Gerimed × BluPax — Cross-Reference Analysis

Headlines at a glance

Cross-reference of Gerimed GPO members against BluPax customer base, with 16 months of chargeback-aware sales detail.

How much overlap is there?

1,505 matched DEAs out of 3,731 active Gerimed members. DEA is the only practical cross-reference key.

• Match rate is uniform across bed-count buckets (36–43%) — facility size didn't drive who became a BluPax customer in the past.
• HIN fallback adds zero matches (BluPax HIN coverage 0.8%). DEA is the only operative cross-reference key.
• 3 active Gerimed members have no DEA on file — uncrossable by any method.

What kind of relationship do we have with the matched 1,505?

Split by most recent invoice date at the matched-DEA level.

Question 2 16-month sales — revenue and margin per customer

"How are we doing with them? How many are buying, how much, how many items per customer?" 16-month invoiced sales, chargeback-net.

Per-customer order shape (median across 603 buyers)

Top 10 matched customers by 16-month revenue

⚠ Valu Merchandisers (top) is a regional wholesale intermediary serving 7 Gerimed-listed pharmacies (MO/OH/TN). Real revenue, but attribution is one step removed from end pharmacy.

Brand + Vaccine visibility

Brand and vaccine items in the matched Gerimed cohort — 56% of cohort revenue.

Question 1 Flu vaccine prebooks

"How many prebooks do these customers have for flu vaccines (Fluad / Flucelvax, vendor Seqirus)?" Open backorders via Sales_Order_Open_Items_List.

Bed-count correlation: still none (Mann-Whitney p=0.71). Flu prebook is a vendor-relationship decision, not a facility-size one.

Question 3 Bed-count layer — what does it tell us?

"Notate beds served. What does >100 beds mean? Do they have a large bed count but low BluPax sales?" Match rate, buy rate, and revenue per buyer by bed-count bucket.

Top 100+ bed Gerimed members with ZERO 16-mo revenue (largest prospecting tier)

1,329 matched-or-unmatched Gerimed members with 100+ beds have ZERO BluPax revenue in 16 months. Full list in the XLSX tab 9.

Question 4 Sales-rep gap — coverage analysis

"Maybe bed count doesn't matter because there's a sales-rep marketing gap." Buying rate per rep on matched Gerimed accounts. The 16-month "any buying rate" is the fair measure; the 3-month "recent" column shows current cadence.

Spread between top performer (D'Attilo 69%) and median (~40%) is 30 percentage points. Closing half that gap on the Blupax Pharma house-bucket (360 accts at 39.7%) = potentially 50+ additional active customers.

Growth signal: first 8 months vs last 8 months

"How are we doing — growing month over month?" Per-customer change between P1 (Jan–Aug 2025) and P2 (Sept 2025–Apr 2026).

• +$5.0M growth, -$1.5M attrition — net cohort change favors growth by $3.5M.
• 151 new customers added in P2 vs. 78 lost from P1. +73 net new customers.
• The growth is concentrated in the COVID vaccine push (Sept-Oct 2025). The post-vaccine baseline (Q1 2026) is a smaller, more recurring business.
• P1 vs P2 split is a synthetic comparison — periods aren't seasonally adjusted. With one more year of data we could compare same-period year-over-year.

Question 5 Seasonality — are purchases seasonal or consistent?

Monthly cohort revenue + net profit across all 16 months. Baseline run-rate ~$500K/month at 20–25% margin; the COVID vaccine surge dominates Sept–Oct 2025.

Reactivation targets

Matched customers by recency of last invoice. The reachable reactivation universe is the cooling + dark buckets (last 4–12 months).

The untapped 2,226 — prospecting universe

Gerimed members with no BluPax customer record at all. State distribution below; full ranked list (by beds) lives in the XLSX.

CA stays the standout single-state opportunity. Full unmatched list (2,226 rows ranked by beds) in the XLSX tab 4.

Methodology — how it was built

The audit-trail layer. Every number traces to a specific transformation.

Cross-reference chain

Gerimed roster (3,731 active members with DEA)
   ↓ DEA join (HIN fallback adds zero)
Customer_Ship_To_Addresses (18,380 unique BluPax DEAs)
   ↓ BP Code join
Customer_Info_Report (channel / group / rep — NOT for sales — Sales rollup unreliable)
   ↓ BP Code join
Sales Detail with Chargeback (16 monthly CSVs Jan 2025–Apr 2026, 608,643 real data rows)
   + Sales_Order_Open_Items_List (Fluad / Flucelvax prebook backorders)

Net Profit formula (canonical)

Net Profit per line = (Ext Price − Ext Cost) + Ext Flat Rebate + Ext. % Rebate + Ext % Cntrt Cost Rebate

• Non-chargeback lines: rebate columns are 0, formula reduces to standard Gross GP.
• Normal chargeback lines: Apparent GP captures contract-aware margin, rebate columns add manufacturer incentives.
• WAC anomaly chargeback lines: Apparent GP captures the windfall naturally (no date-flagging needed). Verified on Comirnaty Jan-Mar 2026 (31-34% margin vs. 7% baseline).
• DO NOT add Final Ext. Rebate to this formula — that's the chargeback CLAIM amount (includes neutralization), not profit.

Other rules

• CMs negated: Document Type='CM' rows are stored as positive but represent reversals. 3,508 rows negated → $5.2M revenue reduction.
• Trailing empty pad rows filtered: SAP exports pad files with blank rows. Filter on DocNum non-empty.
• Comma-stripping: thousands-separator commas in number columns silently zero out 2.6% of rows if not stripped before numeric conversion.
• String-sort on date columns lies: always parse to datetime before .min()/.max(). Three trap-pairs documented in [[LESSONS_DATA]].

Audit results

Where this goes next

The reusable pattern

Same shape (GPO roster → DEA cross-reference → chargeback-aware sales → analytical layer) applies to:

• SPS Health — files already loaded in Reference/Avi/.
• Any future GPO partnership.
• Any "find me my best customers in [X cohort]" question.

Worth noting: this analysis is gross of expenses (shipping 8–9%, CC 3%, commission 2–3%). True bottom-line for the cohort after expenses is ~7–9% net-net, vs. the company average ~5–7%. Gerimed remains an above-average-margin cohort even after expense load.