Data first. Then AI.

Most company AI is installed on data that cannot support it. We do the opposite: we measure your data, fix what matters, then build the engine on it, and we prove it on your own documents before you commit.

Why it fails

Gartner expects 60% of AI projects without AI-ready data to be abandoned. MIT found only about 5% of company AI pilots deliver measurable results. The AI was rarely the problem. The data was: duplicated, outdated, unowned, defined three different ways in three departments. An engine built on that data answers confidently and wrongly, and the company stops trusting it within a month.

The six pillars

The Readiness Audit measures your data across six pillars. For each: what we check, and what ready looks like.

Pillar
What we check
What ready looks like

Governance

Who can see what; user groups; where sensitive data sits
Every document has an access rule the engine can follow, taken from the permissions you already run

Lineage

Where each figure and policy comes from; versions; sources of truth
Every answer can cite a document and a page; one current version per document

Ownership

Who owns each folder, policy, price list and template
Every important document has a named owner who keeps it current

Quality

Duplicates, stale copies, scans without text, broken files
Duplicates below an agreed threshold; every document readable by the engine, in Arabic and English

Metrics

How KPIs, terms and units are defined across departments
One definition per KPI and term, agreed by management and written down

Trust

What the engine may answer from, what it must refuse, how it is tested
An agreed source list and a 50-question acceptance test written by your team

The Readiness Score

One score, 0 to 100, per department. Six pillars, equal weight, scored on evidence we collect, not on what people tell us. The scorecard shows the score, the three gaps that cost the most, the pilot department we recommend and why, and an ROI estimate built from your own numbers. Two weeks from kick-off to scorecard.

Audit and foundation

The audit and the Knowledge Foundation are different work with different outputs.

Readiness Audit
Knowledge Foundation
What it works on
The data you already have: documents, files, systems
The knowledge you have never written down: rules, definitions, playbooks
The question it answers
Is our data ready for AI, and what must we fix?
How does our company actually work, in one agreed source of truth?
What you receive
A diagnosis: scorecard, gap list, pilot plan, ROI estimate
A new asset: codified strategy, KPI definitions, pricing and approval rules, workflows, policies, playbooks
When
First, always. Two weeks
Optional, after the audit. Four to six weeks

The acceptance test

Before a department goes live, your team writes 50 real questions and tasks with known correct answers. The engine passes when it answers them with the right sources. You sign off; then it is live. It is the simplest guarantee we can give, and the only honest one.

See it on ten of your own documents.

A 45-minute session at your office. Bring contracts, policies or price lists; leave with answers, finished work and their sources.

We reply within one business day.