Insights

Notes on doing it properly.

Short pieces on methodological decisions that change the outcome of a study. Written for whoever will use the figure, not for whoever produces it.

• Methodological note · 2026

  Why a sample of 150 physicians cannot read six segments

  The most common error in pharmaceutical research is not in the field, it is in the cut. How to size a sample when segment-level analysis is the real objective.

  Request the document

• Analysis · 2026

  The eligible universe is not the prevalence

  The distance between population prevalence and effectively treatable patients is often an order of magnitude. How to build the funnel without inheriting unsourced figures.

  Request the document

• Methodological note · 2026

  Trackers that celebrate noise

  A three-point variation with n = 200 is almost never significant. What a brand tracker must report for the wave reading to be defensible.

  Request the document

• Analysis · 2025

  What the public sector changes in prescribing habit

  Averaging IMSS, ISSSTE, IMSS-Bienestar, state services and private destroys the most useful information in the study. Evidence for why sector stratification is mandatory in Mexico.

  Request the document

• Point of view · 2025

  The value dossier starts before the model

  Most evaluator observations are not about the economic model but about the evidence feeding it. How to order the file from the start.

  Request the document

Full documents are sent by email on request. If a topic you care about is not on the list, say so and we will consider it for the next piece.