Sustaining Momentum: Avoiding Pilot Purgatory and Initiative Fatigue
Deliberate Academy Editorial Team
Reviewed for accuracy and professional relevance
You're 8 lessons in — don't lose your progress.
Sign up free to save where you are and earn a verified certificate when you pass.
- Identify the predictable timeline at which AI transformation programs most commonly lose momentum, and explain the organizational dynamics behind it
- Distinguish initiative fatigue from legitimate program underperformance, and explain why treating one as the other leads to the wrong corrective action
- Apply a structured approach to renewing stakeholder energy and executive attention in the second and third year of a multi-year transformation program
- Design a momentum-sustaining checkpoint plan for a described transformation program entering its second year
Most AI transformation programs do not fail in month three, when skepticism is highest and scrutiny is closest. They fail in month eighteen to month thirty, when the original executive sponsor has moved to a new priority, the early adopter teams have plateaued, and the organization has quietly stopped treating AI transformation as a program with attention and instead treats it as background infrastructure that runs itself. This lesson addresses the distinctive leadership challenge of sustaining a multi-year transformation program past the point where its initial energy naturally fades — a different problem from the launch-phase challenges covered earlier in this course.
Why Momentum Fades on a Predictable Timeline
Transformation programs tend to lose momentum in a recognizable pattern. In the first six months, novelty and executive attention carry the program — early wins get visibility, and the sponsor actively defends resourcing. Between months six and eighteen, the program either builds real organizational habit (the hybrid structures and reskilling pathways from earlier lessons take hold) or it does not, and the gap between programs that sustain and programs that stall becomes visible. Past month eighteen, the specific risk is different: even genuinely successful early efforts face a fatigue period, where the people who drove the initial pilots have been reassigned, promoted, or moved on, and the organization has begun treating "we are doing AI transformation" as a completed fact rather than an ongoing program requiring continued attention.
This dynamic compounds with reorganizations and executive turnover, which are common enough over an eighteen-to-thirty-month window to derail even a well-designed program if its continuity does not survive a change in sponsor. A program whose success depends entirely on one executive's personal attention is fragile by design, regardless of how good that executive's early leadership was.
A telling early warning sign of momentum loss is when AI initiative progress reports shift from specific milestones and metrics to vague status language — "continuing to roll out," "ongoing adoption efforts" — without a specific number or date attached. This shift often precedes a visible stall by several months and is a more reliable early signal than waiting for adoption metrics to actually decline.
An AI transformation program is 22 months into a planned three-year timeline. The original executive sponsor was promoted to a different division eight months ago. Quarterly progress reports have shifted from specific initiative-level metrics to general statements about 'continued progress across the organization.' No formal decision has been made to end the program. What does this pattern most likely indicate?
Select one answer.
Distinguishing Fatigue From Real Underperformance
A critical judgment call for a leader at this stage is telling initiative fatigue apart from genuine program underperformance, because the corrective actions are opposite. Fatigue looks like: initiatives that were performing well on their original metrics but have simply stopped receiving attention, executive reporting, or renewed resourcing — the substance is fine, the energy is gone. Underperformance looks like: initiatives that are not meeting their original success thresholds regardless of how much attention they receive, and where the honest answer is not renewed sponsorship but redesign or discontinuation.
Treating fatigue as underperformance leads to unnecessary program cuts that abandon initiatives that were actually working. Treating underperformance as fatigue leads to renewed cheerleading and resourcing for an initiative that needed a fundamentally different approach, not more energy. The distinction requires looking at the actual metrics, not just the level of organizational attention — an initiative still meeting its original success threshold with declining visibility is a fatigue case; an initiative that has never met its threshold despite sustained attention is an underperformance case.
Build a structured eighteen-month and thirty-month checkpoint into every multi-year AI transformation program from the start, not as an afterthought. At each checkpoint, review every active initiative against its original success threshold and require an explicit answer to one question: is this initiative still on track, and does it still have a named, active owner distinct from the original transformation sponsor? Initiatives without a current, engaged owner are the clearest fatigue signal.
Renewing a Stalling Transformation Program at a Telecommunications Provider
Context
Twenty months into a planned three-year AI transformation program, the original executive sponsor had left for an external role, and no formal successor had been named. Progress reporting had become infrequent and general. Adoption metrics for the program's four most mature initiatives, when the new CTrO investigated, were actually still strong — three of the four remained above their original success thresholds — but two newer initiatives launched around month fifteen had stalled with almost no measurable adoption.
Action
The new CTrO ran the distinction this lesson describes: she confirmed the four mature initiatives were fatigue cases, not underperformance cases, since their metrics remained on track despite reduced visibility, and renewed executive sponsorship and quarterly reporting for those without changing their design. For the two newer, genuinely underperforming initiatives, she commissioned a focused review that found both had been launched without adequate stakeholder buy-in work from Lesson 2 of this course, and redesigned their rollout plans rather than simply renewing attention on the existing approach.
Outcome
Within two quarters, the four mature initiatives regained board visibility and their metrics held steady, confirming the fatigue diagnosis had been correct. The two redesigned initiatives, relaunched with proper stakeholder sequencing, reached 54% and 61% adoption respectively within five months of relaunch — up from under 10% each before the redesign. The CTrO reported the explicit fatigue-versus-underperformance distinction as the single most useful diagnostic tool in recovering the stalled program.
Why does the lesson recommend distinguishing initiative fatigue from genuine program underperformance before deciding on a corrective action?
Select one answer.
Exercise
Your Task
For a multi-year AI transformation program you are involved with, or a realistic one you construct, list its active initiatives at the eighteen-month mark. For each, record its original success threshold, its current actual performance against that threshold, and whether it currently has a named, actively engaged owner. Classify each initiative as healthy, fatigue-affected (meeting threshold but losing attention or ownership), or genuinely underperforming (missing threshold regardless of attention), and write one sentence on the specific corrective action each classification implies.
Success looks like
- Each initiative's classification is based on its actual metric against its original threshold, not on how much attention it is currently receiving
- The corrective action proposed for fatigue-affected initiatives (renewed sponsorship, no redesign) is clearly different from the action proposed for underperforming ones (redesign or discontinuation)
Watch out for
- Classifying an initiative as underperforming simply because it has lost visibility, without checking whether its actual metrics remain on track
- Renewing attention on a genuinely underperforming initiative without addressing the design problem that caused it to miss its threshold in the first place
Hint
If an initiative is meeting its original threshold but you cannot name its current active owner, that is a strong signal you are looking at a fatigue case, not an underperformance case.
A Common Failure Mode: The Program Without a Renewal Mechanism
The failure mode underlying this entire lesson is a transformation program designed as a launch event rather than an ongoing capability — one with a detailed plan for the first six months and no structural mechanism for renewing sponsorship, attention, or evaluation past that point. Programs designed this way survive as long as the original sponsor's personal attention lasts and no longer. The correction, consistent with the recommendation in this lesson, is to build scheduled checkpoints and an explicit sponsorship succession plan into the program's design from the beginning — treating momentum maintenance as a designed feature of the program, not something to improvise if and when attention starts to fade.
- AI transformation programs most commonly lose momentum in the eighteen-to-thirty-month window, driven by sponsor turnover, plateauing early-adopter teams, and the organization treating the program as a completed fact rather than ongoing work.
- A shift from specific milestone reporting to vague status language is an early warning sign of momentum loss that often precedes a measurable decline in adoption metrics.
- Distinguish initiative fatigue (healthy metrics, declining attention and ownership) from genuine underperformance (metrics below threshold regardless of attention) — the corrective actions are opposite, and misdiagnosing one as the other wastes effort or abandons working initiatives.
- Build scheduled eighteen-month and thirty-month checkpoints into every multi-year transformation program from the start, reviewing each initiative against its original success threshold and confirming it has a current, engaged owner.
- Design sponsorship succession into the program from the beginning — a program that depends entirely on one executive's personal, ongoing attention is fragile by design, regardless of how strong that executive's initial leadership was.