Digital Transformation
The transformation already started
In most companies it began without a decision: tools adopted one at a time, renewals nobody reviewed, and AI already in daily use under no policy. Digital transformation is the work of deciding what a company should automate, replace, or leave alone – in what order, at what cost, and with what changing around it. The software market is crowded with good products. Most of what goes wrong happens before any of them are installed, and after.
Spending has risen for a decade. The share of projects that fail has barely moved.
The reasons are consistent, and almost none of them are technical: unclear goals, no owner, data that will not support what is being built on it, processes automated exactly as they were, and adoption that never happens because nobody asked the people doing the work.
There is a second pattern now, particular to this moment. In most companies the transformation has already started without a decision being made. Staff have adopted AI tools on their own, using whatever data the task required, under no policy and no review. The question for many owners has shifted from whether to begin to how to govern what is already running.
This practice sits before procurement and before implementation. It establishes what the business actually needs, what it already has, what its data and processes can support, and what sequence makes sense. It ends with a recommendation that can include spending nothing.
Most companies can answer two or three of these with confidence.
Which of the tools we pay for are actually in use, and by whom?
Who here is already using AI, with what data, and under whose approval?
Is our data in a condition that would support what we propose to build on it?
What would we redesign if we were starting now, and what are we about to automate exactly as it is?
Which problem is this purchase solving, and who decided that?
What happens to this project if the person championing it leaves?
If it works, what does the business stop doing?
How the work takes shape.
Technology and spend review
What the company pays for, what is actually used, what overlaps, and what renews automatically without anyone deciding to renew it.
AI readiness and current use
What AI is already being used for inside the business, by whom, and with which data. What the company's data would and would not support. Where automation would pay for itself, and where it would mostly produce impressive demonstrations.
Process review before automation
Which workflows are worth keeping, which are worth redesigning, and which exist only because an earlier system required them. Automating a process preserves it, including its defects, at higher speed and greater cost to reverse.
Governance and acceptable use
Who may use which tools, with what categories of data, and what has to be reviewed by a person before it leaves the building. Most companies of this size have no policy at all, which is a decision that has been made by default rather than on purpose.
Sequencing and business case
What to do first, what it will cost including the time of the people involved, what would count as evidence that it worked, and what to decline. Written so that a change in conditions triggers a review rather than a restart.
We do not sell software, resell it, implement it, or take a commission on anything recommended here.
That removes the conflict that sits inside most technology advice. A firm that earns its margin on implementation has a reason to find that implementation is needed, and the finding tends to arrive. Our output is a decision, and the decision is sometimes that the current systems are adequate and the money is better spent elsewhere.
The second commitment is sequencing. Most transformation programs fail at the point where a pilot is supposed to become ordinary operation, so the work is planned backwards from that transition rather than toward a launch.
The third is honesty about what a technology will not fix. A tool can make an existing process faster. It cannot decide which processes should exist.
We do not implement any of this.
No software is sold here and no commission is earned on anything recommended. Implementation goes to whoever is best suited to it – sometimes a vendor already in the building, sometimes the company's own people, and sometimes nobody, because the answer was to wait.
Book a consultation