[Follow our blog posts, obsession with data, and original articles on Twitter @RJMetrics]

Over the past two years, I have developed a growing fascination with lawn signs. Not the ones advertising politicians or plumbers, but the ones advertising websites. Dating websites.

These signs are so prevalent in my area that I decided to launch a private investigation into who was behind them and just how far they stretched. What I found started in my small home town and led me all the way to the secret guerilla marketing infrastructure of a multimillion-dollar company…

Background:

In the fall of 2007, I was about a year into an Analyst gig at a large web-focused private equity firm. My job description was simple: do whatever it takes to find interesting companies who are making lots of money on the internet.

During this time, every radio commercial, billboard, and t-shirt bearing a domain name held a special meaning: it represented an opportunity to find the next big deal. As you can imagine, the same names kept popping up again and again. I was looking for new deals everywhere.

One weekend, I trekked down to South Jersey to visit my parents in my hometown of Glassboro. The town sits about 30 minutes southeast of Philadelphia and has a population of less than 20,000. As I drove past my old high school, my deal-hunting subconscious noticed something bizarre. Stuck in the grass by the curb was a white lawn sign about a foot tall with a very simple message in black Times New Roman: “Single? www.GlassboroSingles.ORG“

It looked like something the local contractor would ask to stick in your front yard while he replaced your roof. Except… well, it was plugging a dating website. I had about a million questions, but two immediately simmered to the top:

• Glassboro is a tiny market of nominal interest to even local advertisers. Who would register a domain, let alone build a website, to target our tiny population?
• Who in their right mind advertises websites with lawn signs?

By the time I pulled into my parents’ driveway, I had convinced myself that the site was the product of some overzealous local entrepreneur. I wrote off the lawn sign as an amateurish stab at guerilla marketing. When I drove out of town the next day, the sign was gone.

Fast forward a few weeks. I was back in New York, rushing up 5^th Avenue on my way to work in Midtown. As I wedged myself through the usual crowd, something stopped me in my tracks. Eight feet in the air, tied to a lamp post, was a white sign with black Times New Roman: “Single? www.FifthAvenueSingles.COM“

I promptly morphed into one of those sidewalk-obstructing idiots who stares up into the sky and infuriates the people who actually have to be somewhere. Aside from the URL, this sign was identical to the one I had seen in Glassboro. By the time I got to my work, I decided that there were four possibilities:

• This was a complete coincidence and these were the efforts of two completely separate businesses with identically unorthodox advertising methods (unlikely).
• The overzealous Glassboro entrepreneur had loaded up his car with lawn signs and decided to extend his guerilla marketing scheme to the Big Apple (less likely).
• Some NY-based business had done some marketing in the tri-state area and decided Glassboro was a ripe market (even less likely).
• There was something bigger going on. This struck me as the most likely case, but raised a question that made my head hurt: if whoever is doing this has the ability to target New York City but somehow made their way down to Glassboro, how many of the towns in between have also been hit?

Not long after, I stumbled onto another clue. I was in Central New Jersey on my way to give a guest lecture at Princeton University, which is about the geographical midpoint between New York and Glassboro. As my cab rolled through neighboring West Windsor Township, I saw a familiar-looking lawn sign wedged in the grass alongside the road: “Single? www.WindsorSingles.ORG“.

That one did it for me. At the absolute least, I was now convinced that this lawn sign business had its tentacles stretched into almost every town in the state of New Jersey. It was worth spending some time to learn more.

Industry Research:

In talking to a few colleagues about this fascinating business, I learned that most private equity shops shy away from dating sites for a number of reasons:

• Dating sites are known for tremendously high churn rates (if your product works, your customers never have to come back; if it doesn’t they see no reason to come back). This means dating sites have to keep a steady flow of new customers coming into the top of the funnel in order to survive, let alone grow revenue and profit.
• High churn rates mean new customers have low, volatile expected lifetime values. This has a negative impact on the equity value of each customer, making it difficult to justify the valuation multiples seen by membership-driven websites in other verticals.
• The need to keep more and more new customers coming in creates a necessity for massive marketing budgets that often involve aggressive affiliate marketing (i.e. paying third parties to bring you new customers). This further damages the perceived value of the user base to a potential investor or acquirer.
• Like social networking, “online dating” is a natural monopoly (or, at best, a natural oligopoly). A dating site’s quality is determined by the number and quality of matches it can provide a new user, which is directly tied to the size of its membership base. This makes it extremely difficult to enter the market.

However, just because something isn’t a great investment prospect doesn’t mean it’s a bad business. Many, many people have become obscenely wealthy in this industry (both online and offline). The technology required to connect two people is trivial, meaning your only real expense is the cost of customer acquisition. If you are part of the natural oligopoly, your product quality will be high and people will seek you out. This cycle lowers your costs and sends your margins skyrocketing.

Furthermore, the online dating industry has made a lot of secondary players wealthy thanks to affiliate marketing. At times, online dating sites have paid as much as $100 per head for new paying customers, and routinely pay out at least a few dollars for new “free trial” users or other prospects. This means anyone with the power to herd single internet users can potentially tap into a strong monetization engine.

With this information in-hand, I started to see some beauty in the lawn sign model. Since virtually all dating sites are national, even ones with millions of members can under-serve certain geographic regions. The “YourTownSingles.com” approach leads potential members to believe their area will be extremely well represented in the site’s population. This creates the perception of high-quality matches, even if the total user base is small.

I had visited each of the URLs I saw on the lawn signs, and each contained a multi-step form asking for a bunch of personal contact information. This led me to suspect that the business wasn’t running its own site, but was acting as an affiliate marketer. Often times, an affiliate’s commission is tiered based on the level of pre-qualification of their referrals. This means that an affiliate can make a lot more money selling my information to a third-party dating site if they have my name, e-mail, phone number, age and gender than if they simply have my e-mail address.

Given the large amount of information these pages required, I became fairly confident that I had figured out what was going on (for the most part). The business makes a small investment ($50-100) in buying a domain name and a few dozen lawn signs for a given town. Then they put up a form landing page at the URL, plant the signs, and see what kind of return they generate by luring the townsfolk to their site and then passing their information on to user-hungry dating websites. If the ROI is positive, they keep at it. If not, they try another town.

After reaching these unverified conclusions, the lawn sign business slipped out of my mind for quite some time. When I quit my private equity job to found RJMetrics this past July, however, my interest was reignited. The reason: I still see these signs everywhere.

Relapse

I seriously can’t take a ten minute drive without passing one of these signs. What fascinates me more, however, is that they never seem to last more than a few days in one spot. In most cases, I’ll see a “Single?” sign somewhere and the next time I drive by it will be gone. I can only assume that these signs are being taken down by whomever maintains the property where they are placed (they are almost always stuck in the lawn of a public park or building).

The fact that these signs are still so prevalent today, more than a year after I saw the first, means two amazing things:

• Despite their short shelf-life (or, lawn-life), sticking these plastic signs into the ground in small towns has proven financially viable (I can’t imagine that a year’s worth of data to the contrary would result in the business continuing to print and plant these signs).
• Someone must be monitoring and replacing these signs as they are taken down. When you consider the number of towns likely involved in this system, it’s clear that this is far from a one-man show.

Then, just this past Friday, I saw the most amazing sign yet. A town not far from my house is called Haddon Heights, NJ. It is a miniscule town that occupies just 1.6 square miles of land and has a population of barely 7,000 people. Furthermore, 56% of the population is married and 25% is under the age of 18 (thanks Wikipedia!). Not exactly a ripe market for a dating business. Nonetheless, as I drove through the town, I saw (no exaggeration) twenty signs that read “Single? www.HaddonHeightsDating.COM“

Since the town seemed to have such a surplus, I decided to pull over and pick up a souvenir:

My new souvenir on the RJMetrics couch

Stealing a lawn sign brought me to a realization: these signs are driving me insane and I have to figure out who is behind them, how big this system is, and whether they are actually making any money by doing this. It was time to do some real digging.

How Big Is It?

First stop: Google. Dozens and dozens of crafted queries designed to find the slightest mention of one of these yard signs anywhere on the internet turned up dry. Absolutely no one out there seemed to be aware of these things (and those that were didn’t seem to care).

Maybe this was a smaller operation than I thought. To answer that question, I set out to discover just how many of these websites were actually out there. Since these sites are all just basic lead-gen landing pages, I speculated that the infrastructure of this system was a lot like a domain parking business, where a single web application feeds different content to a large number of domains based on which domain is accessed. If this was the case, all of the domains would likely correspond to the same IP address.

I pinged each of the four websites I could recall to see what IP addresses were serving them. Here are the results:

• GlassboroSingles.com: 200.46.241.132
• HaddonHeightsDating.com: 66.252.239.220
• FifthAvenueSingles.com: 69.41.228.6
• WindsorSingles.org: 66.252.239.220

Four domains and three different IPs. It appeared I might be wrong about the parking servers, but the fact that one of them showed up twice gave me some hope. It was still possible that these domains were parked in massive batches on various servers. However, they weren’t all in the exact same place. To get a sense of where the servers are actually sitting, I used an IP lookup and traceroute tool. As it turns out, here is where these servers reside:

• 69.41.228.6 is in Dallas, Texas
• 66.252.239.220 is in Miami, Florida
• 200.46.241.132 is in Panama

Why the scattered infrastructure? Is it possible these identical road signs that all appeared within 100 miles of each other are actually operated by different companies?

Regardless, I had the data I needed to size up the operation. With the IPs in hand, I turned to the “Reverse IP” tool at domaintools.com. The tool is simple: you provide an IP address and they tell you how many websites reside there. If you want, you can buy a complete list for a few bucks. The results:

• 66.252.239.220 hosts about 5,100 domains
• 200.46.241.132 hosts about 3,800 domains
• 69.41.228.6 hosts about 500 domains

Wow-is it really possible this lawn sign network includes so many domain names? I whipped out my credit card and purchased the answer. I stared in disbelief at an Excel file containing every domain name hosted across these three servers: 8,870 of them. They all fit the formula: a town name and a dating keyword.

I wrote some Macros in Excel to pull out each domain’s TLD (i.e. .com, .net, .org), name scheme, and the name of the town it represented. Here are the results:

• Unique Domains: 8,870
• Unique Town Names: 5,902
• Top Level Domains:
  • .COM: 65.7%
  • .ORG: 29.6%
  • .NET: 4.4%
  • .US: 0.3%
  • .INFO: <0.1%
• Sign Text Patterns (____ is the name of a town or city):
  • ____SINGLES: 78.0%
  • ____DATING: 12.2%
  • ____MATCH: 3.0%
  • ____DATES: 2.0%
  • ____PERSONALS: 1.8%
  • SINGLEIN____: 1.0%
  • ____SINGLE: 0.5%
  • ____CHRISTIANSINGLES: 0.5%
  • ____GREATEXPECTATIONS: 0.4%
  • DATING____: 0.2%
  • SINGLES____: 0.2%
  • ____PERSONAL: <0.1%
  • ____ASIANS: <0.1%

It’s important to note that this massive list is almost certainly not the list in full. I remembered four signs I had seen and they turned up three servers full of domains. It’s extremely likely that more signs might yield more servers.

I remembered originally thinking that this business was something homegrown to South Jersey- clearly I was wrong. However, now I had the information I needed to determine just how far these yard signs stretched.

I loaded a master list of every city in the country and wrote an algorithm to link each domain to a specific state. Naturally, there are some town names that appear in multiple states (which are a jackpot to these guys, since one landing page can serve multiple markets), so in those cases I assigned the domain’s state as “multiple.” Also some domains contained city nicknames or abbreviations, making it tricky to classify them in a quick batch process. Despite these issues, I was able to classify over 80% of the domains.

As it turns out, there were domains corresponding to all 50 states and Puerto Rico. Additionally, the names were distributed surprisingly evenly across the states. No single state represented more than 8.4% of the domain population, and the top 20% of states only represented 54% of the domains (I would have guessed an 80/20 distribution or worse). The top ten most referenced states are listed below:

• Texas: 8.4%
• Wisconsin: 7.8%
• West Virginia: 5.7%
• Pennsylvania: 5.7%
• California: 5.6%
• New Jersey: 5.0%
• Virginia: 4.7%
• New York: 4.7%
• Washington: 3.3%
• Tennessee: 3.2%

Clearly, this effort isn’t isolated to my tri-state area (in fact it appears to be even more prevalent in states like Texas and Wisconsin).

Let’s take a moment to consider the potential scale of this operation. Given the number of domains registered and the frequency with which signs appear to be replaced, I don’t think it’s unreasonable to speculate that whoever is responsible may have placed literally hundreds of thousands of signs into American soil over the past two years.

Who Is Behind It?

As I mentioned before, I was intrigued by the fact that these domains live on three separate web servers from separate hosting providers in separate geographic locations (Florida, Texas, and Panama). In order to find out who is behind this operation, I first needed to confirm that these domains and servers were in fact managed by the same company.

I visited several of the landing pages across the separate servers searching for clues that could tie back to an owner. I found nothing by way of contact information, but I did see enough to conclude that all three servers are related. While the designs vary slightly from domain to domain, all of the sites end up asking you the same exact questions about yourself and offering the same exact drop-down lists of answer choices.

Since I couldn’t find contact information on the webpages themselves, I decided to go find it on my own.

My first stop: WHOIS. The WHOIS database is designed to serve as a master directory of all domain name owners, although in my experience there are no controls enforcing the submission of valid information. Moreover, most domain name registrars now allow registrants to register anonymously by serving as a proxy for the registrant. Despite these shortcomings, it’s always a good place to start.

I looked up a random domain from my list and was disappointed to see it registered to “Domains by Proxy, Inc”-which is basically GoDaddy’s anonymous domain name registration service. Disappointed, I decided to try a few more just in case. Most were by Proxy, but eventually found some variations.

Many of the domains on the Panama server turned out to be registered to NuStar Solutions, S.A. of Panama City. On the other servers, those that weren’t registered by proxy were registered to IMAT Group of Vadodara, India.

The records didn’t provide domain names for these companies (their e-mail addresses were Yahoo or Gmail). However, Google was able to come to the rescue and turn up their sites: www.imatgroup.com and www.nustarsolutions.com. The sites revealed that both companies are offshore development shops. Both are centered around web design services, but offer additional services as well. NuStar mentions “Sign/Banner Advertising,” while IMAT mentions a more exhaustive list that includes “Guerilla Marketing Services,” “Sign/Banner Marketing,” “Localized Campaigns,” and “Direct Response Lead Generation.” Sound familiar?

At this point, I came to the realization that every question I answered seemed to introduce two more. In this case, they were “did someone hire these firms or are they acting on their own?” and, more confusingly, “how did a web design firm in Panama or India get a lawn sign physically planted in the front lawn of my high school in South Jersey?”

To find the answer, I decided that I wasn’t done with my WHOIS search. Back when I was cracking into companies in New York, I discovered a tried and true trick for getting a CEO’s cell phone number: historical WHOIS records. Often, the original registrant of a company’s domain name is its founder. As such, it’s common for the founder’s personal contact information to exist at some point in the WHOIS database. As the company grows, they change the record, but by that time (often unbeknownst to them) companies like domaintools have already saved the old information in their archive.

I went back to the list of domains and did an exhaustive search of WHOIS records looking for domains with WHOIS record changes. I quickly noticed that all of the Panama domains were registered in 2006, whereas the India names were registered from about the Summer of 2007 onward. This might explain the different servers; if a company hired these firms to register the names and administer the websites, perhaps they switched providers in 2007, leaving their infrastructure split in half.

Pretty soon, I hit the jackpot.

CLINTONDATING.ORG, which sits on the Panama server, was not registered to NuStar Solutions or by proxy-instead, it was registered to Terry Fitzpatrick at a company called “The Right One” in Norwell, Massachusetts. At long last, I had a company name. A closer look revealed another little present: Terry’s email contained a company website: therightone.com.

I visited the site and began to find some answers. “The Right One” is a matchmaking service. You provide them with basic information about yourself and they use it to match you up with a prospective mate. Clicking the “Get Started” link on their website leads you to a familiar sight: a web form asking you the same exact questions that appear on the domain landing pages. These were definitely our guys.

With one more click, I learn that The Right One has franchise offices across the country. Check out this map from the “Locations” page on their website:

Look familiar? If you need a hint, scroll up and look at the top 10 domains by state that I listed above. It’s a perfect match. Even more telling to my personal story are these particular franchise locations:

• Cherry Hill, NJ: Neighbors Haddon Heights, NJ and is a 20 minute drive from Glassboro, NJ
• Lawrenceville, NJ: Neighbors West Windsor Township in Central NJ
• New York, NY (420 Lexington): Midtown Manhattan, two blocks from my subway stop

Some quick research on the company itself yields a pretty complex business structure. The Right One is owned by a company called PAFCO International, which is itself a subsidiary of Together Management Group, Inc, which goes by the aliases TD Management Group and Together Dating Services. The combined company employs over 500 people in over 80 offices across the United States. If you assume the headquarters has at least a few dozen employees, that implies that any given satellite office is just a handful of people.

A January Inc. article mentions Together Dating as a client of Texas-based Instinct Marketing, a response-based marketing company that specializes in “vertically-focused websites.” Texas, as you may recall, is the home of one of the company’s three web servers. However, while Instinct Marketing may be Together Dating’s partner on the technology side, I find it hard to believe they’re the ones putting the signs in the ground. The geographic correlation between sign spottings and franchise offices is simply too strong.

Does Yard Sign Marketing Work?

Remember when I said that certain dating businesses can print money? This is one of them. A 2006 reveals that the company was then bringing in about $45 Million of revenue a year. I also found an entry from a disgruntled ex-employee on ripoffreport.com that claims the company charges their customers from $3,000 to over $15,000 for their matchmaking services.

At those prices, the economics of the yard sign strategy start to make sense. Yard signs like those cost about $1 apiece to print and if franchise office has a slow day they can send out their extra “relationships experts” to plant lawn signs within a 50-mile radius. One successful new lead can bankroll thousands of new signs.

Some Closure (and a Shameless Plug)

In the end, I consider this mystery solved. My years-long fascination with these bizarre signs that seemed to follow me up and down the east coast is over. As a bonus, I discovered an interesting company with an even more interesting marketing strategy.

Regardless of how you feel about Together Dating’s industry or its methods, you have to appreciate their tremendous, low-profile marketing machine and the data-driven technological infrastructure that supports it. This company brings in 8 solid figures of revenue every year using nothing but yard signs, some parked domains, and a firm grasp of the data that drives their growth. CEO Paul Falzone explained the importance of such data in a recent interview.

Even in this offbeat corner of the business world, success is a function of intelligence, strategy, and analytical decision-making. The online business intelligence tools we provide here at RJMetrics are designed to empower all online businesses with these strengths. Our tools provide new echelons of accessibility to your existing data, allowing your business to measure, manage, and monetize better. If you’re interested in learning more, contact us for a free discussion about how we can drive growth in your online business (no lawn signs required).