DX / Differentials / EYE CARE GROUPS

D-04 — MULTI-LOCATION MARKETING · EYE CARE GROUPS

One playbook, different results by location — eye care

Same brand, same frame boards — and three times the revenue per exam between offices

DIFFERENTIAL — LIKELY CAUSES, MOST LIKELY FIRST

1. 01

   Revenue per exam is an operating variable, not a scale or market variable

   The industry's own benchmark data shows a very large spread in revenue per exam that is not explained by practice size. That means a site's number is telling you about how that site operates, not about the town it sits in — which is the whole reason cross-site comparison works here.

   Gross revenue per complete exam: median $306, top decile $500 (top 5% $529), bottom decile $159 — a 3.1× spread on identical exam volume. The publisher states it plainly: "There is only a weak positive correlation between practice size and revenue per exam." It calls the metric "perhaps the single most useful measure of practice productivity."

   MBA "Key Metrics," pp.5–6 — n>1,900 US private optometric practices, full decile distribution published. 2015 edition, data largely 2010–2012: use the decile placement, not the dollar level.

   NOT THIS IF — Your inter-site spread tracks median household income or panel age cleanly across every location. Then you are looking at a demographic difference and the fix is per-site product mix, not a uniform playbook.

2. 02

   Capture rate differs by site — and it, too, is size-independent

   The percentage of exams that convert to an eyewear purchase is set by the handoff, the board and the people, all of which are local. Two sites running the same brand can be 30 points apart on capture and nothing in the group's reporting will show it, because three-quarters of practices do not measure capture at all.

   Eyewear Rxes per 100 complete exams: median 61, top decile 109, bottom decile 30 — and "the ratio does not vary significantly by practice size." Only 23% of practices ever track it. Modern transaction-derived reference: 60.1% capture across 1,000 practices and 7M+ exams in 2024.

   MBA "Key Metrics," p.21 (n>1,900, 2015 edition) and GPN Technologies, GPN VISIONS (2024). The two use different denominators — MBA counts eyewear Rxes ÷ complete exams; GPN counts purchasing patients ÷ examined patients. Compare each site against the same denominator, never against the other one.

   NOT THIS IF — Capture is within a few points across sites and the variance sits in exam volume instead. Then look at recall and OD-hour utilisation.

3. 03

   Payer mix differs by site, and payer mix moves both the fee and the capture rate

   A site with a heavy managed vision care panel collects roughly half the exam fee of a direct-pay site — but converts eyewear at a higher rate. A site with an older self-pay panel does the reverse. Two mechanical effects pulling in opposite directions, neither of which is visible in a blended group report.

   Share of exams carrying a managed-care discount runs from 9% at the 5th percentile to 95% at the 95th, median 65% — a genuine order-of-magnitude spread across practices. Collected exam fee: $66 insured against $134 direct-pay. On the capture side, 66.0% for patients with a vision plan against 54.3% self-pay; at 65+, 70.6% with a plan against 37.3% self-pay.

   MBA "Key Metrics," pp.17, 19 (n>1,900) for the discount distribution and fee gap; GPN Technologies (1,000 practices, 7M+ exams, 2024) for capture by payer and age. The primary notes directly that "the sources of payments vary widely across practices."

   NOT THIS IF — Payer mix is within a few points across all sites. Then the fee gap is not driving the spread and you should look at product mix and process.

4. 04

   OD-hour utilisation differs by site

   The same doctor hours produce very different exam counts depending on the schedule template, the pre-testing workflow and the technician support. This is the least glamorous cause and the most common — and it shows up as revenue per OD hour, which is the number that actually pays for the doctor.

   Complete exams per OD hour: median 1.10, top decile 2.18, bottom decile 0.50 — against a 15–20 minute exam, which the publisher reads as "much excess capacity in typical optometric offices." Gross revenue per OD hour: median $330, top decile $610. Within solo practices, $750K+ offices run 1.48 exams/hour against 0.76 below $500K. The primary notes a productivity plateau above roughly $650,000 in annual revenue that holds until about $2 million.

   MBA "Key Metrics," pp.7, 9–10 — n>1,900. The 1.10 median is independently corroborated in the primary against the AOA's median for all practicing ODs and against Practice Advancement Associates' corporate-OD surveys.

   NOT THIS IF — Exams per OD hour is within about 0.1 across every site. Utilisation is uniform and the spread is downstream of the exam.

5. 05

   Recall discipline differs by site — and recall is also size-independent

   Each site has its own active-patient file and its own habits around pre-appointing and recall outreach. Because the metric does not correlate with size, a group's biggest office is as likely to be its worst recaller as its best.

   Complete exams per 100 active patients: median 43 — a 28-month average interval — with a top decile of 76 and a bottom decile of 22. The publisher: "There is no correlation between this ratio and practice size and the ratio has not changed over six years of measurement." Annual gross revenue per active patient follows: median $133, top decile $270.

   MBA "Key Metrics," pp.11–12 — n>1,900, 2015 edition.

   NOT THIS IF — Recall ratios match across sites within a few points. Then the variance is in what happens during and after the exam, not in getting patients back.

HOW TO TELL THEM APART

How to tell these apart in your own numbers

Each of these is a measurement you can run yourself, without us.

WHAT RESOLVES EACH

What resolves this, and how you will know it resolved

Revenue per exam is an operating variable, not a scale or market variable	Rx 04 · marketing attribution →	One readout, identical fields, every site, monthly — placed against the published decile table rather than against the group average. An internal average tells a site it is fine relative to its siblings; the decile table tells it where it sits in the industry.
Capture rate differs by site — and it, too, is size-independent	Rx 03 · patient conversion →	Capture variance across sites running one playbook is almost always handoff variance. Instrument the transition from exam room to dispensary at each site and compare the process, not the personalities.
Payer mix differs by site, and payer mix moves both the fee and the capture rate	Rx 04 · marketing attribution →	Segment before you conclude. What does not work: applying one capture target across sites with materially different payer mixes — the plan-heavy site will beat it on capture and lose on fee, and the team that is actually underperforming will be invisible.
OD-hour utilisation differs by site	Rx 03 · patient conversion →	Schedule template, pre-testing workflow and technician support are the three levers, and they transfer between sites cleanly because they are process rather than market. Note the published plateau: OD productivity per hour flattens above roughly $650K of annual revenue and does not move again until about $2M, so a site already at the plateau will not be fixed by template work alone.
Recall discipline differs by site — and recall is also size-independent	Rx 03 · patient conversion →	One recall engine for the group, one active-patient definition, one cadence — rather than each front desk working the list when it has time. This is the cheapest same-store growth in the vertical and it does not require any media at all.

HOW THIS DIFFERS BY SCALE

How this differs by scale

Single site	There is no multi-site variance, but there is provider variance and it uses the same tests. Run exams per OD hour, capture and revenue per exam by doctor and by optician. A two-optician office with a 25-point capture spread has the same problem as a four-site group, at one address — and the same three numbers do the whole job in multi-office eye care groups.
Group	This is the native scale for the problem. The three metrics that matter most — revenue per exam, capture rate and recall — are all stated by the primary to be independent of practice size, which makes cross-site comparison genuinely diagnostic rather than merely descriptive. Standardise the active-patient definition and the capture denominator before comparing anything.
Platform	Variance arrives with the acquisitions. Each acquired practice brings its own payer mix, its own dispensary square footage (median 25% of office space; smaller practices run 30%+, $1.5M+ practices 22%; under 500 square feet risks being perceived as a limited frame selection) and its own recall habits. Expect the spread at close and underwrite the compression, not the average — which is the difference between an owner-operator's arithmetic and a consolidator's underwriting: the same earnings price at roughly 2.5×–4.5× SDE at single-OD scale against 10×–14× adjusted EBITDA at platform scale, per a compilation of broker and advisory data.

OTHER PRESENTATIONS — EYE CARE GROUPS

• Exam volume is flat — and the unserved demand is already in the recall file
• Cost per exam is climbing — and the exam is worth less than it was
• The exams happen. The purchase happens somewhere else.
• The earnings are real. The system that produces them is not visible.