“It all started above a deli in Staten Island.” And so begins one of the lesser known and more  intriguing stories in the world of online lead generation, one that has resulted in not just one successful exit but potentially three. Junkies of online lead generation will know the name ClassesUSA, and they will know it not for some of its innovative work on display today but for owning display before anyone, even the schools themselves, understood that it could be done.  And, it all started above a deli in Staten Island.

The story which is today SeniorsForLiving.com began in 2000 when Jordon Keltz, who ran an internet consulting company ,became introduced to Luciano Rammairone, the founder of CollegBound. Today, they rank among the largest of online lead generation providers, easily in the top there, but in 2000, CollegeBound was an offline business, a series of publications for students and adults interested in higher education. Web guy meets education guy. They name their online venture ClassesUSA. The rest is history. Right? If only it were that easy. And, if it were that easy, the story wouldn’t be nearly as interesting.

The year 2000 was early for online education lead generation, but it wasn’t early for several of the larger for profit institutions. Apollo Group’s University of Phoenix for instance had just entered its third decade of existence, but unlike today, where liquidity / monetization for those entering the marketing of online education, there was no field guide for ClassesUSA aka “Classes.” Their first site in fact had none of the usual suspects that we find on today’s online education properties. Luciano knew he wanted to create an online business, hence partnering with Jordon and creating Classes. But, with neither knowing what exactly that entailed what do you do? In their case, the logical choice was creating a showcase of classes that could be taken online. And rather than compensation occurring on a per lead basis, they received payment on a per transaction level. These weren’t degree programs but vocational courses; think of it as an early version of EduFire.

Site built. Next was traffic, and this is where things start to click into place for amateur historians. In 2002, Jordon made a deal with MSN for placement. This wasn’t a straight banner deal but a more robust integration of their content and course listings. It was a cpc deal for MSN, where clicks on the courses would earn them money. It was still an arbitrage play for ClassesUSA, because they only received payment on a purchase. What they didn’t know at the time was that this deal and style of placement would become one of the best things that happened.  He certainly would have guessed it when they were losing money each month over month. But that would soon change, and it would change the entire focus of their business as well.  The placement attracted the attention of AIU who saw the Classes content as a natural fit for its certificates. That AIU wanted to pay on a lead basis was an initial stumbling block for a company that operated on an ecommerce model, but they went ahead and tried it.

Not long after, Jordon found himself hearing from Capella and the University of Phoenix. They too had seen the site, and now with AIU on there, they wanted placement as well. And, it was then that the business started to shift. Leads worked. They went from losing money to making money, and even better they had the premier entities coming to them as opposed to the other way around. The content deal started working too well in fact. Jordon negotiated a cap on the number of clicks they would pay out to MSN. After that cap, it was pure profit. Around the time that they began renegotiating the deal was the time that MSN introduced PerformancePlus, their display on a CPA program that was open to a select few big spenders, one of whom was Classes. Now, they started running individual school ads on display and on other placements throughout the site. And, guess what happened when they put Phoenix on the home page? Everyone wanted the same – from schools to other aggregators. The gatekeeper to that traffic for a period became Classes and on the path to exponential growth.

The next step became to own all of elearning on MSN, which they did, and then to apply that formula to other properties. Jordon proceeded to create a partnership with AOL, and in the next two years, he had crafted arrangements with almost all job and career sites. In 2005, Experian gave Classes an offer they couldn’t refuse, becoming part of Experian Interactive along with LowerMyBills and AffiliateFuel. Jordon moved Classes from its Staten Island headquarters that they shared with Luciano’s other business, CollegBound Network to the city, Manhattan. At about the same time, an old client from pre-Classes days was pitching a lead generation idea to him that Jordon invested in, and when his contract with Experian ended, he decided to take over the helm and apply his learnings to the senior housing lead generation space… which brings us to SeniorsForLiving.

The online lead generation space for seniors is a fascinating and tantalizing one. It has all the right makings for a successful vertical – a large audience (boomers and their parents), a high ticket price (senior housing is not cheap), a transactional component, and a decision that requires follow-up, i.e. you don’t buy it online. As any who has tried to get into a new vertical can tell you, it’s a grueling process. There is a huge sales undertaking as you work to get coverage for the leads and as you work on educating buyers who in the past haven’t used leads as part of their business. It is according to Jordon where education was about ten years ago. There are some national brands that buy leads, but the mid to long tail isn’t there. And, senior housing lead generation is more akin to ground schools lead gen as a physical presence is needed. If there is anyone who will have success beating down doors, though, it’s he. A key to success he tells me is making sure institutions understand performance based marketing is the way to go, and trying to institute some online dna into marketing teams that are often focused only offline. Sounds familiar to many, I’m sure.

Overall, Jordon is sticking to the formula he knows and believes in – user choice and exclusive leads. Their model focuses on a portal based approach where everything relies on the user making informed choices and understanding what happens next. That user doesn’t fill out a generic form but one for a specific institution. Not new to the game, SeniorsForLiving is focusing on quality of leads. As for traffic generation, they have made themselves very publisher friendly, and given what he did in the past, I wouldn’t be surprised if there weren’t some interesting partnerships in the works.

LeadsCon has a small number of followers, and not surprisingly, it generates almost all of its activity, around the time of the event.  Going through those tweets, though, a few started to stand out. Happy as I was to see the additional volume, they just didn’t make enough sense. Here is one such tweet:

It wasn’t until a few started appearing that I began to get curious.  Here’s a snapshot of them in context.

You most likely noticed two from different users with the exact same text but without the re-tweet. Again, for those not familiar with the event, it would seem a normal thing to say.  Except as the brand owner, I know it isn’t. That comment was four months too late, but it’s not the tardiness of the context that is the real problem. The link being promoted is. Having been in the performance marketing, online customer acquisition space for quite some time, I am used to seeing crafty affiliate tactics. This ranks up there. A click on the link takes you to a business, just not one relevant to the conference goer, unless we are talking about what a conference goer does on their personal time. Yes, you guessed it. The link goes to an adult dating site – XXXBlackbook.

Like a webmail account, signing up for a Twitter account is a frictionless process. That’ makes deciding the real users from the fake that much harder. A click on one of the profiles above doesn’t go to a blank page with no friends/followers and zero tweets. It goes to a page resembling an active user at first glance. And, while it might seem like a manual process, those doing this type of spam have it all automated, from the signups to the followers to the tweets. And, while you can rid the system of spam, you can’t rid it of human nature. Hard to say if this will become a chronic problem, something akin to Google developing Quality Score, or simply a passing fad. Having finally made Twitter a part of my business, and increasingly valuable part, I hope they can squash this sooner than later. With the upcoming $100 million investment and increasing omnipresence, something tells me they will.

Paid Content ran a blurb today regarding somewhat stealthy startup, Inadco, titled “Lead Generation Firm Inadco Raises $5 Million; Spends $3 Million Of It On Acquisition.” Alas, we know no more than they do regarding the firm behind the funding or the company that was rumored to have been acquired. What the story did do, though, is bring to light one of the best trends we are seeing in online lead generation – the platform play.

There are a several different approaches that those in the online lead generation world have taken. The first is arguably the oldest and most classic approach – a direct marketing / arbitrage business. In this model, a company spends money to make money. It is fair to say that the vast majority of revenues flowing through online customer acquisition today follow-this model. They include some storied firms with impressive revenues. Quinstreet, for example, took the bold step of publicly disclosed its fiscal year 2008 earnings – an unusual step for a coming that has historically kept all information tightly guarded. (FYI, from the article, “QuinStreet said it generated revenue of $261 million, up 36% from a year ago, and earnings before interest, taxes, depreciation and amortization, or Ebitda, of $57 million, up 57%.”) Regardless of why they shared the numbers, it’s great to see, as it they wouldn’t do so were they not confident in their business and the industry. Sneaky large companies such as CollegeBound also fall into this direct response based model.

The second model is the brand building approach. It’s one that is very new and epitomized by companies like BillShrink, CreditKarma, and KnowBeforeYouApply.com. The premise sounds easy enough – create a compelling consumer proposition such that users choose to visit your site. They either hear about you through PR / friends, or they find it through organic search results. While many companies own well ranked organic sites, Quinstreet chief among them, what differentiates that SEO approach from the brand approach is the investment in recall, in building a service that users don’t just use once but come back to time again. Among contenders, BillShrink has had the most high profile attention (hard to beat inclusion in a national T-Mobile commercial), but none can yet claim victory for the approach. If lead generation is in the second inning, then the brand approach is just getting drafted. Like the disclosure of earnings by Quinstreet, that companies like these exist is a boon to the long-term future of online customer acquisition, because it signals investor’s willingness to embrace the sector and a real focus on the user.

Third comes the impetus for the post, Inadco and the platform approach. When we look out over the landscape of internet advertising companies, we see plenty that fall into the above categories – from direct response firms to consumer plays. The largest plays, though, have often been platform plays. Google, for example, makes all of their money by creating efficiencies for advertisers to reach publishers and for publishers to earn money off ads. They just happen to be their own publisher as well. Creating a platform is no easy feat, but doing so means being able to achieve a scale that most other businesses can’t – in reach and revenue. The best are technology driven and can create true barriers to entry. Such is what Inadco certainly hopes to build. And, while we have no shortage of platforms offering this level of efficiency on a CPM/CPC basis to advertisers, we have very few doing it on a per lead basis. The belief is that they can create a monetization engine that for a good chunk of inventory will perform better for publishers than the click to a third-party site options today. It’s not an easy business to build, because it takes time to get to scale, and there are no dumb money short-cuts that exist with some other forms of online ad spend. Inadco will earn their success, but when they do, they will really have earned it. Two others focusing on the platform approach – Pontiflex and Performline. What’s great is that each has the chance to succeed without needing the other to fail.

Years ago, I remember speaking to someone who worked in online mortgage lead generation, and he half joked, half lamented the fact that he wouldn’t recommend his company’s service to a friend looking to refinance. And, even today the same sentiment exists. The main reason – in many of the large verticals today, consumers go through a sub-par user experience. It would seem as though it should be easy to fix, but like so many things, it’s not as easy as it seems.

While individual brands and agents/brokers/contractors, etc. will advertise on their own, the vast majority of leads come through the aggregators. (I am a fan of aggregation, so any negatives uncovered are not inherent flaws with the model. ) The customer experience goes like this. A user lands on the site, provides the necessary information needed, and then hits submit. They are then contacted by up to eight lead buyers, some times right away. Depending on the model, the user may or may not learn of the individuals that will be contacting them. When I went through the process for moving, the site sent them to me in this email.

As for insurance, it is not widely the practice to disclose from who you will hear. That’s also the case for refinance, debt, home improvement, and virtually all non-branded experiences. Online education is the big exception, as users must more proactively select which schools with whom they wish to speak (even though the list from which they choose is often heavily influenced by the aggregator).

Online lead generation for auto insurance is among the more unique verticals in operation primarily due to the high degree of co-opetition that exists. Aggregators commonly sell unsold leads to other aggregators to increase their revenue / recoup costs. As a result, a lead buyer who works with Company A might be receiving a lead from Company B. And, a user who visits a site operated by Company C, could very well hear from both A and B. While this helps make sure that all parties make as much as possible, it creates little incentive for a buyer to develop as deep a relationship as they might, and it means the consumer can easily be overwhelmed when it comes to speak with someone.

The shopper with policy in hand can make the most of the current system, but the casual shopper who clicked on a banner ad won’t be. Why? For them, the value proposition isn’t fulfilled. They come to sites run by aggregators most often after seeing an ad leading them to think they will see rates at the end of the process. Most use verbiage similar to, “See how much you can save?”, “Are you paying too much?” , “The average driver can save x”, and so on. If the ads very clearly said, enter your information, we’ll throw it into a black box, and then wash our hands of you only for you to get called by up to eight people who will try and sell you on insurance, would they do it? Getting back to an earlier point, is this something  you would suggest a friend do? Again, it’s not that there is anything wrong with lead aggregation, and we shouldn’t just blame them for any shortcomings in the user experience.

Proactive vs. Reactive Selling

Auto insurance is not something that people think about all the time. They don’t always realize that they can switch to a new provider or that they might save more. It takes the vast majority of them being reminded that they could save for them to make the effort to look into it. That’s what the branded TV commercials do and what the online ads do. They get people’s attention and make them take action they might not have otherwise. It’s the difference between Proactive Selling and Reactive Selling, where the latter is for the small majority who are actively searching / thinking about the process. And, as a result of this process, more people end up saving and the ecosystem does more business than were it simply to wait until it crosses people’s minds.

Everyone in the ecosystem relies on this model, and it’s the only way for instance that a lead generator can be competitive. There is some serious doubt that it will last, though. One of the biggest questions facing our industry is should the users get to decide which of the available lead buyers gets to contact them instead of being “matched” (and bombarded)? And, is that really good for anyone, including the lead buyer? It certainly isn’t great for the lead generators, because the current practices make it too easy for even more competitors to enter.

Blind Matching vs. Total Choice

A handful of companies take a truly transparent approach to the matching of user and lead buyers. Zillow is one such example. Luckily, for Zillow, they don’t have to make money like vast majority of companies. Being the a specific type of heavily funded company, they have the luxury, or at least permission, to operate in whatever manner they feel is best. For Zillow, this means no money spent on user acquisition and an emphasis on transparency. It’s a great model if you can make it work. For the rest of us, it’s simply not practical to emmulate. But, while user choice makes a lot of intuitive sense, it doesn’t necessarily mean an easy transition is possible.

Given that most companies must spend money to make money, they need to have a way to make money once its spent. For lead generators, this means guaranteed buyers. If users make it almost all the way through the process, i.e., fill out their information and are presented with those who could meet their needs, but end up not selecting any one, that’s a real problem. Or, they may decide to select only one. Again, that’s suboptimal, because they will get better service by having some competition. The real challenge are the unknown economics for the lead buyer and seller. If a buyer pays $10 for a shared lead, what do they pay for what InsWeb is calling with their BestInsuranceMatch.com product as a referral lead? Do they pay 50% more, two-times more, etc.?

The knee jerk answer is that they should pay similar to how they do now, that is, a percentage of the average expected profit. Today that calculation includes a certain cost based on number of leads needed to close. But, if by the user chosing with whom to speak, shouldn’t that mean a higher close rate for the buyer? And in addition to the higher close rate making the lead worth more, isn’t the hassle of not spending so much time on lower converting leads also worth a premium? That’s the questions being asked, but the industry faces a real headwind in getting to the answer. If the economics don’t work, i.e. they don’t close that much better to reduce the number of leads sold per inquiry, then we are left with fewer leads generated (because companies can’t be as competitive for media) and with a user experience that is still sub-par. But, it’s a user experience that so far maximizes economic utility. I just hope for the sake of online lead generation, that companies keep trying to prove alternate models exist where the user gets a better experience, the buyer gets a high converting lead, and the seller / aggregator can profitably purchase media.

Targeting

Both LinkedIn and Facebook offer unique targeting options. Few sites have the data that theses sites do, which makes up that targeting foundation. Even Google, which still stands as the gold standard, can’t match (without getting into some real trouble) the same level of gender and demographic specificity. Between these two social networks, which targeting you prefer, will depend on what you want to advertise. Overall, given it is a new, more flexible platform, Facebook offers a much greater range of options. The one in particular that I like is the ability to target by company. LinkedIn has the data, but they haven’t made it available yet. Below is a screenshot of the LinkedIn targeting page. Besides not allowing an option for company, the biggest weakness is the restriction with respect to how many targeting selects one can select. They allow you up to three. My guess is that such a restriction has to do with their not wanting to segment the audience too small. Facebook on the other hand, allows you to practically advertise to a single person if desired.

According to LinkedIn, my audience is almost 550,000 people. Good for them but no so good for me where specificity over reach matter. Here is a screenshot of the parallel screen in the Facebook process.

My audience here is less than 15,000. That’s fine with me, and if anything, I suspect it’s actually inflated.

Next up in the evaluation process is what the user sees and how the process was for me to set up.

Ad Unit / Creative Shown

Both companies have a long, long way to go before their platform’s scalability matches that of their audience. Facebook in particular will benefit greatly once they open up an API for ad creation. It’s a very manually process for both, more so for Facebook only because they allow you to slice and dice the audience so well. It makes you want to test. There are a handful of third-party tools which in effect just take your data points then enter them onto the Facebook site as if they were you. That speeds things up, but the entire process is the equivalent of a lead generator giving leads to a client by having to fill out the client’s form for each lead they collected, instead of being able to post the data directly.

As for the ad creation process, both are relatively straightforward.  First-up Facebook.

And now, LinkedIn’s:

Both, let you copy existing ads for use in a new ad, which is very helpful especially if you have some specific targeting options selected.  The one thing we won’t really discuss which has become relatively standardized is the ability to run a variety of different products/services under one account. As a commerce platform though, Facebook is the winner here because they designed their system around supporting multiple campaigns.  I can run many ads for LeadsCon and many ads for something unrelated but have them visually separated and reported differently. Not so with the current LinkedIn platform.  As for the ads themselves:

Despite the personal touch of the LinkedIn ad, whose language tries to increase the trust and value of the ad by inserting a level of accountability, I prefer the Facebook one. Facebook has two big advantages.
1) Size of the image – much bigger, and
2) Amount of text – you can say much more than you can with LinkedIn. And LInkedIn doesn’t help you as you create the ad by letting you know how many characters can go on each line.

Pricing

While you could make a case that Facebook snuck out a close victory on each of the previous rounds, this one doesn’t even come close. Both sites share similar pricing models in that you can pay either on a cost per click basis or on a cost per thousand basis. The sites differ dramatically on those minimums. Facebook uses their own algorithm to determine minimum cpc’s and cpm’s. In the end, the company tries to hit an effective revenue per thousand, so paying per click becomes more affordable for those with higher click through rate ads. A high click through rate means you don’t need as high of a click cost for them to hit their target. The better your perform, the more impressions you see. It’s the same market place dynamics we see in any competitive ad landscape. The nice with Facebook is that the cost to start playing is relatively affordable. They are willing to show your ad without them necessarily making a lot. If your ad doesn’t perform well initially, then it will quickly lose impressions, and the only option becomes to increase the payout or design a new ad that will presumably yield a higher effective return for Facebook.

With Facebook on my B2B ad, the estimated cost per click came in at a not too cheap $.70. Were I to pay on a CPM instead of a CPC basis, I could have paid in the $.50. I chose CPC and set the minimum bid to several dollars though knowing they would need some room to play with if I were to get any distribution. Here is a snippet. It’s still a modest spend despite my having a $100 limit. There are alas only so many people who will click on a LeadsCon ad.

LinkedIn has dramatically higher pricing – good for LinkedIn but not as immediately good for me, especially given the less granular targeting. My ad shows to a broader group for more money. As for pricing, the minimum CPM is $3.00. The minimum CPC is $2.00. Doing the math, if you ad will get a better than 0.15% CTR, pay on a CPM. If not, do CPC. I didn’t think I would beat that CTR but I still chose CPM. Chosing CPM wasn’t a great choice as we can see from the below:

Results

Unfortunately for me, both Facebook and LinkedIn suck so far for promoting LeadsCon. Despite the call to action of potentially free admission, none of the clicks have yielded in a short form submission. I have created several different landing pages, both single step and multi-step with the same overall result. While I potentially haven’t hit statistical significant, at $2/click+, it’s already hitting pocketbook significance.

For me, the soft incentive is one where if executed properly, it is an upsell to an action that you would already take. It’s very different than a direct tradeoff, such as buy this to get this. It’s more of a teaser. The challenge of course is making sure that the person filling out the form understands what is involved. Recently, we came across another fairly common soft incentive, an ebook, but it’s execution might lead to less than qualified leads. The good thing about an eBook from a giveaway perspective is the cost. Outside of production costs, generally low, each incremental copy costs very little; yet, the perceived value can be high especially if on a topic that someone values. (It’s no wonder we see so many work from ebooks pitched as ways to make fast money…reselling the same ebook.) Not all ebook’s are necessarily about working from home. Here is one used in a debt consolidation lead generation campaign.

It’s a very clean and well executed page. It’s a good model for a higher converting page, one done in the more classical sense of a landing page that leads a user to a specific action without giving them many oppotunities to look and navigate to destinations outside the intended path. In other words, there is nothing wrong with this page from a functional standpoint. The issue deals with the message and expectation setting to the user. The headline reads, “Getting OUt of Debt Is Easy” followed by “Our eBook Will Show You How.” Next to the form rests a picture of a book, presumably the ebook on top of which reads, “Fill out the form to a FREE copy of our ebook.” And, under the ebook are some testimonials to the effectiveness of the book. That’s great if this were only about giving away a free ebook.  You get the sense that something else might happen by reading the fine print, i.e. “DebtExpertAdvice.com representatives are waiting to take your application. Our expert debt counselors are offering to settle all late accounts to raise cash this month and cover growing defaults. There is no obligation to receive free debt help, so contact us today and have your debt relieved!” (The best part might be the stating that filling out the form is accepting their privacy policy which doesn’t exist on the site.)

Fill it out to get your ebook, and you see this next:

Sure enough, it’s not really an ebook form but a standard lead form with a nice upsell to a credit monitoring offer. Stated quite prominently on the thank you page is their intention to follow-up with you and how “by phone and/or email.” The ebook isn’t an auto download. You have to hunt for the link on the page. (Curious about the ebook? Here it is.) Since this is a debt offer, the callers are used to dealing with a lead that doesn’t want to pick up the phone. So, they might not notice the degredation in quality or be as sensitive to it. Unfortunately, the lack of honesty and transparency to the user ultimately harms all of lead generation because it makes people less trusting and makes those in our industry look less trustworthy.

Distilled, the affiliate tactic used for the incentive promotion ads involves creative ways to a) buy traffic on keywords not directly relevant to the core business and b) pay less for traffic than they would if they tried to bid directly.

The online education landscape is among the most competitive, especially in paid search. It’s also full of schools being heavily wooed by Google to spend directly with Google and rely less on the aggregators. Google is even reportedly offering pricing incentives. One area, albeit with less traffic and arguably lower intent, where there aren’t the same brand / direct advertiser issues involves scholarships. Below is another example, this one from the online education space, courtesy of Vantage Media, although they are far from the only ones. (Education Dynamics for instance has run a very successful and well-branded scholarship promotion they call Project Working Mom.)

The Ad:

Clicking on this ad, takes you to the follwing landning page:

The page above is interesting. Being experts in the online education space, Vantage knows that schools don’t want just leads. They want enrollments. Enough people fill out forms with no real commitment to enrolling as is, so making it even easier to do that (offering them an incentive) makes the liklihood for conversion even lower. To combat that, they put the following disclaimers to discourage just those wanting money:

• Who Can Apply: Anyone planning to enroll in a school of higher education
• Requirements: Winner must be enrolled in a school when the scholarship is awarded*.

Fill out the essay, and it takes you to:

While they collect some lead data here, outside of the check box asking users to join their email list, they don’t suggest that it will go to a school. That opportunity comes later, on the third page:

Notice the subtle prompting of the headline saying, “Which school would you use a $1000 scholarship at?”