Evaluate a pharmaceutical pricing data provider on the questions your team already asks, not on a dataset the vendor has chosen. That means testing coverage across the markets you work in, the depth behind each figure, how fast a data query comes back with a usable answer, and what the commercial model does as your team grows. A demo shows you the platform. Those four take longer to establish.
That test can be difficult to run, and the reasons are structural rather than anything to do with the providers themselves, so we’ve outlined some steps you can follow below to get the best out of your current setup.
Why does pricing data rarely get evaluated properly?
A pharmaceutical pricing data provider is usually bought on an annual or multi-year subscription. Once it is in place, three things make a genuine comparison difficult.
The first is timing. A meaningful evaluation takes weeks, because you need real questions to arrive in the normal course of work rather than questions invented for a trial. Renewal decisions rarely allow that much runway, so the choice arrives as renew or do not renew, with no third option in front of it.
The second is cost. Running two providers at once means paying twice for overlapping coverage, which is difficult to justify to a budget holder who has already approved the incumbent. Cancelling first and evaluating afterwards means going without pricing intelligence during the gap, which is difficult to accept when a submission or a launch date is already in the diary.
The third is dependency. When one provider supplies the price references that feed your models, your reporting and your payer submissions, there is no independent view against which to check them. If a figure is wrong, or a market is thinner than it appears, the gap surfaces when someone external challenges it rather than when you could have done something about it.
None of this is an argument that your existing data set is wrong. It is an argument for running the comparison while you still have both, so you can see where the gaps sit.
What should a fair evaluation test?
A demo generally runs on a dataset and a set of markets chosen in advance. An evaluation starts from your own work instead, which is why the two answer different questions. Seven things are worth testing on any pharmaceutical pricing data provider.
- Coverage where you operate. Not the headline market count, but whether the markets in your launch sequence are covered at the granularity you need, including the smaller ones that tend to be thin.
- Granularity of the price itself. Pack level or country average, which price categories are carried, and whether historical series exist far enough back to show how a market has moved.
- Interpretation attached to reference pricing rules. Raw international reference pricing rules are widely available, and the WHO guideline on country pharmaceutical pricing policies sets out how external reference pricing is applied across jurisdictions. What decides whether a rule changes your price is the country-level conditions around it: which countries sit in the basket, which price type is referenced, how often it is revised and what the exemptions are. Check whether that analysis comes with the rule or is left to you to decipher.
- Sourcing you can show someone. Every figure that reaches a payer submission or a price sign-off should be traceable to a source you can name when challenged.
- Cost of treatment method. Whether it is calculated from dosing and duration or inferred from pack price. The two produce materially different numbers and only one of them survives scrutiny from a health economist.
- Response time on a real query. Send your hardest open question and time the answer. Then judge whether the answer came from someone who understood the question or from someone reading a database.
- What the commercial model does over time. Whether the price is agreed once for an agreed scope, or grows through modules, seats and add-ons as your team’s use of the platform increases.
Run those seven against your current pharmaceutical pricing data provider, before the renewal decision rather than after it.

Run those seven tests against us
Keep your current provider and run GPI alongside it, on your own markets and therapy areas, with no charge for the evaluation period. Nothing gets cancelled and there is no gap in coverage.
How long should an evaluation run?
Long enough for real work to pass through it. A fortnight tends to produce impressions rather than evidence, because the questions that expose a dataset’s limits are the awkward ones, and awkward questions arrive on their own schedule. Run the evaluation alongside an existing subscription rather than compressing it into a renewal window, so those questions arrive in their own time rather than being invented for a trial.
What if you do not have a provider today?
Where there is no provider, the gap sometimes gets filled with spreadsheets, free public sources and occasional consultants, and that arrangement works until a launch price, a licensing decision or a payer submission needs figures that will survive being questioned.
The same seven tests apply when you are choosing a pharmaceutical pricing data provider for the first time. The difference is what you compare against. Rather than running two providers in parallel, take one asset and one set of markets where you already know what a good answer looks like, and see whether the output matches. If it does not, you have learned something cheaply. If it does, you have a defensible basis for the decision in front of you.
Where GPI sits on this
We have had the same discussion with enough prospective clients that it seemed worth publishing this article.
Global Pricing Innovations works only in pharmaceutical pricing and market access. That focus and area of expertise is the reason we are comfortable offering this evaluation: our data carries expert interpretation rather than raw figures, our clients work with a named Customer Success Manager rather than a shared queue, and the price is agreed once for an agreed scope rather than growing module by module.
So we have removed the cost of finding out. If you are under contract with another provider, run GPI alongside it through to the end of that contract, at no charge for the evaluation period, and compare the two on your own markets and therapy areas. If you have no provider, bring us one asset and one set of markets and we will run it and show you the output. If you would rather see the data before any conversation, ask for a sample report built on the markets you care about.
None of those routes asks you to cancel anything.
Get a second opinion on your pricing data – find out more about our latest offering to help you analyse your current position and achieve more from your pharmaceutical pricing data and tools.
FAQs
A: Test the provider on your own questions rather than a vendor demo. Check coverage in the markets you operate in, whether prices are held at pack level with history, whether international reference pricing rules arrive with country-level interpretation attached, whether every figure carries a traceable source, how cost of treatment is calculated, how quickly a difficult data query is answered, and whether the commercial model stays fixed as your team’s usage grows.
A: Long enough for real work to pass through it. A two-week trial produces impressions rather than evidence, because the questions that expose the limits of a dataset are the awkward ones and they arrive on their own schedule. An evaluation running alongside an existing subscription over several months tells you more than one compressed into a renewal window.
A: Yes, by running them in parallel rather than sequentially. A parallel run avoids any gap in coverage and removes the need to wait for a renewal window, so the comparison happens on live work rather than on a sample dataset. The obstacle is usually cost, since it means paying twice for overlapping coverage during the evaluation.

