The new system goes live on Monday. By Thursday, the old spreadsheet is back. Nobody announced it. It simply returned, passed from one colleague to another, because a report was due and the new screen didn’t show what people needed.
Moments like this rarely appear in project plans, and they’re easy to dismiss as resistance. More often, they’re information. Something about the change doesn’t yet fit the work.
A statistic worth leaving out
Articles on this subject often open by claiming that 70% of change initiatives fail. It’s a memorable figure, but a 2011 journal review found nothing valid or reliable behind it. We don’t need it.
More than a change of tools
In our view, digital transformation is rarely about replacing one tool with another. It’s about improving how work, information and responsibility move between people, processes and systems: who picks up a task, where the facts live, who decides, and who hears when something changes.
New software can leave all of that untouched. When it does, the project may be finished, but the change hasn’t happened.
Start with the work as it really happens
Most projects document the official process. The more useful version is usually the unofficial one: the shared inbox only one person understands, the file kept “just in case”, the phone call that resolves what the form can’t.
Before choosing or configuring anything, spend time with the people doing the work and ask:
- What do you do when the system doesn’t give you what you need?
- Which task takes longer than it should, and why?
- What would you refuse to give up?
- Who do you ask when you’re stuck?
Each workaround solves a real problem. If the new system ignores that problem, the workaround will come back.
Useful comes before easy
Training is often where projects put their effort, and it matters. But people first need a reason to bother.
Back in 1989, Fred Davis pointed to two beliefs that shape whether people use a system: that it will help them do their job, and that it will be easy to use. Usefulness had the stronger link with actual use. People will tolerate a clumsy tool that really helps them, and they will avoid an elegant one that doesn’t.
So describe the benefit from the user’s position. “Better data visibility for management” helps management. “You won’t have to re-enter the same order in two places” helps the person doing the entering. If you can’t write that second kind of sentence for a group of users, stop and find out why. They may be taking on extra work so that somebody else benefits, which can be fair but needs saying openly.
Find out where people are stuck
“People are resisting” is too vague to act on. Prosci’s ADKAR model is more precise, following a person’s progress through awareness, desire, knowledge, ability and reinforcement. What matters is the first one that’s missing, because each calls for a different response.
Consider three people in the same team. One doesn’t know why the change is happening, and needs an honest explanation. Another understands but thinks it will make their job worse; more training won’t help, but a real conversation might, and perhaps a design change. A third has attended the training but freezes when a real customer is waiting, and needs practice and someone nearby to ask.
Treating all three with the same email and the same webinar helps none of them.
Make it safe to say it isn’t working
The earliest warnings about a change usually come from the people using it. Whether those warnings reach anyone depends on whether speaking up feels safe.
Amy Edmondson’s 1999 study of 51 work teams found that where people shared the belief that it was safe to take interpersonal risks, they were also more likely to ask for help and talk about errors. That matters during a rollout, when everyone is temporarily worse at their job than they were last month.
Say in advance that the first weeks will be uneven. Ask “what’s getting in your way?” rather than “why aren’t you using it?” And when someone reports a problem, fix something visibly and tell them you did.
What leaders actually need to do
Sending the launch message isn’t sponsorship. In our view, three leadership actions matter more:
Use it yourself. If the new reporting tool is mandatory but the management meeting still runs on the old slides, everyone will draw the obvious conclusion.
Make room for it. A change landed on top of a full workload usually loses. Decide what can wait during the transition, and give any named champions the time to do it.
Set an end date for the old way. Running two systems indefinitely invites people to stay on the familiar one. Announce when the old route closes and what happens to its data. Then keep to it, unless the evidence says the new way isn’t ready.
Communicate the answers people are looking for
People rarely need more messages. They need answers:
- What’s changing for me, and when?
- What stops?
- Who can help me, and how?
- What happens if it goes wrong?
Be plain about what isn’t decided yet. “We don’t know yet, and we’ll tell you by Friday” builds more trust than a confident message that later has to be corrected.
Live, but not yet adopted
Here’s an invented example. A company rolls out a new CRM. The data is migrated on schedule, the integrations work and every salesperson completes the training. Technically, it’s a success.
Six months later, salespeople update their records on Friday afternoons because their manager asks them to. The real customer history is still in inboxes. Service colleagues phone the account owner before speaking to a customer, and the forecast meeting still opens with an argument about whose number is right.
The technology hasn’t failed. But the intended change hasn’t happened. Information moves as it did before. Adoption would look different: a service colleague opens the record and finds the last conversation, and an account changes hands without a long handover call.
Measure use, not launch
Go-live is a date on a plan. Adoption is what happens afterwards, and log-ins alone won’t show it. More useful signs include:
- The share of the work that’s completed in the new process from start to finish.
- Whether the old files and side channels are still in use.
- How long a typical task takes now, compared with a baseline taken before launch.
- What people are asking for help with, and whether those questions change over time.
A team that lags behind may be stubborn, or it may have a legitimate case the design missed.
Before you go live
If you’re planning a change now, five questions can show how ready you are:
- Have we watched the work being done, not just read the process document?
- Can each affected group say what’s in this for them?
- Do we know what the old way’s end date is?
- Do managers have time to support their teams through the transition?
- Have we agreed how we’ll measure use once the launch is over?
If the honest answer to several of these is no, the technology probably isn’t your main risk. These gaps are usually easier to close before launch than after, and where it helps to have support, that’s the purpose of our Organizational Change Management service.
A successful go-live shows that the system works. A successful transformation shows up later, when work, information and responsibility move more easily between the people and systems that depend on them. The first can be announced on a single day. The second you’ll see in how people work once nobody is watching the launch.