Designing agentic AI into your journeys.
A one-day programme on where an AI agent belongs in your journey, when it hands off to a human, and what that handoff feels like. Run against your own service.
A decade of design training. Companies that have paid us to train their people.
Built for teams already shipping agents.
Not an introduction to AI. It assumes you are already building agentic features and need to make specific design decisions about them.
Design teams already shipping agents
PMs and engineers designing agent behaviour
Service and content designers
Teams whose gap sits upstream of the design system
Six things you have to get right.
Run against your own live journey. Every block ends in something written down.
Where an agent belongs
Map the journey, then mark where an agent earns its place.
→ a scored shortlist of candidate stepsThe handoff to a human
What triggers it, what context carries across, what the human receives.
→ handoff specs for one real flowTrust, error and recovery
What earns trust, and how a user takes back control.
→ trust and recovery patterns, agreedNon-deterministic behaviour
The states you cannot specify in advance, written down anyway.
→ states and rules ready for sign-offTesting an agentic flow
Measure a range, not pass or fail.
→ a testing approach you can run yourselvesMaking the case internally
Turn the decisions into an argument the business accepts.
→ the tradeoffs made explicitBefore you enquire.
What does agentic AI in a user journey mean?
Agentic AI in a user journey is an AI agent that takes action on a user's behalf inside a product, rather than only answering questions. Designing for it means deciding which steps the agent should handle, when it should hand control back to a person, and how that handoff is communicated to the user.
When should an AI agent hand off to a human?
There's no universal rule, and part of the day is building one for your own service: usually the trigger is some mix of the agent's own confidence, the stakes of getting the step wrong, and whether the user has signalled they want a person. We work through your journey to decide where that line sits for you, rather than importing someone else's answer.
How do you test a flow when the AI gives a different answer each time?
You stop treating a single session as representative and design the test around a range of plausible agent responses instead of one script. We cover how to pick tasks that expose the range, how many sessions you actually need, and what to measure when "did they complete the task" isn't a stable enough signal on its own.
What does the team actually leave with?
Written artefacts, not notes. A mapped journey with the agent's role marked on it, a documented handoff design for at least one real flow from your product, a set of trust and recovery patterns the team agreed on together, and a testing approach you can run without external recruitment.
Do our designers need to know how to build AI to take this?
No. This is a design and product training, not an engineering one. You don't need to know how the underlying model works to make good decisions about where it sits in a journey and how it behaves toward a user.
Have you delivered this programme before?
Yes. We ran it in October 2026 for a commercial insurer's team, on designing agentic AI into their own journeys. Behind it is over a decade of in-house UX training for product, design and technology teams, including the BBC, Google, GroupM, Wataniya Insurance, Deloitte and Lidl. It applies that experience to a problem the whole industry is working out at the same time.
Can this be combined with UX research or service design training?
Enquire about this programme.
Tell us about your team, the journey you want to work on, and your preferred timing. We'll get back to you within 2 business days.
*Advertised price assumes 10 or more participants. Prices vary depending on the number of participants.