AI Design Sprint

Rapidly scope an AI project and validate it with users

An AI Design Sprint is a 4-day workshop with the aim to define, prototype and get customer feedback for an AI-centered product or major feature. The format is based on the classical Google Design Sprint with a focus on AI issues.

Deliverables & impact

• A prototype, tested with real users
• Optimisation targets for your AI engineers
• A team aligned on a concrete vision
• Clear insights as to the next steps

When to run an AI Design Sprint?

The right moment is when you are ready to get specific with your AI project. Usually this means you have:

• Identified a meaningful AI use case
• Validated the business case and rough technical feasibility

Who should take part?

Between 3 and 7 of your key team members. Typically stakeholders, product owners, domain experts, UX designers and AI engineers.

Format

Day #1: Map
The participating team is brought up to speed ,knowledge is shared and outcomes/goals are defined.

Day #2: Design
Solutions are ideated and sketched out. Particular attention is given to AI specific issues, such as optimisation targets, fallbacks, trust and explicability and conversational input.

Day #3: Prototype
The retained solution is prototyped, typically by using humans to simulate the interactions using the Wizard of Oz or Concierge approach.

Day #4: Test
The prototype is put into the hands of real users to get feedback on usability and desirability.

And then ...

Based on user feedback a decision is made to either continue iterating the product, conduct a technical feasibility study or gear up for implementation.

Want to find out more?

Read our blog post on how to run an AI Design Sprint.

Ready to run an AI Design Sprint ...

... or generally seeking help designing your AI products?

Get in touch

We can run your AI Design Sprint in English, German or French