Teams stopped treating AI as a novelty and started using it as a work tool.

We delivered department-by-department AI training across three parallel engagements in 2026, including a Foundation Programme for a Lebanese industrial group with more than 1,000 employees, a C-suite strategy workshop for the same client, and a focused sales-team series for a regional B2B organisation. Every session was delivered in person at the client's site, mapped to the actual workflows of the team in the room. The industrial group programme is now in its second year. Several other training clients have moved into build engagements with us. The metric we cared about wasn't satisfaction at the end of the day. It was whether the team still used what they learned three months later.

What was actually broken

The teams we worked with weren't sceptical about AI. They were stuck.

They'd heard about it. They'd tried a few tools. They'd ended up back where they started, using AI chatbots as a slightly faster search bar, wondering why it wasn't making a real difference to anything that mattered. The gap wasn't awareness. It was application. Nobody had shown them how to connect AI to the actual work they do every day: the reporting, the outreach, the content drafts, the internal communication.

Generic courses hadn't done it. The standard MOOC-style "Introduction to AI for Business" gets people excited and leaves them no closer to using AI for anything specific. Slide decks hadn't done it. The "future of work" keynote produces enthusiasm without translation. Vendor-led training hadn't done it either, because vendors teach their product, not the team's job.

The pattern we kept seeing was that organisations had bought AI tools, watched usage stall, and were one or two failed initiatives away from concluding that AI didn't apply to them. Which is the wrong conclusion. The problem wasn't AI. The problem was that nobody had bridged the gap between the tools and the work.

What we designed

Department-by-department sessions, each mapped to the workflows that team actually runs.

For the Lebanese industrial group, we built a Foundation Programme that ran in waves across the organisation. Each department got its own workshop, with case studies, exercises, and prompts drawn from the kind of work that team does. Finance worked on variance analysis and reporting. HR worked on policy drafts and candidate communications. Operations worked on process documentation and SOP review. The customer-facing teams worked on outreach, follow-ups, and account research.

In parallel, we ran a strategic workshop for the C-suite and senior leadership of the same client. That session wasn't about how to use AI tools. It was about how to make the right decisions about where AI belongs in the organisation, and where it doesn't. Where the leverage is. Where the risk is. What "AI strategy" means for a company that makes physical products at scale.

For a regional B2B sales organisation, we ran a focused training series for the sales team. Built around a specific AI tool they'd adopted, with practical exercises drawn from their actual pipeline. Between sessions, we collected structured feedback and used it to shape the next one. The feedback loop was the thing that made the series adaptive rather than scripted.

In the room, not on a screen

Every session was delivered in person at the client's site. That was a deliberate choice, and it cost more than running remote sessions. It was worth it.

Remote AI training has a specific failure mode: people log in, half-attend, do a few exercises with the tool, and forget most of it by the next morning. The presence problem matters because adoption is social. When the team learns together, in the same room, with their actual colleagues running their actual workflows, the new behaviour gets reinforced by everyone seeing each other do it. The training becomes part of the team's culture in a way that remote sessions almost never achieve.

We delivered across Lebanon, the UAE, and Saudi Arabia. Different cities. Different cultures. The same in-person format. The energy is different in different rooms, but the principle holds: people learn by doing, in front of each other, with someone in the room to answer the questions that come up two seconds after the slide moves on.

The feedback loop between sessions

The thing that separates a training programme from a series of workshops is whether what happens in session three is shaped by what happened in sessions one and two.

After every session, we collected structured feedback from participants and from their managers. Not a smile sheet. Specific questions about what worked, what didn't, what they actually used between sessions, what blocked them. We rebuilt the next session around the answers.

For the sales team, that meant the second session focused on a specific objection-handling workflow that the first session had revealed wasn't being applied. The third session moved on to outbound research because the team had mastered the first two and was ready for the next thing.

Rigid curricula don't survive contact with real teams. The plan for week four is always less important than what week three taught you.

The C-suite track, separate from the team training

The thing nobody plans for is the leadership problem.

You train a team to use AI. The team gets faster at their work. Their leadership, who hasn't been trained, doesn't know what to expect from an AI-augmented team. They keep prioritising the same projects, asking for the same outputs at the same pace, because that's what they know how to lead to. The team's new capacity gets absorbed into the same backlog, and the training looks less effective than it was.

We ran a separate session for the C-suite of the industrial group, deliberately not technical. The agenda was three questions: where in the business does AI compound, where is it irrelevant, and where is it actively risky. The answers don't fit on a slide. They came out of two hours of structured discussion grounded in the leadership's actual decisions for the next year.

That session was the one several of the senior people told us, afterwards, was the part they wished they'd had three years ago.

What we measured

The honest measurement of training isn't how many people completed it. It's whether the work has changed.

For the industrial group programme, the programme moved from proposal to active departmental rollout, and into a second year. That's the metric that matters: not whether people enjoyed the session, but whether the organisation trusted the methodology enough to fund what came next.

The sales team started the engagement treating AI as an optional extra. They finished it with specific workflows they use daily, and the team's adoption of the underlying AI tool went from sporadic to embedded inside three months.

Several organisations that went through training are now scoping AI build projects with us, because they saw what was possible during training and wanted more than a workshop. That's the second-order signal: training that ends in a build engagement means the team got specific enough about what they wanted to act on it.

What we'd do differently

Two things.

We'd run the C-suite session before the team training, not in parallel. Doing them at the same time meant the leadership and the teams were on slightly different timelines, and we caught some misalignment that we could have prevented. Future engagements: leadership first, by a week or two.

We'd budget for follow-up sessions explicitly. The team finishes the programme with momentum and a list of things they want to try. Without a structured 30-day check-in, some of that momentum gets lost to operational work. We've added a built-in follow-up session to the Foundation Programme as a result.

If your team is stuck somewhere between "we have the tools" and "the tools are changing the work," the answer is not another generic course. It's training built around your team's actual workflows, delivered in person, with a feedback loop that adapts as the team learns.