The challenge
The group ran twelve oral surgery locations and watched the metric every multi-site operator watches: total case volume. It looked fine. But an oral surgery practice lives on referrals from general dentists, and a healthy aggregate can sit on top of a referral base that is quietly narrowing. The leadership had no way to see whether the volume was coming from a broad, resilient set of referring GPs or from a shrinking few working harder — because the number they tracked, by design, couldn't tell them.
The approach
Innovation Park reconstructed the full referral history and scored every referring GP against its own baseline cadence, then mapped concentration across the twelve locations. This is the step the aggregate cannot perform: instead of one case-volume number, the group got a relationship-level picture — which GPs used to refer and had gone quiet, and how much of the network's volume depended on how few sources.
The result
The scan surfaced two findings that reframed the group's growth question. First, dormancy: 34% of historically active GP referrers had sent zero cases in 90 or more days — a third of the base, gone quiet without ever being formally lost. Second, concentration: just 18 practices were carrying the network, a fragility invisible in the healthy-looking total. Together they showed the group's growth wasn't a marketing problem; it was a dormant, recoverable referral base and a concentration risk that needed managing.
