Patterson — turning a Contact Us form into a revenue channel in six weeks.

NCR.com had a real traffic problem that wasn’t being converted. Visitors landed on product and solution pages, found what they needed, and then routed themselves to a Contact Us form — a static dead end that handed them to a sales queue with no context, no qualification, no immediate engagement. The form worked, in the sense that it collected names and email addresses. It didn’t work in the sense that it created any momentum.

Chatbots were an obvious category to consider, and Drift was an established platform with Salesforce integration, A/B testing, and lead routing already built in. But NCR had none of the surrounding infrastructure — no playbooks, no trained team — and a calendar that left roughly six weeks before the next significant revenue window. That was the constraint: ship something production-ready in six weeks, or wait another quarter.

Delay was expensive. The hospitality vertical was in an active sales cycle, the Restaurants site had just relaunched, and visitors were arriving with intent. Every week without an engagement layer was pipeline left on the floor.

The first question wasn’t what to build. It was what the site’s traffic was actually doing. Held against overall traffic, the Contact Us conversion rate made one thing plain: a large share of visitors never reached the form at all. They browsed, they left, they didn’t convert — and some of them were qualified. The form asked for too much intent; it selected for people already far into a buying decision.

A chatbot changes that dynamic. A prompt at the right moment can turn passive browsing into a conversation before someone decides to leave. Drift let me script playbooks for specific paths — a visitor on the restaurant-technology pages would meet a different entry point than one on the corporate homepage. That kind of segmentation simply wasn’t possible with a static form.

The real decision was launch breadth versus launch quality. NCR runs several verticals; standing a chatbot up across all of them in six weeks was not realistic if the goal was a clean, tested experience. I scoped the launch to hospitality and the general NCR experience, with the explicit intent to build a business case for expansion if the numbers came in.

I built and launched Patterson in six weeks, from initial platform configuration through production deployment on ncr.com. It shipped with scripted conversation playbooks for the hospitality vertical, lead capture with email collection, Salesforce integration for routing and attribution, and A/B testing across engagement variants.

Patterson handled two kinds of interaction: scripted playbooks for visitors on known paths, and freeform questions from those who wanted to ask something specific. Freeform was the weak point at launch — without an AI FAQ layer, off-script questions got thin answers. I documented that limitation from the start and folded it into the expansion business case rather than letting it surface later as a surprise.

At launch it ran alongside the Contact Us form as a parallel path. Within eleven weeks it was handling 26% of all conversions to the form’s 74%.

The numbers made the case for what came next: a Drift AI FAQ module to handle freeform questions at scale, and Banking and Retail MVP chatbots — positioning Patterson as the proof of concept for a company-wide conversation layer rather than a hospitality experiment. I locked in the two-year Drift renewal at existing pricing before year-end, protecting the program budget before rate increases could apply.