SAMPLE: Upstream Bio × BI Consulting Services — Claims-Data Market Insight Board

Where do the treated patients sit in the airway — and how many reach a biologic?

A single Power BI model over refreshed medical, pharmacy and eligibility claims: diagnosed prevalence, line-of-therapy flow, biologic share, payer mix and site of care for severe asthma, CRSwNP and COPD — the three indications in verekitug's development plan. Filter by indication, US census region and claims refresh.

1Line-of-therapy flow, drawn as the airway

Each branch is a patient segment; branch width is proportional to patient count. Trachea = diagnosed population.

2Biologic-treated patients, trailing 24 months

Monthly count of patients with ≥1 biologic claim in the month, vs. planning baseline.

Actual (claims)Planning baseline

3Biologic share by mechanism

Share of biologic-treated patients by target class (pharmacy + medical J-code claims).

4Payer mix of biologic-treated patients

Primary payer on the most recent biologic claim.

5Prescriber specialty & site of care

Rendering-provider taxonomy on biologic claims, plus the HCP count behind each.

6Treated patients per 100k population, by region

Treated = ≥2 indication-coded claims and ≥1 maintenance-therapy fill in the refresh window.

7Claims feed refresh status

Structure for the refreshed claims drop: load counts, completeness against expected volume, and script run state.

Source	Rows loaded	Last refresh	Completeness	Script	State

8What this board answers

Three new-product-planning questions, each answered from the current filter.

9Ask the data

Natural-language layer over the same model (Copilot-style). Illustrative response.

Private AI on your own server

The "ask the data" panel above is built to run on a GPU server inside your own network: an open-weight model over your own documents, with every answer citing the file it came from. No per-seat licence, works offline, and nothing leaves the building.

How it sits in your environment

One appliance-class server on your rack or in a secured closet. Your files are indexed where they already live; the model runs locally; answers come back with the source page attached.

What that means in practice

1
Your data only, on your hardwareAn open model over your own documents and exports. No third-party AI processor, no data-transfer agreement to review.
2
Every answer cites its sourceReplies point to the file and page they were drawn from, so people check the answer instead of trusting it blind.
3
No per-seat licenceThe whole team can ask questions. Cost is the server and our build, not a monthly number that grows with headcount.
4
Works offlineIf the internet connection goes down, the answers still come. Nothing depends on a vendor's uptime or pricing changes.