A digital transformation that disappoints rarely ends in a declared failure. Projects continue to deliver, committees to meet, the budget to be consumed. It is the announced effects that do not arrive: processes remain as they were, the new tools duplicate the old ones, and nobody can say what the whole has brought in.
A successful digital transformation then depends, essentially, on five management mechanisms: governance that decides, a portfolio held to real capacity, arbitrations made on time, adoption prepared with the business and benefits measured by those who promised them. The choice of technologies comes afterwards. The diagnosis proposed here starts from observable signals, traces back to the failing mechanism, then ranks the remedies from the least costly to the most demanding.
The signals of a transformation that is no longer moving forward
None of these signals is proof on its own. Their accumulation guides the diagnosis.
In the governance bodies, the transformation committee breaks up without a record of decisions. The roadmap is presented at each meeting with the same milestones, pushed back. Progress is expressed as budget consumed or number of projects launched.
In the teams, every initiative is said to be a priority. Launches follow one another, closures are rare. Pilots conclude favourably and are never rolled out.
On the ground, the tools delivered coexist with those they were meant to replace. Business departments talk about “the IT department’s project” or “the digital project”. And when you ask what a project delivered last year has changed, nobody has a measurement to cite or an owner to name.
Tracing back from the signal to the cause
Each signal points to one of the five mechanisms. For each, a simple check confirms or rules out the hypothesis.
Governance without an arbiter
A transformation cuts across several departments. If nobody below executive management has the power to arbitrate between them, each disagreement waits or is settled by a compromise that satisfies no objective. The sponsor sends a representative, and the committee brings together people who keep each other informed. To check, take the last disagreement between two departments: who settled it, and how many meetings did it take?
A portfolio larger than capacity
The transformation was added to existing projects and day-to-day business, and nothing was postponed or stopped to make room for it. The check consists of counting the initiatives open under the transformation label, then the people who hold the scarce skills on which they all depend. The criteria, the reading of capacity and the body that make it possible to decide fall under project portfolio management.
Arbitrations that wait
Delay forms between projects, where no schedule shows it: a platform awaited by several applications, a master data repository, an architecture choice. These subjects have no owner, and requests for decisions circulate without a deadline. The check is a list: the pending decisions, with for each the day from which waiting costs a delay.
Adoption that stops at go-live
Adoption was entrusted to the projects, whose mandate ends at go-live. After that day, nobody is responsible for checking that the new practice has replaced the old one, and front-line management has not been prepared to require it. For a project already delivered, the check comes down to two findings: the number of people who actually do their day-to-day work in the new tool, and the state of the old system, closed or still open. Our page on change management describes how this work begins from the framing stage.
Benefits without an owner
The benefits appear in the file that justified the budget, written by the project team. No business department has taken them up in its own objectives. The baseline has not been recorded, so that a real gain could not even be demonstrated. The check concerns each announced benefit: a name on the business side, an initial measurement, a date for taking the reading.
In what order to act for a successful digital transformation
The first remedies produce the information the following ones need: a portfolio of which nobody has the complete list cannot be arbitrated.
- The inventory comes first, because it requires no decision. Write on a single page all the initiatives, with for each its sponsor, the target result, the benefit owner, the scarce skills used and the next decision expected. The empty boxes indicate which of the five mechanisms is most lacking.
- The existing body is then put in a position to decide, without creating a committee. Its agenda consists of the decisions requested, each presented with its options and their consequences, and the sponsor attends in person. Our article on the project review details this format.
- Owners are appointed. The overall trajectory receives one, with a written mandate. Each benefit receives a business owner, an indicator, a baseline and a deadline.
- The portfolio can then be arbitrated on facts. Demand is compared with capacity, and initiatives are postponed or stopped until those that remain can be completed. The effort is of another order, because it involves giving things up, which the executive committee must stand by in front of its own teams.
- The trajectory is broken down again into stages. A stage is only considered reached when a real population uses what it has delivered, has been supported and the old system has an announced shutdown date.
- The last remedy concerns the ambition itself. When neither the capacity nor the support of senior management exists to reach the target, the target is reduced: fewer areas covered, a more distant horizon. This decision falls to executive management. Taken early, it leaves the choice of what is abandoned.
The case where the obstacle is technical
The diagnosis does not always conclude that management is at fault. Unusable data, core applications impossible to open up or the absence of a software delivery pipeline are real obstacles. One sign distinguishes them: properly governed and resourced projects run into the same point, one after another.
The remedy is then a foundation workstream, for example application modernisation or a data platform. It enters the portfolio like the other initiatives, with a sponsor, expected benefits and a place in the order of priorities. Funded separately and without arbitration, it would reproduce the causes described above.
Judging success stage by stage
The six remedies have one thing in common: they assume that someone holds the whole together. This is the purpose of programme management, which our service page describes from framing to the handover of results.
A transformation run in this way does not wait until it ends to be judged. At the end of each stage, the same question is put to the business departments: what has changed in the way your teams work, and what has this change brought in?