Prosci's Best Practices in Change Management research has tracked what actually predicts project success across more than two decades and 11 separate studies. In every one of them, active and visible sponsorship comes out as the single largest contributor — cited far more often than training, communication, or any other factor. The gap in outcomes is stark: projects with extremely effective sponsors were 79% likely to meet their objectives, compared with just 27% for projects with extremely ineffective ones. That's not a marginal difference. It's close to the whole game.
But sponsorship is only the most visible failure point. Underneath it sit four more roles that rarely get the same attention, and any one of them can quietly stall what the sponsor set in motion.
1. The sponsor who signs off but never shows up
An executive sponsor who approves the budget, sends a kickoff email, and then disappears isn't sponsoring — they're permitting. Prosci's research is specific about what "active and visible" actually requires: the sponsor has to participate visibly throughout the project, build a coalition of other leaders who back it, and communicate directly with employees about why the change matters, not just that it's happening. Skip any of those three and the label "sponsor" stops meaning much. Employees take their cues from what leadership does repeatedly, not what it announced once.
2. The middle manager caught in the squeeze
Middle managers are where most rollouts actually live or die day to day, because they're the ones fielding questions, modeling the new process, and absorbing the complaints their teams don't want to take upstairs. Gartner's July 2024 survey of 473 HR leaders found that 74% believe their managers aren't equipped to lead their teams through change — not unwilling, unequipped. That's a capability gap, not an attitude problem, and it means most managers need direct coaching on how to talk about the change before you can expect them to reinforce it.
A manager who doesn't understand why a change is happening can't convincingly explain it to their team — and their team can tell the difference between genuine backing and a forwarded memo.
If your rollout plan reaches executives and end users but skips managers in between, you've left the busiest translation layer without a script.
3. The other three: IT, the frontline, and the informal influencer
Three more roles round out the list, and each fails a rollout in its own way:
- The IT or technical owner who inherits the system after go-live. Brought in only to execute someone else's decisions, they end up maintaining a configuration they never had a chance to flag problems in — and the workarounds start early.
- The frontline end user who does the work the system is meant to support. If the new process makes their job measurably harder in month one, no amount of messaging about long-term benefit will stop them from reverting to the old way the moment no one's watching.
- The informal influencer — not always the most senior person in the room, but the one colleagues actually ask before they trust something new. Win this person over and they do half your change management for you; ignore them and they become the most credible source of doubt in the building.
Managing five people, not one plan
The instinct on most projects is to write a single communication plan and broadcast it to everyone. That's why sponsorship, manager readiness, and frontline usability keep showing up separately in the research as distinct failure points — they require distinct interventions, not one message repeated five ways. A sponsor needs a role to play publicly and repeatedly, not a signature. A manager needs to understand the change well enough to answer their team's questions, which means coaching before rollout, not a slide deck at launch. Frontline users need the new process to be at least as workable as the old one on day one, not eventually. And influencers need to be identified and looped in early enough that they're advocating before the rest of the organization has formed an opinion.
None of this requires a bigger budget. It requires treating these five roles as five separate risks to manage, each with its own plan, rather than five audiences for the same rollout announcement.
