The phrase "digital transformation" has done enormous damage to a lot of medium-sized businesses. It implies that there's a transformation to be made — a before and after, a project with a start and end date, a moment when you're done. That framing is almost always wrong, and it leads companies to invest in the wrong things in the wrong order.
Most digital transformations fail not because of the technology, but because of the sequence. Companies try to automate before they've standardised. They try to scale before they've measured.
The sequencing problem
We see the same pattern repeatedly. A company buys an ERP because they've outgrown spreadsheets. The implementation takes 18 months and twice the budget. Three years later, the ERP is being used as a glorified spreadsheet, with most of the advanced features turned off because the underlying processes were never standardised before the software was installed.
Or a company invests in AI tooling before they have reliable data. The AI has nothing to learn from, nothing to optimise against, and the project quietly dies after the pilot.
The right sequence
- Standardise the process before you automate it
- Measure what matters before you try to optimise it
- Fix the data before you build AI on top of it
- Get adoption before you scale the rollout
- Build for the next 18 months, not the next 10 years
The companies that execute digital transformation well treat it as a series of small, sequential, measurable improvements — not a single big-bang project. Each step makes the next step possible. Each system is introduced after the process it supports has been defined and validated.
Technology doesn't fix broken processes. It amplifies them.
The good news is that when you get the sequence right, the individual steps are much smaller and less risky than a traditional transformation programme. The investment is distributed over time, the ROI is visible at each stage, and the risk of catastrophic failure drops dramatically.