· 5 min read
Requests such as "train our engineers on AI" or "we need cloud skills" are starting points, not briefs. A short needs analysis turns them into something a training provider can scope and an organisation can measure.
Start with the business outcome
Write down what should be different in six to twelve months. Examples include faster delivery of a platform migration, a team able to own an AI feature in production, or managers who can review AI-assisted work. If the outcome cannot be described in one sentence, the training scope is not ready.
Separate roles before you separate topics
Group participants by what they do, not by title. Leaders need decision-making context, practitioners need applied skills, and builders need depth. A single cohort mixing all three usually serves none of them well.
Establish a baseline
- Review recent project work and incident reviews for recurring capability gaps.
- Use short skills assessments or structured manager conversations rather than self-ratings alone.
- Record the tools, languages and platforms people actually use, so labs match reality.
Identify constraints early
Confirm data restrictions, tool access, working hours, language needs and delivery locations before designing content. Constraints change format decisions far more than topic lists do.
Agree how you will measure success
Choose indicators before training begins: assessment movement, application in real work, manager observations or delivery metrics your teams already track. Completion alone does not show capability.