They asked for a heat map. I deployed PinID, a closed-loop attribution platform I had built in 1995 and already integrated into LoanLink.com, MortgageSites.com, and dozens of other sites, and extended it into a full direct response system connecting direct mail to personalized web experiences to call center routing, all with individual-level attribution. It ran for over a decade and generated over $500 million in attributed revenue.
In 1998, Terminix came to my team at Stirling Bridge Group with a straightforward request. They wanted a heat map. They had direct mail campaigns going out to millions of households, and they wanted to see, on a map, where their response rates were highest. That was the entire brief.
We could have built the heat map, collected the check, and moved on. Instead, I looked at what was actually happening in their operation and saw something much bigger. They were mailing millions of pieces with no way to connect a specific mail piece to a specific response. No way to know which households called, which calls converted, or what any individual campaign actually returned at the household level. They were spending tens of millions of dollars a year on direct mail with aggregate-level reporting and no closed loop.
The heat map they asked for was a symptom. The real problem was that their entire direct response operation was flying blind at the individual level.
"They asked for a heat map. What they actually needed was a system that could trace a single mail piece through a phone call, an appointment, and a sale, then report the revenue back to the original household."
Rather than delivering a heat map and walking away, I deployed and extended PinID, the closed-loop attribution platform I had built in 1995 and already integrated into LoanLink.com, MortgageSites.com, and dozens of other sites. For Terminix, I expanded it into a complete direct response system. Nobody was doing individual-level attribution at this scale. The system wouldn't become standard industry practice for another decade.
PinID was a closed-loop attribution platform I had built in 1995 when Dennis Stover brought me the concept. I told him I couldn't build it, then built it. By the time the Terminix engagement started, PinID was already integrated into LoanLink.com, MortgageSites.com, and dozens of other sites. For Terminix, I applied its core mechanic to direct mail: a unique identifier assigned to every recipient. When that person responded, whether by visiting a personalized URL, calling a tracked phone number, or both, the system captured their identity and linked it back to the original mail piece, the campaign, the offer, and the geography. Every touchpoint was trackable to the individual.
Each recipient who visited their personalized URL landed on a web experience customized to their specific situation: their location, the service being offered, the seasonal context, the offer variant they received. This wasn't just tracking. It was personalization at individual scale, years before the industry had a name for it.
Attribution doesn't stop at the click. Most of Terminix's conversions happened on the phone, not on the web. So the system had to extend into the call center. I built the integration layer that connected inbound calls to PinID records. When someone called the number on their mail piece, the system knew who they were before the agent picked up. The caller's campaign history, mail piece variant, and web activity were all available to the agent in real time.
The outbound side was just as critical. Households that received a mail piece but didn't respond within a defined window were queued for outbound call follow-up. The call center agents had full context: which piece was sent, when it was delivered, whether the recipient had visited the personalized URL. The follow-up wasn't cold. It was informed.
The system didn't treat direct mail as a one-shot medium. Based on the response behavior of each household, automated follow-up mail sequences were triggered. A non-responder got a different second touch than someone who visited the URL but didn't call. Someone who called but didn't convert received a targeted re-engagement piece. Each follow-up was driven by the behavioral data flowing through PinID, turning direct mail into a multi-touch, data-driven channel years before digital marketers popularized the concept of "drip campaigns."
The entire system fed back into a single attribution layer. Every dollar of revenue could be traced backward through the sale, the appointment, the call or web visit, the campaign, the mail piece, and the individual household. Terminix could finally see, at any level of granularity they wanted, what was working, what wasn't, and exactly how much each campaign, each geography, and each offer variant was returning.
This wasn't web analytics. This was full-funnel, offline-to-online-to-offline attribution. The sophistication of what we deployed at Terminix in 1998 wouldn't become standard practice until the mid-2000s, and most companies still can't do it today.
The PinID platform I deployed for Terminix in 1998 ran for over a decade. It became the backbone of their lead generation operation. Over that span, the platform generated more than $500 million in directly attributed revenue. Not estimated. Attributed. Because the system was built from day one to close the loop between every dollar spent and every dollar earned.
As the platform proved itself at Terminix, the underlying PinID technology was deployed across 200+ enterprise customers in recruiting, mortgage, insurance, and other industries where connecting offline marketing to online behavior to phone-based conversion was the difference between guessing and knowing. The same architecture, adapted for different verticals, each deployment reinforcing the core principle: if you can't trace a specific marketing dollar to a specific revenue outcome at the individual level, you're not measuring. You're estimating.
PinID was built in 1995. The Terminix deployment, its largest and most demanding implementation, launched in 1998. Google AdWords didn't exist yet. The term "attribution modeling" hadn't been coined. Facebook was eight years away. The martech industry, the one that would eventually spend billions trying to solve the attribution problem, didn't exist.
I built the solution before the industry knew it had the problem. That's not a timing accident. It's a pattern. When you understand both the commercial mechanics and the technical architecture, you can see where the gaps are before anyone else names them. You build systems that arrive years early because you're not waiting for the market to tell you what to build. You're watching where revenue gets lost and building the thing that stops it.
The PinID story is the clearest example of the CTMO discipline at work. Not marketing. Not technology. The structural layer between them where the real money lives.
"I built the solution before the industry knew it had the problem. Full-funnel, individual-level attribution in 1995, deployed at scale in 1998. The market wouldn't catch up for a decade."