Beyond Go-Live Day
by Braden Kelley and Art Inteligencia
What Metrics Prove Change Worked Beyond Go-Live? (Short Answer)
Six metrics that prove change worked beyond go-live day: (1) named behavior adoption, (2) old-path / workaround retirement, (3) time-to-confidence, (4) post-hypercare relapse rate, (5) BAU reinforcement coverage, and (6) outcome movement on the funded human or business result. Completions, cake, and “system up” photograph the project. These metrics score whether the new way of working survived contact with real work.
Go-live proves the system shipped. These six metrics prove humans changed — and stayed changed.
Why Isn’t a Green Cutover Proof That Change Worked?
I have watched rooms celebrate a green cutover while the median person still lived on the spreadsheet, the shadow path stayed open “just in case,” and nobody in BAU owned reinforcement after the PMO left. Cutover is a technical event. Proof that change worked is behavioral and durable.
Anti-metrics that can inform but do not prove change worked: training completions, town-hall attendance, cutover RAG, license/login occupancy, and “awareness” pulse scores. Soft landings score maintenance and behavior — not cake.
Leaders who still treat go-live as transformation should pair this scorecard with 8 Change Myths Leaders Still Believe in 2026. For the method spine behind Maintenance, see 5 Steps of Human-Centered Change.
| Metric | Replaces | Proves |
|---|---|---|
| 1. Named behavior adoption | Training completion %; awareness scores | The success behavior is happening in BAU |
| 2. Old-path retirement | Endless dual-run celebrated as kindness | The easy old way actually died |
| 3. Time-to-confidence | Speed-to-cutover; e-learning hours | Median person succeeds without heroics |
| 4. Post-hypercare relapse | Hypercare ticket volume alone | Behavior stuck after applause |
| 5. BAU reinforcement coverage | “Project closed”; RAG with no BAU owner | Someone still owns the new way |
| 6. Outcome movement | Milestone burn-down; “launched on time” | The funded human/business result moved |
1. How Do You Measure Named Behavior Adoption After Go-Live?
Proves: The success behavior written into the change plan is happening in BAU — not only in training.
Replaces: Training completion % and “aware / informed” pulse scores.
Measure: One named behavior per change — for example, complete in the new system without a shadow spreadsheet, use the new handoff, or stop dual entry. Sample or use system logs. Trend after day one, day thirty, and day ninety.
Soft-landing use: If behavior is flat, diagnose design and incentives — not “resistance.” Adoption without a named behavior is theater with better attendance.
2. How Do You Prove Old-Path and Workaround Retirement?
Proves: Adoption was not optional because the shadow path stayed open forever.
Replaces: “Go-live” with dual-running celebrated as kindness with no kill date.
Measure: Kill-date honor rate; volume on the legacy process toward zero or a hard floor; count of retired workarounds. Immortal shadows turn good tools into shelfware — see 12 Adoption Mistakes That Turn Good Tools Into Shelfware.
Soft-landing use: Time-boxed coexistence with an explicit stop. Retirement is part of value, not cruelty.
3. What Is Time-to-Confidence — and Why Does It Matter?
Proves: Enablement worked — practice and redesigned work, not only broadcasts.
Replaces: Speed-to-cutover and hours of e-learning.
Measure: Time from first exposure to unsupervised successful completion — or self-reported confidence paired with observed success. Cohort by role. The median matters more than the hero cohort.
Soft-landing use: Long tails mean managers and jobs need redesign, not another cascade deck.
4. How Do You Track Post-Hypercare Relapse Rate?
Proves: Maintenance is real — the change did not relapse when the PMO left.
Replaces: Hypercare ticket volume as the only post-launch dashboard.
Measure: Return-to-old-path rate; reopened shadow tools; drop-off in new-way usage thirty to ninety days after hypercare ends. Lower is better.
Soft-landing use: Relapse triggers reinforcement, incentive checks, and owner escalation — not surprise at day ninety. Many “stalls after the pilot” are relapse with a different name; see 9 Reasons Digital Transformations Stall After the Pilot.
5. What Is BAU Reinforcement Coverage?
Proves: Change maintenance has an operator — rituals, coaching, and drift response are running.
Replaces: “Project closed” and RAG green with nobody in BAU accountable.
Measure: Named owner exists; reinforcement rituals held as planned; drift signals reviewed; percent of sites or teams with active reinforcement.
Soft-landing use: No owner = no proof. Coverage is a leading indicator of sustained adoption. Maintenance is a step in human-centered change, not a footnote after cake.
6. How Do You Measure Outcome Movement on the Funded Result?
Proves: The change delivered the human or business outcome it was sold on — effort, quality, safety, cycle time, customer or employee success — not only a new system.
Replaces: Milestone burn-down, vendor acceptance, and “we launched on time.”
Measure: Baseline → after on one primary outcome tied to the original case for change. Ranges are fine. Fake precision is not required. (Innovation portfolios need a different scoreboard — see 12 Metrics That Actually Measure Innovation Value — this metric is the change-funded result.)
Soft-landing use: Behavior without outcome invites a harder question — wrong behavior or wrong bet. Outcome without behavior invites theater and luck.
What Should You Ask in the Next Steering Review After Cutover?
Five questions before the next steering review after go-live:
- Which named behavior moved — and how do we know?
- Is the old path actually dead on a date?
- What is time-to-confidence by role?
- What is the relapse rate since hypercare ended?
- Who owns reinforcement in BAU — and did the funded outcome move?
Mantra: Stop proving you launched. Start proving humans changed — and stayed changed.
FAQ: Change Management Metrics Beyond Go-Live
How do you measure if change management worked?
Measure whether change management worked with behavior adoption, old-path retirement, time-to-confidence, post-hypercare relapse, BAU reinforcement coverage, and movement on the funded outcome — not training completions, town halls, or a green cutover alone.
What metrics prove change adoption?
Metrics that prove change adoption include a named success behavior trending in BAU, legacy and workaround volume falling after a kill date, and low relapse after hypercare — paired with a BAU owner still reinforcing the new way.
Why is go-live not enough to prove change worked?
Go-live proves the system shipped and cutover ran. It does not prove the median person adopted the new behavior, retired the old path, reached confidence without heroics, or sustained the change after the project team left.
What is change maintenance measurement?
Change maintenance measurement tracks whether a named BAU owner runs reinforcement rituals, reviews drift, and keeps relapse low after hypercare — so the new way of working survives when the PMO closes the project.
How do you measure behavior change after go-live?
Measure behavior change after go-live by naming one success behavior, sampling or logging it at day 1 / 30 / 90, tracking return to the old path after hypercare, and confirming time-to-confidence for the median person in each role — not awareness scores alone.
Image credits: Gemini
Content Authenticity Statement: The topic area, key elements to focus on, etc. were decisions made by Braden Kelley, with a little help from Google Gemini and Cursor to clean up the article, add images and create infographics.
Sign up here to get Human-Centered Change & Innovation Weekly delivered to your inbox every week.