Brand analysis rebuilt for every request became a configured pipeline.
Holmusk, NeuroBlu Analytics · January to June 2026.
Is this you?
A healthcare data company’s commercial team needed brand-specific analysis for pharma conversations after each data release.
- Project
- Each brand analysis was custom analytical work that took four to eight weeks.
- Why now
- Commercial colleagues kept asking for the same analysis for one drug after another. The one dashboard built by hand could not keep up, and each new drug took a data scientist away from other work.
- What they tried
- A data scientist built one brand dashboard by hand. The data team ran its release decks from workbooks, with a drug or diagnosis as the input and Excel templates linked to PowerPoint.
- Why it fell short
- Every new drug meant another build by hand, and the decks still waited on one person. After one data release, the decks were two to three weeks late.
about 45 seconds per configured drug from the initial pipeline.
Initial-pipeline output, not an end-to-end rebuild time. The earlier custom work took four to eight weeks per brand; the two are not the same measure.
One repeatable pipeline: branching queries, structured configuration, product logic and branded outputs in PowerPoint and PDF.
Kevin: Built the initial configurable pipeline, reusing the data team’s earlier analyses.
The team: The data team’s earlier analyses were the base. Commercial colleagues took the outputs into pharma conversations.
A product analyst on Kevin’s team reviewed and merged changes in the team’s analytics toolkit. Commercial and training teams used the outputs.
Not end-to-end. Configuration and review are separate steps, and four to eight weeks describes the earlier custom work, so there is no speedup ratio. No staffing saving or revenue was measured.
Experience Library, “2. Brand Dashboard” and the October 29 Granola review · Experience Map H7 · resume 2026-08-25 · meeting notes: Daily Standup 2026-01-23, data-team review 2026-02-27, data-team review 2026-05-13, Product Team 2026-05-19, Platform Leads 2026-07-29.
Bring the one your team repeats.
If this is the shape of what your team rebuilds by hand, the next step is a direct conversation.
The Data Product Builder
What I’m building, what broke, and what I’d do differently. Healthcare data products, decision infrastructure, and the messy middle of building alone.
No spam. Unsubscribe anytime.