The Bot That Started It All Still Doesn't Know Your Regulars

The webchat pioneer helped invent commercial webchat, incorporating in Delaware in November 1995 and going public in April 2000. That is a genuinely early, category-defining position. But the webchat model it pioneered was built to greet a first-time, anonymous website visitor, not to recognize the regular customer who has called five times this year. Being first does not make a platform good at remembering people. Those are separate design problems. The industry solved the first one decades before it seriously attempted the second.

Why was early webchat built for strangers instead of regulars?

Webchat in the late 1990s and 2000s existed to catch anonymous site traffic at the moment of hesitation. It offered a proactive invite on a product page or a "need help?" popup. The entire architecture assumed no prior relationship. It was built to authenticate nothing, ask everything, and resolve the issue in one session. That design choice made sense for its era and its channel. It becomes a liability when you stretch that same underlying pattern to serve call- and appointment-driven businesses. The regular at the dealership who bought two cars in three years expects recognition. So does the family that calls the same furniture store every time they move.

What does the acquisition-and-service-failure research show about "regulars" specifically?

McLelland and Goldsmith's 2014 study of consumer reactions in the wake of a merger examined how established customers respond when a familiar service relationship changes ownership or direction. They found that customers with a longer prior relationship reacted more strongly, positively or negatively, than new customers when service continuity broke down (McLelland & Goldsmith, 2014). Regulars have more history invested. They notice discontinuity faster than first-time visitors do. A platform architected around the anonymous first contact has no mechanism to detect that discontinuity because it never modeled the ongoing relationship in the first place.

Does personalization research back up why this matters commercially?

Ball, Coelho and Vilares (2006) found personalization drives loyalty specifically through trust built from repeated, remembered interactions over time, not from any single polished exchange (Ball, Coelho & Vilares, 2006). That mechanism requires a system that knows a regular customer is a regular. A chat platform that treats every session as a cold start cannot build the trust the research describes, regardless of how enterprise-grade its infrastructure is elsewhere. The webchat pioneer possesses genuine strength in large-scale, regulated messaging deployments. That does not change the equation. Scale and recognition are different engineering problems. The company's origin was built to solve the former.

Is it fair to hold a 1990s-era product to today's memory expectations?

It is fair to hold the current product to that standard. That is the actual question a buyer must ask. The historical origin explains why the architecture leans anonymous-first. It does not excuse a 2026 platform serving call- and appointment-driven businesses from carrying that legacy forward. The counterargument that "every platform started somewhere" is true and irrelevant. The question is not where the webchat pioneer began. The question is whether its current product recognizes a specific returning customer across visits, channels, and months. That is a testable claim, not a historical one.

What does this look like for an actual regular customer?

A family has bought two vehicles from the same dealership in three years and used its webchat both times to schedule service. The father messages again this spring about a recall notice. The chat opens with a generic greeting and a request for his VIN and contact information. This is the exact intake flow a first-time visitor sees. He has to explain his history again, dig out his old paperwork for the VIN, and re-describe a car the dealership has serviced twice before. The dealership has not done anything wrong by any compliance or security measure. The platform was simply never built to notice this is the fourth or fifth time this specific family has shown up. Its founding architecture treats every chat session as an anonymous, first-time event.

Isn't it unreasonable to expect a chat platform to replace a full CRM?

That is a fair-sounding objection. It is also answering a different question than the one at issue. No one is asking a chat widget to become a customer relationship management system. The question is narrower. Can the chat platform surface information the business already has at the moment a conversation starts? It should know this person bought two cars here or called about a recall last month, instead of treating every session as a blank slate. McLelland and Goldsmith's research on regulars found that established customers notice and react more strongly to discontinuity precisely because they have more history invested in the relationship (McLelland & Goldsmith, 2014). A platform does not need to be a CRM to avoid manufacturing that discontinuity. It just needs to check the record that almost certainly exists somewhere in the business's systems before starting from zero. We explore the same logic one level deeper: why digital-first is not customer-first. Where this goes next: message scale versus customer memory.

Anonymous-first design vs. regular-aware design

How should a call- or appointment-driven business test for this before buying?

Run the two-contact test above before signing a contract. It takes ten minutes and reveals more about a platform's actual architecture than any pedigree claim in a sales deck. See how this plays out against the webchat pioneer specifically on our comparison page. Then, read our related piece on memory as the real moat for businesses built on repeat customers.