Thirty years on referrals. Then a site people actually call from.

Situation

At around $50M a year, Vitality is the largest firm we work with and the one that had done the least about it. Licences, trucks, a thirty-year name in commercial mechanical across metro Detroit, and all of it built on word of mouth. They had never really done marketing. The website was a GoHighLevel template somebody had set up years earlier and then left alone: no tracking on it, no campaigns pointing at it, no reason for anyone who did not already know the name to find it. Work came in by phone from people who had used them before. Growing meant waiting for the next referral.

So this was not a fix. Before us there was nothing online to fix. It was the first online presence the business had ever actually had.

Strategy

We put the bulk of our fee on the outcome rather than on the hours, and said so on the first call. Robert had spent thirty years not paying for marketing; the only version he was going to try was one where we did not get paid unless it worked.

Then we built the whole thing from the ground up, in order.

First, the site. Not a redesign of the old one, a replacement: fast, one phone number that rings dispatch, a page for each kind of work they do rather than one page that says "commercial HVAC" and stops, and a quote path for the planned work where a mechanical contractor's margin actually lives. Built to be landed on by a stranger and called from, which is a different job from a site built to be shown to someone who already knows you.

Second, the data layer under it. Every call, form and quote view on the new site emits a structured event carrying the equipment type, the building type and whether the enquiry is a breakdown or a planned replacement. The old site sent nothing anywhere. This is the part nobody sees and the part everything else stands on: without it, an ad platform can tell that something happened and nothing about what it was worth.

Third, the first paid search campaigns the company had ever run. Emergency demand and planned replacement are the same lead on a form and wildly different jobs to a contractor, so they were separated, budgeted and bid as two businesses from day one, each with its own tracked number.

Identity resolution came after, once there was traffic to work with. Matching people across devices and sessions means a facilities manager who priced a replacement in March can be reached in the buying window rather than the week after, and anyone already on the books drops out of the audience instead of being paid for twice.

The loop is the point. Richer intent going into the platforms means the optimisation is chasing invoiced work rather than form fills, return on ad spend rises because the budget is aimed better, and our share is taken out of that increase, which is why the reporting is built around it. None of that loop existed before the site did.

Results

Revenue growth of +25% across 2 months, on a business doing around $50M a year, from a standing start: there was no paid channel and no measurable online demand to build on. Same trucks, same technicians, same phone number. A single-digit share of tracked new revenue came to us out of that growth, which is the part of our fee that exists only if the number moves.