Skip to content

Advisory

AI adoption that sticks

Four forces need to align before AI use becomes routine in the organization. When one is missing, the pilot never turns into practice. An engagement starts by finding out which one.

Four forces of AI implementation

↓ From the top

Lead

Leadership wants this, says so in public, and owns the budget line.

↑ From below

Enable

People have the tools, the data access and the permission to actually do it.

→ Across the teams

Train

Skills built for the work people really do, not for the tool.

← Back into the business

Implement

Someone who turns the idea into something that runs in production.

Where they meet

AI adoption that sticks

Four services, one per force

From the top

Lead

"The pilot won an internal award and then nothing happened. Nobody at board level owned the budget line, so phase two never got one."

Work alongside the executive team on where AI actually belongs in your business, which use cases are worth the capital, and what you are prepared to say no to. The output is a decision, not a deck.

Formats: Strategy sprint, one to two weeks. Full strategy, four to eight weeks. Executive advisory on a monthly retainer.

You keep: A prioritized use case portfolio, and every decision on record including the option you rejected.

From below

Enable

"Staff were told to use AI. They had no licences, no access to the data and no policy telling them what was allowed, so they used private accounts. You now have a compliance problem where you wanted a capability."

Vendor and platform selection with the reasoning written down. An honest look at whether your data is in a state AI can use. A usage policy short enough that people read it.

Formats: Half a day to three days, depending on how much of the estate is in scope.

You keep: A tooling decision you can defend, and a policy your people can follow.

Across the teams

Train

"The licences were bought. Six weeks on, the usage report shows the same eleven people. Everyone else sat through the training, understood it, and could not see what it had to do with their Tuesday."

Training built around the work your teams do rather than around the tool. Executives need to make decisions about AI. Practitioners need to get things done with it. Those are two different rooms.

Formats: Executive AI literacy, half a day. Team upskilling, one to two days. Practitioner cohorts running alongside a live project.

You keep: A curriculum you re-run yourself for the next group.

Back into the business

Implement

"The business has ideas and no route to production. IT has capacity and no way to tell which of the ideas matter. Both sides are being entirely reasonable and nothing ships."

Someone sitting between the two. Shaping the pilot so it can be judged, weighing what you build against what you buy, and staying close enough to delivery to see it stall before you do. Where the work is modernizing legacy software with AI, the case study is already published, numbers included.

Formats: Advisory retainer, two to four days a month. Embedded, three to six months at two to three days a week.

You keep: Something running in production with a cost model attached.

Finding out which one

Which force is missing is usually clear to the people inside the organization, and hard to say out loud. That is a conversation, not a form.

Get in touch

Prefer to see the wider picture first? The AI Readiness Assessment scores your organization across six dimensions of readiness. Six questions, five minutes, no login.

Take the assessment

How an engagement runs

Start narrow

One force, one part of the business, something that can be finished. Not a transformation programme with a two-year runway.

Work in the open

Your people are in the room and you see the reasoning, not only the conclusion. Decisions made without you leave you dependent on whoever made them.

Honest checkpoints

What is working, what costs more than it should, and what should stop. That includes work we proposed ourselves.

Hand over

The point is to make you self-reliant. If a second engagement happens, it should be because you chose it.

Start with a conversation

Tell us what is stuck.
If we cannot help directly, we will tell you straight away and point you to someone better equipped to help.