How the research is built.

The Big AI Secret is a research programme that reveals how businesses actually use AI. It is editorial and independent. It reports what the data shows and cites its sources. It is not a set of dated editions. It is a continuously updated body of research, refreshed in near real time as new interviews come in.

The research draws on over 1,500 in-depth interviews with business leaders across sectors, including agency, commercial and developers, professional services, retail, financial services, manufacturing, healthcare and technology. It is validated against external datasets, and it is updated in near real time as new interviews come in. It is qualitative depth at scale: real operators describing what they run, what works, and what does not.

Two models organise the findings. A six-dimension evaluation model scores how ready an organisation is to get value from AI, across the Tasks in scope, the Agents in use, the workflows they sit inside, the handoffs between them, the governance around them, and the measurement of the hours they move. A four-stage maturity model then places organisations on the path from experimenting to embedding: Explorers, Implementers, Integrators, and Transformers. Most organisations, the research finds, are still in the first two stages.

Every finding in the programme originates in a recorded interview or in a structured questionnaire completed by a named leader within a firm. Interviews are anonymised by role and sector before publication. Interviewees see the quotes attributed to them before anything is published. Nothing is printed that has not been checked.

Much of the picture today rests on self-reported estimates. Leaders tell us what they run and roughly what it costs them. We normalise those answers and benchmark them against organisation size and maturity, but it remains estimation, and we treat it as such. Every finding that depends on estimation is labelled that way, next to the figure.

The research is moving from estimated to measured. As businesses choose to share real records of the work their AI actually completes, it shifts from asking what AI did to recording it. Taking part is how that picture sharpens.