Everyone in the building already uses AI. Month-end still waits for three people retyping what a supplier already sent you.

What there is to buy

Two of them are work. One is a system you keep.

Most of what follows on this page is the first of these. The other two are here because a reader deciding whether to spend an hour with us should know what the hour could lead to.

The documents we work on

Named one by one, because a system that reads a permit to work is not the same system that reads an invoice.

Twenty years of procedures across SharePoint and four shared drives. Your new engineer asks three colleagues, then uses an old revision because nobody can tell her which one is current.

We are not a horizontal AI consultancy that also serves industry. The lane is the industrial back office - finance, procurement, operations, HSE and shared services. The paper that moves between them is where we start, because it is where the cost is easiest to count and hardest to argue with.

The smallest thing you can buy

Four stages. Each ends in a decision, not an assumption. You can stop after any of them.

Building is no longer the hard part - modern tooling made it cheap to build almost anything. The expensive decision is what is worth building, in what order, and whether the organisation can absorb it.

An audit is genuinely capable of concluding that no AI work is warranted, and sometimes does. That conclusion is worth the fee - it is far cheaper than discovering the same thing eighteen months into a build. No obligation to build with us.

One result, in full

One engagement, not an average. The same table, with the same three columns, is how every result on this site is reported.

A petrochemicals operator in the Gulf. Reception and the switchboard were the front door to the plant’s office - location questions, contact lookups, facility enquiries - and every one of them interrupted a person.

Figures are drawn from the practice's own delivery records for the engagement named, measured against the process that preceded it. They have not been through third-party audit, and none is presented as an average across clients.

85+ documented deliveries sit behind this practice, organised by problem shape rather than by client sector. See the proof →

One thing already exists

Of 79 firms surveyed at our size, two have a real product rather than advice or build-to-order work.

The AI Workspace is not a slide. We built it for ourselves, we run it, and the numbers below are our own operating benchmark rather than a client's result.

The problem it solves is consolidation, not capability. Knowledge sat in separate repositories, each with its own search behaviour, and AI tooling kept arriving as more destinations rather than fewer. Another capable assistant added to that estate would have made it marginally worse.

The hard part, said plainly: consolidating repositories means consolidating permission models that disagree with each other, and an index that ignores those differences will surface a document to someone entitled to see the repository but not the file. That is the work. Any vendor presenting it as a connector exercise has not done it.

How the workspace works

Audit before automate

Published estimates put enterprise AI failure at 70-95%. Usually from automating a process that was already broken.

On a bottling line we ran a defect-data audit before anyone proposed a camera. It found three failure modes accounting for roughly 90% of the cost, against twelve listed in the QA manual. We solved the three.

Eight countries, one team

Plotted by coordinate. Eight dots, and no ninth we cannot stand behind.

What a client buys is the people who will do the work, and this is where those people have already done it - across government and private sector, in Arabic and in English, on both sides of the Atlantic.

Who this is not for

Published up front, so nobody spends a call finding out.

We would rather lose a conversation early than decline in week two.