A program status report can stay green for months while the work underneath it is already off track. Gartner puts the share of recently implemented ERP initiatives that will fail to fully meet their original business case goals by 2027 at more than 70%. That gap between what the deck says and what the program is actually doing is exactly where an outside review earns its cost, and seven checkable signs tend to show up before the deck catches up to reality.
The seven signs, and the record that confirms each one
Each sign below has a specific record behind it, not a feeling. Pulling the record first, before asking the program team how things are going, is what keeps a review from just restating the status deck in different words.
| Sign | What it looks like | Record to pull |
|---|---|---|
| Defects rising while status stays green | Severity 1 and 2 count climbing for three or more weeks | The open-defect count by week, last eight weeks |
| Too few full-volume data mock loads | Fewer than three planned before go-live | The mock-load schedule and the record load success rate on the last run |
| Signed-off designs being reopened | Change requests citing an already-approved design | Change requests by process area, cross-checked against sign-off dates |
| SI key-role turnover | More than 20% of named roles replaced in six months | The statement of work's staffing list against who is actually on the program today |
| Steering committee stops deciding | Meetings run short with no decision recorded | The decision log and attendance for the last three meetings |
| Scope slipping to an unfunded phase two | Features moved out of scope without a funded plan to deliver them | The phase two list against the approved business case and its funding |
| Process owners missing design and test | The business side is not in the room for its own decisions | Attendance by named owner against who actually signed the design documents |
Three or more signs tripped at once is the threshold worth acting on. One sign in isolation is common in any program under normal pressure. Three together, on our view, is the point where a program's own reporting can no longer be trusted to catch the trend on its own.
What a review costs against what a rescue costs
A point-in-time review is a bounded piece of work: one to three weeks, depending on program size, producing a written red, amber, or green call with the records behind it. Set against a $5,000 to $20,000 a month PMO rate as a reference point, a single review is a small fraction of a month of program spend.
A rescue is a different order of cost. Deliberate Directions reports NetSuite program rescues running 30 to 50% of the original budget, on top of the money already spent reaching the point where a rescue was needed. That figure is from one systems integrator's rescue practice, so treat it as directional for other platforms rather than a universal rate; the direction, that fixing it late costs far more than checking it early, holds regardless of the exact multiple.
Who the review should report to
A review that reports into the program manager it is reviewing tends to land softly, because the person receiving the finding is also the person whose work produced it. The useful version reports to the sponsor, the steering committee, or whoever outside the program holds the budget and the go-live date. That single structural choice, more than the reviewer's résumé, decides whether a finding changes anything.