NHS digital programmes do not fail on dashboards. They fail on deployment

NHS digital programmes do not fail on dashboards. They fail on deployment

scarce resource. It is not. Between ePACT2, NHSBSA prescribing dashboards, national medicines optimisation metrics, Open Data Portal releases and the Secondary Care Medicines Data set, the service already has a substantial public-data base for identifying variation, price pressure and switching opportunities. NHSBSA’s own documentation is strikingly honest about this. The national medicines optimisation opportunities dashboard says its metrics are only a starting point and that further national and local data, alongside more sophisticated comparisons and local knowledge, are often essential to interpret any opportunity. Secondary Care Medicines Data, which is now accessible in ePACT2, gives monthly standardised data from NHS trust pharmacy stock control systems, but even there the published cost is indicative rather than the full local net position. The signal exists. The hard part is what happens next. This is where too many digital products and internal programmes still fail. They stop at the chart. A medicines digital programme should not be designed around a maze of top-level analytics pages. It should be designed around a decision workflow. In practice that means four linked jobs.

dashboards. Which issues matter most now? Which ones are large enough to warrant action? Which have enough data confidence to justify time? Which are simple wins and which are complex pathway changes? Ranking is not cosmetic. It is how overwhelmed teams protect scarce managerial capacity.

pharmacy, procurement, information governance and, when needed, clinicians. This is the missing layer in many digital products. A graph may show a trend, but an organisation still needs a concise statement of the opportunity, the baseline, the source data, the assumptions, the expected value range, the key caveats and the reason the issue is worth acting on now.

alone. It usually requires some combination of formulary change, site-level engagement, pathway redesign, homecare coordination, procurement follow-through, prescribing support, or shortage mitigation. A serious digital workflow therefore needs owners, milestones, adoption assumptions and a way to track blockers. Without that, the “insight” remains detached from the organisation’s actual machinery.

still has not moved? Digital teams often underinvest here because the reporting feels less exciting than the discovery layer. In reality, this is the part finance and operational leaders trust most. It is also the only way to learn whether the earlier scoring and prioritisation logic was any good. None of this is an argument against analytics. It is an argument for placing analytics in the right position. A chart is an input into a decision workflow, not a product outcome in its own right. There is also an assurance reason to redesign programmes this way. NHS England’s updated DTAC guidance, published in February 2026, emphasises the continued use of a standard national assurance form and the need to assess digital health technologies against baseline requirements covering clinical safety, data protection, technical security, interoperability and usability/accessibility. The same guidance warns buyers not to create amended local versions that duplicate the national baseline. In other words, if a digital supplier is already generating friction at the assurance stage, the product is unlikely to become easier to manage after procurement. Public-data medicine tools bring additional caveats that digital leaders should surface rather than hide. Primary care and secondary care measures are different. Not all datasets are live. Indicative cost is not the same as contracted cost. Benchmarking can identify outliers without proving that they are wrong. Some dashboards do not capture clinical indication, which means the signal cannot be treated as a mandatory switch list. The best digital programmes make these caveats visible with provenance, refresh dates and confidence notes. The worst ones wave them away with the language of “real time” and “full visibility”. That honesty is not a weakness. It is one of the strongest adoption tools available. Digital leaders also need to be more deliberate about sponsorship. Medicines value programmes should almost never be owned by digital alone. The strongest sponsor model is a triangle: pharmacy, finance and digital. Pharmacy defines clinical appropriateness and operational pathway reality. Finance defines the standard of proof for value. Digital defines assurance, integration and product fit. When those three functions work together, the programme moves from “interesting dashboard” to “managed change”. This matters now because the NHS is under intense pressure to show productivity gains without adding avoidable administrative burden. A medicines programme that produces more charts but also more meetings, more reconciliation work and more duplicated reporting is not maturity. It is overhead dressed up as innovation. So what should boards and digital leaders ask for in the first ninety days of a medicines digital programme? Not a platform tour. Not a heat map. Not a promise of artificial intelligence. Ask for a ranked portfolio of live opportunities. Ask for evidence packs that show provenance and assumptions. Ask for a deployment view with owners and dates. Ask for a review page that compares forecast with realised value. Ask for clear refresh dates and data limitations. Ask for the assurance pack early. That is what turns digital from an observational layer into an operating layer. The NHS does not need another dashboard that tells teams something may be wrong. It needs digital workflows that help them decide what to do, defend why they are doing it, and prove what changed as a result. The programmes that fail usually do not fail because the charts were poor. They fail because the chart was where the design ambition stopped. Research base used in drafting