Building a Measurement Framework That Finally Answered “What’s Working?”

The Challenge

The firm's leadership team had a problem they knew about but hadn't been able to solve: they were spending nearly $400K a year on marketing with no reliable way to tell which channels were actually producing new clients and which were producing the appearance of activity without business impact.

Each vendor reported their own metrics — clicks, impressions, event attendance — but none of those metrics connected to actual client acquisition. The firm's CRM tracked new clients and revenue, but there was no bridge between the marketing activity records and the CRM outcomes. When the CEO asked which marketing channels she should invest more in for next year, the honest answer was "we don't know."

The pressure to make better decisions was growing. A competitor had entered their primary market. The referral pipeline that had carried the business for years was showing signs of softening. The window for getting clarity on what was working was narrowing.

Our Approach

Attribution for a service business with long sales cycles is genuinely complex — clients don't click an ad and immediately sign an engagement agreement. The journey from first contact to signed client could take months and involve five or six different touchpoints across digital, in-person, and referral channels. Our measurement framework had to reflect that reality, not oversimplify it.

We started by mapping the full client journey for the last three years of new client acquisitions — working backward from signed engagements to identify every recorded touchpoint we could find. This produced a dataset of 340 new client journeys, each with a partial or complete record of how that client first heard about the firm, what interactions occurred before the initial meeting, and what ultimately converted the conversation into a signed engagement.

From that foundation, we built a multi-touch attribution model that distributed credit across touchpoints using a time-decay weighting approach — giving more credit to the interactions closer to conversion while acknowledging the role of earlier awareness channels. We connected this model to a live data pipeline that updated weekly as new client acquisitions were recorded.

The final layer was an executive dashboard — a single view that showed cost-per-acquired client by channel, conversion rate at each stage of the journey by lead source, and a rolling 90-day projection of pipeline value by channel investment level.

The Results

Within six weeks of engagement start, the firm had its first attribution-based view of marketing performance. The findings were both validating and surprising. The referral program, which had felt expensive relative to its visible activity, was generating clients at a 40% lower cost-per-acquisition than any digital channel. The event program was producing top-of-funnel awareness but rarely contributing to conversion — a useful result to know, even if it wasn't what the events team wanted to hear.

Most significantly, a digital campaign that had been running for 18 months and consuming nearly $80K annually was producing almost no attributable client conversions. It was generating clicks and traffic, but the traffic wasn't converting, and the firm had no way to see that in the vendor's reporting.

Over the following six months, the firm reallocated $200K of its annual marketing budget based on the attribution data — increasing investment in referral programs and targeted LinkedIn outreach, reducing digital display, and restructuring the events calendar to focus on higher-conversion formats. Cost-per-acquired client improved 31% year-over-year. The CEO had the answer she'd been asking for.