A digital transformation showing several of these seven signs shrinking governance meetings, status reports that stay green with less and less detail, RAID items that keep rolling forward unresolved, disagreement about what the original business case promised, deprioritized change management, data problems surfacing later and later in the timeline, and unstable ownership of benefits tracking is very likely heading toward the roughly 65–70% failure rate that research from McKinsey and other firms has consistently found across digital transformation initiatives. The good news: these signs are observable well before failure becomes undeniable, which means there's still time to intervene if you catch them early.
What Are the Seven Warning Signs?
1. Are Steering Committee Meetings Getting Shorter or Cancelled?
When governance meetings shrink from an hour to fifteen minutes, or get "moved to next week" repeatedly, it usually signals nobody wants to surface the hard decisions that meeting exists to make. A steering committee that's stopped meeting its full scheduled time is rarely a sign of things running smoothly it's far more often a sign that difficult topics are being avoided.
2. Do Status Reports Stay Green With Less Detail Each Cycle?
A status report that's been "on track" for months, with vaguer language each cycle, is often masking problems the team doesn't yet have language to explain, or doesn't want to be the one to raise. Watch for reports that shrink in specificity even as the program supposedly stays on schedule.
3. Do the Same RAID Items Keep Rolling Forward Unresolved?
A RAID log where the same risks appear "still open" review after review is one of the most reliable predictors of program trouble. It means the log has become documentation rather than a working governance tool actively driving decisions.
4. Can Anyone Restate the Original Business Case From Memory?
If the sponsor, the program leader, and the team all give different answers about what the transformation was supposed to achieve, the program has likely drifted from its original scope without anyone deciding that on purpose. This is a quiet but serious warning sign.
5. Is Change Management Getting Quietly Deprioritized Under Schedule Pressure?
When training and communication activities are the first things cut whenever the timeline slips, it's a strong predictor of an adoption problem waiting at go-live. Change management being treated as optional under pressure rarely reduces true program cost it just defers it to a harder, more visible failure later.
6. Are Data Quality Issues Being Discovered Later and Later in the Timeline?
Each time a new data problem surfaces closer to cutover than the last one did, it usually means data validation isn't happening early enough to catch issues while they're still cheap to fix. A rising trend of late-discovered data issues is a leading indicator worth tracking explicitly.
7. Has Ownership of Benefits Tracking Changed More Than Once?
If accountability for proving ROI has moved between people or teams more than once, benefits realization has likely become an afterthought rather than an active discipline. Unstable benefit ownership is a strong predictor of ending up among the roughly 73% of organizations that can't prove ROI after go-live.
Summary Table: What Each Sign Indicates
| Warning Sign | What It Usually Indicates | AMIGA Dimension at Risk |
|---|---|---|
| Shorter or cancelled steering meetings | Difficult decisions are being avoided | Governance |
| Vaguer "green" status reports | Problems exist but aren't yet being named | Governance |
| Stale, recurring RAID items | The RAID log isn't driving real decisions | Governance |
| No shared memory of the business case | Scope has drifted without a decision to allow it | Value |
| Change management deprioritized | Adoption risk is building for go-live | People |
| Later and later data discoveries | Validation is happening too late to be cheap | Data |
| Unstable benefits tracking ownership | ROI likely won't be provable after go-live | Value |
Why Do These Signs Matter More Than the Root Causes Alone?
Knowing the five root-cause failure patterns governance gaps, people neglect, process chaos, data disasters, value evaporation is useful, but those patterns are often invisible until they've already caused damage. PMI's own research on project governance has found that governance failure is frequently cited as the root cause once a program has already underperformed by which point correction is far more expensive. These seven signs are the observable symptoms that show up while there's still time to intervene, which is exactly why they're worth watching for deliberately rather than discovering the root causes only in a post-mortem.
How Do You Run a Quick Health Check on Your Transformation?
1. Review your RAID log for stale items. Count how many risks or issues have appeared "still open" for three or more consecutive reviews a rising count is an early warning sign in itself.
2. Check your governance meeting cadence against its original schedule. Compare actual meeting length and frequency over the last two months against what was originally planned.
3. Ask three people to restate the business case from memory. Ask the sponsor, the program leader, and a workstream lead independently what the transformation is supposed to achieve — significant disagreement is a signal worth investigating.
4. Check whether change management activities have been rescheduled or cut in the last quarter. Compare the current change management plan against the original one for gaps.
5. Plot the timing of data issues discovered so far. If issues are being found progressively closer to the planned cutover date, validation processes need to move earlier in the schedule.
6. Confirm who currently owns benefits tracking, and how long they've held it. If ownership has changed more than once, or nobody can immediately answer, this needs to be resolved before go-live, not after.
What Should You Do If You're Seeing Several of These?
Seeing one or two of these signs isn't necessarily a crisis every program has rough patches. Seeing several at once, especially the RAID and governance-related ones, is worth an honest program health check before they compound into the kind of failure that's much harder to reverse.
Take the Transformation Readiness Assessment →
Frequently Asked Questions
How many of these signs need to be present before a program is genuinely at risk?
There's no fixed threshold, but signs clustering around governance and RAID specifically (signs 1, 2, and 3) tend to be the most reliable early indicators, since they reflect whether the program's decision-making structure is actually functioning.
Can a program recover after showing several of these signs?
Yes, often — the earlier these signs are addressed, the cheaper the correction. Programs that wait until go-live to address governance drift face a much harder recovery than ones that intervene mid-program.
Who is usually best positioned to notice these signs first?
Program leadership and PMO staff closest to the RAID log and governance cadence tend to notice first, though they're not always the ones with the authority or incentive to raise it — which is itself part of why governance gaps persist.
Is a shrinking status report always a bad sign?
Not automatically, but a pattern of shrinking detail over multiple consecutive reporting cycles, especially alongside an unchanged "on track" status, is worth investigating rather than assuming it reflects genuine stability.
Should this health check be run once or repeated throughout the program?
It's most useful as a recurring check monthly or at each governance cycle — rather than a one-time assessment, since these signs tend to develop gradually and are easiest to correct when caught early.