Medical Content · SolveLetter #6

Who Is Your Deck Actually Talking To?

YP
Yakov Pakhomov, MD, PhD
September 2026 5 min read

Thursday, 20:40. An email from a medical advisor: “Yakov, I have 200 slides from HQ and a country brief due Monday. I don’t know what to cut.”

In ten years I have had that email perhaps fifty times. The cities changed and the molecules changed. Two things never did: “two hundred slides” and “due Monday.”

The industry’s standard answer is “translate it and cut it down.” It is also the worst one available. And the translation is only half of it: the question “what do I cut” is the wrong question to begin with. Nothing needs to be cut. Two hundred slides are not one bloated deck waiting to be trimmed. They are a warehouse you build several different decks from, for several different conversations.

A global deck cannot be translated into a local asset — it can only be rebuilt. Translation preserves the structure, and the structure is the problem: it was not built on your decision, it was not built around your competitor, and it does not answer your payer’s questions.

Three places where a translated deck dies

All three come from our work of the past few months.

Portfolio. A regional team receives the global deck from a major international congress. Accurate, beautiful, legally bulletproof. One detail: most of the slides are about drugs that are not on the market and will not be for years. There is nothing the field can do with it — the physician hears the congress news and asks, “so what do I prescribe on Monday?” The team sits down and builds the local layer itself: from abstracts and posters, with no designers, in two weeks, while also proving to HQ that the region is capable of it. This is not content localisation. This is re-manufacturing content with people who actually have another job. We regularly receive decks assembled this way for rework.

Argument. The global value story is built on cardiovascular outcomes: years of research, tens of thousands of patients, the best journals. An endocrinologist in a growing Central Asian market answers: “There are eight copies of the molecule on this market, and the key question is not ‘prove you are better than placebo’ but ‘prove you are better than the copy.’” The global argument is not weak. It simply answers a question nobody on that market is asking.

Language. A scientific review comes out before launch. Brilliant — and unusable: marketing and the field need the same content, but with claims legible at a glance from the slide, so a manager or a speaker can stand up and deliver without preparation. Translating from “scientific” into “field language” is the most underrated kind of localisation. It is not about vocabulary. It is about the person who will be facing an audience tomorrow.

Now the uncomfortable part: HQ is right

It would be easy to turn these stories into an affiliate complaint about global. But I hear the mirror-image pain from global teams too. Recently, from one of the largest companies: “Affiliates freestyle. One country goes one way on messaging, another goes another. It is all fragmented.” Two hundred slides is not an act of distrust. It is an attempt to hold a single line, to build one solution that works everywhere.

The paradox is that translation-as-localisation is exactly what produces that freestyle. When the material is irrelevant, the field does not argue with HQ — it quietly does its own thing. Veeva Pulse puts a number on it: nearly 80% of approved content is rarely or never used, and the share is growing. And it is not about slide quality. There is no metric that would show which content created by the global team was actually used and which was not.

The solution

A slide constructor lets you assemble 15 local mini-decks of 10 slides each from 200 global slides, every one built for a specific market question. That is what localisation and personalisation for a market actually mean. Rebuilding this way gives global more control than translation does. The affiliate gets relevance, global gets consistency. The only thing that loses is the illusion that you can hand 200 slides to the field, have them “translated,” and be done.

Eight questions to ask the HQ deck before you open PowerPoint

These are the tools and prompt ideas we use ourselves when we build a slide localisation plan for a region.

1. What local decision does this material serve? Who will do what after meeting it: prescribe, add it to a protocol, change a patient pathway? If there is no answer, this is not an asset — it is a slideument, or an archive.

2. How many patients do we have who need this decision? How do you map global epidemiology onto the local picture? How do the patients in our physician’s waiting room compare with the ones the deck was written for? Milder? More severe?

3. Does the evidence extrapolate? Population, doses, clinical practice. Which data travel automatically and which do not?

4. Are all the drugs and indications in the deck available on our market? What do the local guidelines say, and how do they differ from the protocols actually used on the ground? What gets struck out as unavailable, unnecessary or irrelevant?

5. What does the physician use in this situation right now? In the global RCT it is placebo; in practice it is the standard of care. So what do we do? And what if there is already a copy at a third of the price in the room?

6. Who is the slide addressed to, and what argument does that audience need? A budget committee does not read hazard ratios. And who actually needs the trial design?

7. How does this audience consume information? If email open rates among physicians are in the low single digits and they live in messengers, then format is part of localisation too, not just content. Perhaps it is reels and stories for a phone rather than slides.

8. Which 3–5 slides matter for this particular situation? Not “what do I cut” but “what do I take.” Every conversation gets its own mini-deck, and there can be ten of them. A rule of thumb: if you end up with one deck instead of several, you are still translating.

What this looks like in practice: we have collected typical situations and our cases on the site — global evidence that is not local-market ready, and a separate page with all our case studies, from evidence adaptation for CEE markets to value dossiers for MENA.

Colleagues on global and country teams — where does the line run for you between local relevance and losing control of the message? I read every comment.

And if you would like to take one of your own decks apart with us, here are 20 minutes in my calendar — no slides about us.

Newsletter
Get SolveLetter monthly in your inbox
Evidence strategy, advisory board insights, AI tools, and regulatory updates — curated for pharma medical affairs teams.
1–2 articles per month, never more
Practical insights from real projects
Free forever, unsubscribe anytime
Subscribe to SolveLetter
Join 1,200+ pharma professionals. No spam.
Evidence Scanner
Evidence ScannerTM
AI infrastructure

AI-powered.
Expert-validated.

We built AI workflows into our daily practice — not as a marketing claim, but as the infrastructure that lets our medical experts deliver faster without cutting corners.

Research
Structured PubMed queries with narrative or table outputs
Monitoring
Weekly literature digests by drug, target, or topic
AI-Enhanced EDC
Advisory board transcription + structured AI summary
Fact-Checker
Claim verification against your source documents
AI accelerates. Our experts validate.
Every output goes through expert medical review before it reaches your team. AI handles structure and speed — we handle scientific judgement and MLR readiness.
Evidence Scanner · Monitoring module
// Weekly digest: GLP-1 RA publications
monitor("GLP-1 receptor agonist", {
  frequency: "weekly",
  sources: ["pubmed", "congress_abstracts"]
})
Scanning 12 sources...
Weekly Digest · Feb 24–Mar 2
7 new publications found. 2 RCTs, 3 RWE studies, 2 meta-analyses. Key finding: MACE benefit confirmed in CVOT pooled analysis...