AI Impact Lab: Five-Week Workflow Redesign Programme for Enterprise Teams

What it is: a five-week, instructor-led programme (one 90-minute live session per week plus Demo Day) in which cohorts of 25 to 54 participants redesign a real workflow from their own role with the AI tools the organisation has enabled. Each team leaves with an AI Workflow Upgrade: a working prototype, a quantified business case, and a pilot plan with a named owner. Use cases come top-down from managers and bottom-up from participants, then pass gates for impact, feasibility, and scope before any building starts.

Who is the AI Impact Lab for?

People leaders and knowledge workers who already have a tool and need to turn it into changed work, not a new habit of asking a chatbot questions.

The programme sits one level above foundational fluency. Participants have typically completed a 101-style session or have equivalent exposure. The best cohorts are a cross-section of the organisation (the middle of the adoption curve, not only enthusiasts or holdouts), because that is the population whose results generalise when the programme scales.

Who

People leaders and knowledge workers, any function

Format

5 weekly 90-minute live sessions + Demo Day; roughly 70% hands-on

Cohort size

25–54 participants in small teams

Timeline

5 weeks, plus a 4–6 week post-programme pilot

Tools

Tool-agnostic: Microsoft 365 Copilot and Copilot Studio, Gemini, Claude, ChatGPT, client-specific tools

Output

One AI Workflow Upgrade per team: prototype, business case, pilot plan, named owner

Calibration

Before the cohort, we gather use cases from managers and participants and screen them against three gates: business impact, technical feasibility, and scope that fits five weeks.

Workflow identification

Teams map the current process, locate the step where AI changes the economics, and decide how far AI reaches and how much autonomy it gets.

Build

Teams build the upgrade in the organisation's tools (a Gem, a Copilot agent, a Claude Project or Skill, a custom GPT) with live instructors and AI coaches in breakouts.

Business case

Teams quantify time, reach, or quality gains and commit to the evidence that would make them stop.

Demo Day

Each team presents a working upgrade to sponsors and peers. Sponsors decide what proceeds to pilot.

Pilot support

Four to six weeks of coach office hours and asynchronous support while upgrades meet real users, with escalation to a 301 cohort or an engineering team where needed.

AI Workflow Upgrades produced

+73

post-programme NPS, no detractors

100%

of leaders planned to activate their teams

+64%

confidence in identifying an AI-upgradable workflow (pre/post, 5-point scale)

Confidence measured by pre/post self-assessment. Response rates and dates are available on request.

Three of the twelve upgrades, with projected savings pending rollout: an outlier-detection tool for a fleet of roughly 1,200 ATMs replacing a manual dashboard process (about 600 hours a year); a regulation navigator translating a complex compliance rule into actionable guidance (about 500 hours a year in ad hoc research); and a people-leader onboarding tool generating role-specific 30-day plans (about two hours per hire).

Across prior workflow-upgrade programmes, 76 percent of participants showed verified sustained usage at 90 days. A 30-day follow-up at the bank showed stronger durability for this five-week format than for the single-session foundations programme that preceded it.

How is the AI Impact Lab different from other AI training?

The output is a changed workflow with an owner, not a certificate.

Learn by building. Roughly 70 percent of session time is hands-on, on the participant's own work, in the organisation's own tools.
Gated use cases. Nothing is built that has not passed impact, feasibility, and scope screens, so Demo Day produces decisions rather than demos.
Expert support in real time. Live instructors plus embedded AI coaches in every breakout, with weekly office hours during the pilot.
Tool-agnostic foundation. The method (workflow identification, context engineering, iteration, verification) transfers across tools; the tool-specific layer is taught as thin modules on top.
Measured after the end. Pre/post skill deltas, 30-day durability, and 90-day usage from the tool's own data.

A five-week, instructor-led cohort programme in which teams of enterprise employees redesign one real workflow using the AI tools their organisation has enabled, and present a working upgrade with a business case and pilot plan on Demo Day.

Five weekly 90-minute live sessions plus a Demo Day, followed by a four-to-six-week pilot with coach support. Calibration before the cohort takes two to four weeks.

Whichever the client has enabled: Microsoft 365 Copilot and Copilot Studio, Google Gemini, Claude, ChatGPT, or client-specific tools. The method is tool-agnostic; tool-specific instruction is layered on top.

Each team produces an AI Workflow Upgrade: a working prototype, a quantified business case (time, reach, or quality), and a pilot plan with a named business owner.

Pre/post confidence on the skills the programme targets, the number and quality of workflow upgrades that pass to pilot, projected savings validated by managers, and verified sustained usage at 30 and 90 days.

A representative cross-section of people leaders and knowledge workers, typically 25 to 54 per cohort. Cohorts made only of enthusiasts produce results that do not generalise.

Featured:

Turn AI anxiety into action

Featured:

Bolstering technical skills development in MENA

Featured:

Explore new digital career pathways

Featured:

Take action on generative AI

Featured:

Hear what the experts have to say

AI Impact Lab: from individual productivity to workflow redesign

Who is the AI Impact Lab for?

How does the programme work?

Calibration

Workflow identification

Build

Business case

Demo Day

Pilot support

What outcomes has the programme produced?

How is the AI Impact Lab different from other AI training?

How does it fit with the rest of an enablement programme?

Frequently asked questions

AI Champions Network

How we measure AI training ROI

The AI Champion model: why the scout is the role that pays

Enterprise AI Enablement overview

Run an AI Impact Lab with your teams