy."
So maybe that's a common theme for this segment that I identify in these examples. But this segment is also into health and health issues. "Men's Health" and "Health" magazine. Again, you may not-- you saw that FSI was successful with that average audience group. And so maybe a way to boost that up with this top segment would be through some of these publications that are popping here for this group specifically. So another way to optimize there. Just to showcase TV quickly because there's a use case I want to get you guys through as well. You can see what this segment is watching. We saw that DRTV was not an effective channel.
But that could have been based on programming. If the buys were executed against a different flight schedule that didn't include some of this programming that your primary target was watching, that could be a factor for that channel not being successful. And so there's a way-- and media organization can utilize this data in executing those media buys on your behalf to drive those leads through that channel to you. So for example, they're watching soccer. We saw the sports affiliation with nba.com. And they're also watching a lot of ESPN and some of the home based networks. We saw that this is a segment that tends to own their home. So HDTV is popping here for this segment. But really the highlight here is just the depth of data that's available in this platform. This is just scratching the surface. I think I've mentioned a few times that there's 14,000 plus attributes available.
This is a module that allows you to get to that data very quickly and isolate it around segments and audience groups that are important to you, in particular those that are better performing leads. So with that, I want to jump out of the platform and back into the presentation. Kind of what does this all mean? What does it end up looking like? Once you've created these targeted audience groups, you create very detailed profiles for each one. This gives you contacts and visual to use within your internal marketing organization. You can't hand an agency just the platform and say, go to work. I want a campaign built around this audience group 24. You need to really put some definition around who that is through the platform pulling out profiles and attributes that matter, and create more of a persona around those audience groups, and really give the agencies something to work with when developing that creative. That's how you optimize and get more bang for your buck through relevant messaging that's impactful to your target audience. And so let's take a quick look here-- I know we're running a little short on time-- on a real time example, a Lead Gen company in the EDU space that used our platform to prioritize their leads.
They really wanted to understand who they should prioritize. They were already using verification and scoring, but they wanted to add another layer to that. And that was segmentation. So using the platform, they used ElementOne to identify leads that were more likely to be ideal students, those that were coming through as leads that would be valuable to different institutions, and a mechanism for routing to certain institutions. So they identified those segments that were ideal and those that were less than ideal, very similar distribution that we looked in the example in the platform. From that, they used a combination of their likelihood to be an ideal student, and their lead propensity, and the volume that they were seeing from those leads to identify the strategic segments that they really wanted to target.
So this grid here shows performance on the vertical axis and the volume on the horizontal. In this upper right quadrant are those segments exhibiting highly likelihood to be ideal students as well as propensity to be a good lead for them and then high volume also. They're getting a lot from those audiences. So those were the segments that they identified as key and began to target. The first step that they did is they found regions of the country where their key segments were located, just going into that Market Analyzer that showcased in the platform. And they were able to locate where those segments were and where to go after them. They use geotargeting tactics to do that. With all this data and information, they were able to optimize their media plan.
They focused their media efforts on that primary audience group that they identified, the same steps that I just went through, looking at the media channels that were best suited for those audiences, fine tuning on those, using the geography as a component of that, as well as the online behavior and understanding the sites that their targets were going to. I think more important than the sites that they were going to were the sites that they were not going to and cutting those out of their media buys and increasing their buys in the sites that they were going to. It also allowed them to tailor messaging to be more oriented towards those key segments in those profiles and be much more relevant to those prospective students that were more likely to come into through their channel and filter in through. This allowed for real time segment marketing. When they got to the landing page, they were delivering them with messages that were much more relevant to them. They could also do school routing based on that lead performance in certain tiers of quality to certain institutions, which increased their performance with those institutions and overall their sales.
Let's look at those results. This is a real example. By using these optimization techniques, they were able to reduce their overall media spend by nearly 10%. And you're saying OK. Well, they reduced their media spend, but what did that do? Well, they actually increased their enrollment rates by 10% as well, And part of that was through identifying those ideal students and passing those through to the right institutions. And that appropriate matching increased that enrollment. And as a result of that, they increased company revenue by also 10%. So a tremendous success story and an example of how really understanding your target audience and optimizing your media strategies and channel appropriately can lead to a much better result. And so with that, I'm going to wrap up. I do want to confirm that we don't have any questions here, which I've been racing so fast through this that I wasn't able to check everything. But it looks like we don't have any questions. But if you do, please go ahead and submit those now.
OK, I got a question here. And the question is really around the rate sheet. How much does this cost? How is this implemented? And I can't really talk specifics. So I'd ask that you reach out to your TARGUSinfo* account executive for that. But generally, the platform is licensed on an annual basis. And what comes with that license is a certain number of seats or users to the platform. And what comes with that is unlimited data upload. So you can really do as much analysis as you want within the platform and associate as many leads as you want to segment all within the platform and really fine tuning. All the steps I've taken in the platform and showcased to you today are all available through that process. Now there is a difference when you want to come out of the platform and actually act on some of those leads if it was through a direct mail effort or to identify those leads online, which is certainly something that our Ad Advisor product allows you to do. There's different pricing for all of that. But the basic analytics and the platform itself are an annual license. That's a good question. I'm sorry for not clarifying that. I did want to just-- because we're kind of going short here-- highlight that this is a webinar series that's ongoing. This is our second that we've done. We're going to continue to do these throughout the year. Our next one, as Paul mentioned, will be Wednesday, May 30th. And the focus there is putting customer intelligence into action. It will be similar to this one in that respect.
But certainly, if you have feedback or a topic that you're very interested in, we'd be happy to put something together and get that out to you guys. We're looking for your feedback on that. We'll be, as I mentioned before, on a panel at Leads Con talking about this topic. So it's going to be out there. Definitely try to hook up with that. And we'll be at booth number 400 at Leads Con. So please come by and see us. And we can talk more specifics about the platform and the capabilities that TARGUSInfo* has.
And so with that, I think we'll go ahead and wrap up here. Contact information's here. Please don't hesitate to reach out with any additional questions you have about the topics here today. I apologize for rushing through some of it. But certainly don't hesitate to reach out. And I really appreciate your time. An hour is a lot to give up to go through this. And hopefully, you got something out of it. Thank you very much.
*TARGUSinfo was aquired by Neustar in Nov. 2011.