The Simulation Is Over
Success depends on how the team responds when the unexpected appears.
Every production has a final dress rehearsal. The lights are set, the cues are called, and the cast knows every line. Then comes opening night, and the one element no rehearsal can supply takes its seat: the audience.
Something will go differently. A prop fails. A cue comes late. A line lands in silence. No one in the house will know the rehearsal was flawless. They will remember how the company carried on.
President Eisenhower made the same point in 1957, repeating a saying from his Army years: "Plans are worthless, but planning is everything." An emergency, he explained, is by definition unexpected, so it will not happen the way you planned (Eisenhower, 1957).
When opening night is life and death
Now raise the stakes as high as they go.
On July 20, 1969, two men were descending toward the Moon in a spacecraft roughly 240,000 miles from home. No one had ever landed there. The descent had been rehearsed in simulators again and again. There would be no second take.
Minutes from the surface, the computer raised an alarm. Neil Armstrong, a man known for saying little, radioed Houston for a reading on the 1202 program alarm. The choices were stark: abort the first Moon landing, or continue and risk the crew.
The answer had to come in seconds. It came from a 26-year-old guidance officer and his even younger deputy in a back room.
Weeks earlier, in a simulation, the team had called an abort that turned out to be unnecessary. After that failure, they wrote a one-page list of every alarm code and what each one meant. On the day, the deputy had the list in front of him. The alarm meant the computer was overloaded, not that the spacecraft was in danger. The call was go. More alarms followed, and the call held. Eagle landed (Space.com, 2019).
No simulation had run exactly this way. What the simulations had produced was a team that knew what to do when the real descent left the script.
The rehearsals did not prevent the surprise. They prepared the response. In the moment, the mission came down to how the people responded.
As the Apollo 11 astronauts rehearse their lunar landing mission in simulators, they pause in front of a lunar module mockup in the Flight Crew Training Building area. From left, are Command Module Pilot Michael Collins, Commander Neil A. Armstrong, and Lunar Module Pilot Edwin E. Aldrin, Jr.
Image credit: NASA June 19, 1969
A transformation is not an installation
Installing software is a technical event. A process and technology transformation changes how people work, how decisions get made, and how money moves, all on the same morning.
No test environment carries the full weight of live volume, live data, and people doing unfamiliar work under real deadlines. Something unaccounted for will surface. That is not evidence of poor preparation. It is the nature of a launch.
So the question that matters in week one is not "why wasn't this perfect?" It is "how well are we responding?"
What the research supports
Gartner forecasts that by 2027, more than 70% of recently implemented ERP initiatives will fail to fully meet their original business case goals, and as many as 25% will fail catastrophically (Gartner, ERP insights). That is a forecast about business outcomes, not about launch day. Those outcomes are decided in the weeks after go-live.
Gartner's guidance on the go-live period itself comes down to three practices (Gartner, 2017):
Choose the go-live date with the business, and support it with a managed ramp-down and a controlled ramp-up.
Set up hypercare for the period immediately after go-live, with exit criteria defined in advance on both business and IT metrics.
Move into a stabilization period while the business gains competence in the new processes and systems.
Five priorities when the launch gets rough
These come from our own practice leading large transformations. They are consistent with the guidance above, and we offer them as field experience, not as research findings.
Protect what keeps the business running. Paying people, billing customers, collecting cash, meeting compliance obligations. Everything else waits.
Run hypercare like mission control. One intake, one triage, named owners, clear authority to decide.
Let the business set the priorities. The question is not which defect to fix next. It is which interruption hurts the business most.
Freeze scope. No enhancements, no new requests, no side projects until operations are stable.
Measure outcomes and report on a cadence. Count completed pay runs and invoices, not closed tickets. Tell people what happened, who owns it, and when the next update comes.
Three disciplines that decide the outcome
Stay calm and focused. Hypercare preparedness and execution.
Investigate, don't blame. Forensic root cause analysis works. Finger-pointing and overreaction do not.
Run three clocks at once. Get through today, recover from week one, and prepare for week two.
You can do everything right and still have a rough launch. What you do next is the part that counts.
References
Eisenhower, Dwight D. "Remarks at the National Defense Executive Reserve Conference." November 14, 1957. The American Presidency Project.
Gartner. "How to Plan Your Postmodern ERP Project's 'Go-Live' and 'Post-Go-Live'." Research note, July 31, 2017.
Gartner. "Enterprise Resource Planning (ERP) Insights." Citing the 2024 research "What IT Leaders Must Do to Avoid Disappointing ERP Initiatives."
Howell, Elizabeth. "Apollo 11's Momentous Landing Came Down to One Tough Call." Space.com, July 15, 2019.
