Let's talk
Back to AI Transformation

IT & data · 06 / 13

AI roadmap for IT and data

Know before you spend which of the data in your ten systems can be trusted, set an AI policy that passes PDPA, and pick tools with no tie to any vendor.

Talk to an engineer about this
An IT lead dots sticky notes for CRM, POS, ERP and more on a glass wall: 3 of 10 sources trusted, PDPA policy approved. IT & data
“Every department wants AI, but the data sits in ten systems, nobody knows which one to trust, and PDPA is a worry.”

The problem

IT gets AI requests from every department at once. Sales wants a chatbot, accounting wants invoices read, executives want to ask the data questions in plain language. The data every request needs is spread across the ERP, the accounting system, the CRM and the Excel files each department keeps for itself. The same customer is spelled three ways and nobody can say which record is right.

Meanwhile staff have already started pasting customer data into public AI tools, because no policy says whether they may. IT sits between the pressure to start and the PDPA risk, with no time for either.

How we solve it

We start with an audit that goes through the systems one by one with the IT team: what each holds, whether it can be reached through an API or the database, who owns the data and what shape it is in. The result is a one-page data map of the organisation that says, for each AI request, whether it can be done now, needs data prepared first or should wait.

The AI-use policy is written with the legal team as rules staff can actually follow: which kinds of data may go into public tools and which must go through company systems. The first system we build together is a cleaned, shared data layer that other departments connect AI to from one place, with IT seeing every access.

Opening up a new data set and choosing tools stay decisions for IT and the executives. We lay out the options with each vendor’s price and limits, cloud or on-premise, and you choose.

How it runs

Work comes in from
  • ERP, CRM and accounting system
  • Files and Excel from every department
  • AI requests from every department
  • Interviews with IT and legal
What the AI does
  1. Walk every system and grade the data
  2. Write an AI-use policy that passes PDPA
  3. Rank the requests and deliver the 90-day roadmap
  4. Build the shared data layer together
Where it lands
  • A data map of the organisation
  • A policy staff can follow
  • A shared data layer with an audit log
  • An IT team that assesses the next request itself

Before and after

Before
After
Every department asks for AI at once and IT has no way to choose
Every request has a rank and a list of what to prepare first
The same data sits in ten systems and nobody knows which is right
A data map, and one record that counts as official
Staff use public AI on customer data with no rules
A clear policy on which data goes where, and every access logged

What you get

  1. 01

    An AI readiness audit that walks every system in the organisation: what it holds, how it can be connected, which data sets can be trusted and which duplicate or contradict each other

  2. 02

    An AI-use policy for the organisation that says which data may leave the building, which must stay, who approves and how it passes PDPA

  3. 03

    A 90-day roadmap that ranks the AI requests from every department by payback and data readiness, and says what data has to be prepared first

  4. 04

    The first system, built together: a shared data layer other departments can connect AI to without asking for access to every system, with an audit log of who pulled what and when

  5. 05

    A tool and model recommendation chosen on fit and price, cloud or on-premise, plus a workshop so the IT team can assess the next request itself

Who gets what

Business owner

Know before you pay whether each AI project can really be done with the data you have, and stop worrying about where staff are sending customer data.

IT director

We connect every system read-only first and leave the existing ones alone. The shared data layer runs in your cloud account or on-premise, signs in through your existing Active Directory, and logs every pull.

The team using it every day

A department that wants AI asks one place for data instead of filing access requests with four systems, and IT has a ready answer for which requests can be done.

Who this fits

Organisations running more than five systemsGroups with a central IT teamHospitals and financial institutionsManufacturers and multi-branch retailersCompanies planning their first AI project

Connects with what you already run

SAPOracleMicrosoft 365Google WorkspaceActive DirectorySnowflake and BigQueryPower BI

Development process

  1. 1

    Discover

    Requirements, users and success metrics, with scope and price fixed before we start.

  2. 2

    Design

    UX and system architecture; the prototype is approved before anything is built.

  3. 3

    Build

    AI-accelerated sprints with a demo every week, reviewed by senior engineers.

  4. 4

    Test

    QA, security and performance verified against the agreed scope.

  5. 5

    Launch & care

    Production deploy, team training, and a monthly care plan.

Turn your business problem into a system that works for you

Tell us today — get an executive-ready proposal with the plan and budget.

Talk to an engineer about this

More solutions in AI Transformation

See the whole service
Executives & finance AI roadmap for executives and finance Sales teams AI roadmap for sales Operations & production AI roadmap for production and operations HR teams AI roadmap for HR Marketing teams AI roadmap for marketing Organisations rolling out AI Forward Deployed Engineers (FDE) for AI adoption Factories & manufacturers AI roadmap for a factory Hospitals & clinic groups AI roadmap for healthcare providers Banking, insurance & securities AI roadmap for financial institutions Multi-branch retail AI roadmap for retail Logistics companies AI roadmap for transport and warehousing Family businesses & SMEs AI roadmap on a limited budget