Do not buy “transformation”. Buy one closed loop.
Digital programmes stall because they start with a portfolio of initiatives. Start with one painful loop and make it boringly reliable.
In this note
A closed loop is request → decision → execution → evidence → history. Until that exists for one class of work, a “platform strategy” is decoration.
Pick the painful class
Not “digital maintenance”. Corrective work orders on rotating equipment at one site. Not “field transformation”. The chiller calls that breach SLA on Friday afternoon. A programme that starts with twelve workstreams will still be arguing about the data model when the shift needs a parts van.
- One class of work, one site, one owner of the loop.
- Evidence at close-out — not a status update in a steering deck.
- Only then copy the loop to the next class. Do not clone a half-built platform.
Boring is the goal
A reliable loop is unremarkable: the request arrives, the right person decides, the job is done with the current procedure, the record is complete. That boredom is the opposite of a transformation launch. It is also the only thing the plant will still be using next year.
