Fourth installment of the recurring monthly case study series. One anonymized engagement per month — diagnostic, intervention, outcome. Names anonymized; numbers and timelines real.
The brand at intake
$19.4M TTM revenue multi-specialty healthcare group (urgent care + primary care + behavioral health + adjacent specialties across 6 clinic locations). 19% YoY (down from 47%). Channel mix: 48% Google Ads, 24% Meta, 14% SEO, 9% paid podcast/local audio, 5% referral programs. Reported PAC: $134. Telehealth adoption among new patients: 12%. Intake response time: 8 hours avg (range: 2-23 hours). Leadership running on platform-reported metrics only.
Stated problem: "Google Ads ROI keeps dropping. We need to cut paid acquisition by 40% and reinvest in SEO." Actual problem: tracking infrastructure was HIPAA non-compliant (active OCR enforcement risk), PAC was uncalculated for telehealth-vs-in-person paths, and intake was losing 46 patients per year to slow response times.
The Diagnostic (Days 1-30)
90-Day Audit — healthcare adaptation.
Z1 Data & Attribution: 1/6 (P0 — binding constraint). HIPAA Pixel exposure: Meta Pixel deployed on patient-facing pages (symptom checker, telehealth booking, condition pages) without a BAA. Standard GA4 receiving PHI-adjacent data. Active OCR enforcement target. No BAA-signed analytics infrastructure. No server-side tracking. Last-click attribution.
Z2 Acquisition: 4/8. Google Ads running broad-match on high-volume condition keywords with minimal negative-keyword discipline. Meta running on standard Pixel (also non-compliant). No telehealth-vs-in-person campaign separation despite ~3× PAC delta between paths. SEO/content underweighted given trust-vs-CPC math.
Z3 Creative Pipeline: 3/6. Generic clinic photography. No physician-byline content. No condition-specific patient education content optimized for Google AI Overviews or AEO citation.
Z4 Conversion: 2/6 (P0 — binding constraint). Intake response time avg 8 hours. Industry benchmark: under 15 minutes during business hours. Intake form 14 fields. No automated routing. Front-desk staff handling intake alongside in-clinic patient flow.
Z5 Retention: 3/5. Reactivation flows exist but not specialty-specific. Annual-recall outreach manual.
Z6 Operating Model: 3/5. CMO/Director of Marketing reports up to CEO and CFO. No four-metric leadership dashboard. No CMO ↔ Practice Director weekly review cadence.
The true-cost PAC math revealed the depth of the problem:
Reported marketing PAC (blended): $134 Plus intake-handling labor per acquired patient (slow response = high labor): +$22 Plus consultation no-show rate at 23% (lost time slot value): +$18 Plus telehealth/in-person mis-routing overhead: +$13 True-cost blended PAC: $187
Worse, the blended number hid massive variance: telehealth new-patient PAC was ~$71; in-person new-patient PAC was ~$291, a 4× delta. The business was scaling spend on in-person paths without realizing the channel-to-modality mismatch.
The Intervention (Days 31-90)
Rebuilt against the binding constraints in parallel.
Wks 1-2 · HIPAA-compliant attribution infrastructure. Removed Meta Pixel and standard GA4 from all patient-facing pages immediately (compliance triage). Deployed Freshpaint as the BAA-signed customer data platform. Routed conversion events through Freshpaint → server-side → CAPI/Google Enhanced Conversions (PHI scrubbed before transmission). Rebuilt conversion event taxonomy around HIPAA-safe categories (telehealth booking, consultation request, content download) — no condition-specific events touching protected categories. Result by end of week 2: HIPAA compliance audit clean. Attribution functional with proper scrubbing.
Wks 3-4 · Intake response time rebuild. Deployed dedicated intake team (2 FTE) separated from in-clinic flow. Automated routing via Twilio + structured intake form (14 fields → 4 fields). SMS-first response within 15 minutes during business hours; voicemail callback within 60 minutes outside. Telehealth booking direct-to-calendar for telehealth-eligible specialties. Result: avg response time 8 hours → 14 minutes. Intake conversion 17% → 31% within first month.
Wks 5-6 · Telehealth-as-funnel restructure. Separated Google Ads campaigns by modality (telehealth-eligible specialties vs in-person-only specialties). Telehealth campaigns: $1.40 average CPC vs in-person $4.20. Reallocated $24K/month from in-person condition campaigns to telehealth-funnel campaigns. Built telehealth-first patient journey: telehealth consultation → diagnosis → triage to in-person if needed. Day 18 telehealth-as-funnel mechanics applied directly.
Wks 7-12 · AEO + content + ops. Shipped 14 condition-specific patient education pieces with physician bylines and credential markup. Deployed FAQPage schema on top condition queries. Built structured authority content for behavioral health and chronic-care conditions. AEO foundational work — entity hygiene (NPI registry alignment, hospital affiliations on Wikidata), physician E-E-A-T signals. Migrated leadership to four-metric dashboard: true-cost PAC, intake conversion, telehealth-as-funnel adoption rate, payback at 12-month patient LTV cohort.
The Outcome (Day 90)
True-cost blended PAC: $187 → $94 (−49.7%) Telehealth-path PAC: $71 → $52 (−26.8%) In-person-path PAC: $291 → $148 (−49.1%) Intake response time: 8 hours → 14 min (−97%) Intake conversion rate: 17% → 38% (+21 pts) Telehealth funnel adoption: 12% → 41% (+241%) Consultation no-show rate: 23% → 11% (−12 pts) HIPAA tracking compliance: Non-compliant → BAA-signed (audit clean) AI search citation share: 11% → 33% (+200%) Annual patient revenue per new patient: $480 → $890 (+85.4%)
True-cost PAC nearly halved while compliance exposure was resolved. The telehealth-as-funnel restructure produced the largest single lift — separating the modality paths revealed that scaling in-person spend (the historical default) was the wrong allocation, while telehealth funneled to higher revenue per patient because patients who started on telehealth converted to multi-specialty engagement at much higher rates.
The board approved a Q3 spend increase contingent on holding intake response time under 20 minutes and PAC under $110. The CEO stopped describing Google Ads as "structurally inefficient" once the modality-separated PAC numbers became visible.
Three patterns worth internalizing
1. HIPAA Pixel exposure is the silent compliance crisis at most healthcare marketing operations. Active OCR enforcement has produced settlements against healthcare practices using Meta Pixel and standard GA4 on patient-facing pages. Most operators don't realize this is a violation until a notice arrives. The compliance remediation is also the highest-ROI tracking upgrade — BAA-signed infrastructure (Freshpaint, certain CDP configurations) produces cleaner attribution than non-compliant alternatives because PHI scrubbing forces clean event taxonomy.
2. Blended PAC hides 3-4× modality deltas in multi-specialty healthcare. Telehealth-path and in-person-path patient acquisition have radically different economics — telehealth typically $50-100, in-person typically $200-400 depending on specialty. Reporting blended PAC produces wrong-channel-mix decisions. Separate the modality math first; allocate second.
3. Intake response time is the highest-leverage conversion intervention in healthcare. A 5-hour delay costs 46 lost patients per year at average commercial valuation (~$200K revenue). Under-15-minute response separates top-quartile practices from median. Building dedicated intake infrastructure (separated from in-clinic flow) is operational rather than marketing work — but it produces the largest single PAC-effective lift in most engagements.
When this kind of engagement makes sense
If your healthcare operation has any combination of: Meta Pixel or standard GA4 on patient-facing pages, intake response time above 30 minutes, blended PAC reported without telehealth-vs-in-person separation, no BAA-signed analytics infrastructure, or rising patient acquisition cost without obvious channel cause — you're likely in the same multi-constraint pattern.
Start with the 90-Day Audit. The healthcare version weighs Zone 1 (HIPAA-compliant tracking) and Zone 4 (intake response time) most heavily. If you'd rather have an outside team install the compliant attribution stack and intake operating model alongside your in-house team, Praxxii Global runs this pattern across primary care, urgent care, behavioral health, multi-specialty groups, and telehealth-enabled practices. Free 60-minute diagnostic call before any commercial commitment.