Why your team quietly avoids the new system

A team does not decide to avoid the new system. Nobody holds a meeting about it. The old way simply keeps working, the new step gets skipped or done badly, and three months later somebody concludes the staff are resistant. Usually they are not resistant to change. They are resistant to a slower, less certain, less sensible version of their own job.

What are the four reasons a new step gets quietly avoided?

It is slower than what they already know

This is the most common and the least forgiven. A new check-in form that takes four extra minutes a job is not four minutes. It is four minutes every time, on the busiest part of the day, for a benefit somebody else enjoys.

Time it honestly, old way against new way, including setup and including the moment of thinking about what to do:

Does the step pay for itself?
Added minutes × uses per day × working days a month
must be less than the minutes it saves per day

If it does not, the step will be skipped. Not out of disobedience — out of arithmetic, done honestly, by a person with a van to load.

They cannot tell whether they are doing it right

Nobody has seen a good example. The one example they saw was you, on a clean job, with time to spare. Meanwhile an error in the new system shows up on a report somebody senior reads, and the consequence of getting it wrong lands on them.

So they do the minimum that cannot be faulted, which is usually nothing.

Fix it by showing three real examples — including a messy one somebody actually did — and by saying out loud what happens when it is done wrong. If nothing happens, say that too. "The first three times you do it, I will check it with you, and after that nobody will look unless something is odd" removes most of the hesitation in one sentence.

It changes who gets to decide

The technician used to decide the parts and the sequence. Now the office does. The buyer used to order on a Friday afternoon; now the approval has to be back by ten the next morning, which is after the courier leaves.

Nobody objects to this out loud, because it sounds ungrateful. But every judgement call removed without being replaced with something is a demotion, and the people who noticed first are the ones you are losing.

Write down what changed in their authority and what did not. Where you have removed discretion, replace it with something they now own. That is not a bonus, it is the payment for the thing you took.

They cannot see the point from where they sit

The new system exists so that you, the owner, can finally see the numbers. The person filling it in sees extra typing on a Thursday afternoon.

If you cannot name a benefit in that person's own terms, delete the step. A step with no visible payoff has no defence, and the team is right to give it as little as possible.

What does "they will just do it anyway" actually sound like?

You will hear some version of these in the first fortnight. The reply matters more than the sentence.

Why does mandating without listening produce a slow quiet failure?

Because a mandate applied to a bad process produces work that satisfies the letter of the rule and defeats its purpose.

The tracker that is only filled in after the job, once the invoice has gone out, so the numbers look complete. The photos taken of the wrong thing because nobody said what the photos were for. The new form filled in from memory at the end of the week. Nothing is refused, nothing is escalated, and no rule is technically broken.

Then the data is wrong, and everybody blames the system. The system is the easiest thing in the building to blame, and it cannot answer back.

The fix is not a firmer deadline. It is removing the incentives that make the honest version expensive.

What changes when the people who do the work design the step?

This is the single change with the highest return, and it costs you a conversation rather than a budget.

1. Pick one step. Not the rollout, not the software — one step somebody has to perform.
2. Sit with the two people who perform it most, for fifteen minutes, at their normal workstation, not in the office.
3. Ask them to write out the step as they would actually do it, on paper, including what they would skip.
4. Ask what information is missing that forces them to guess.
5. Write their version down. Use it.

The person who knows why the process has three odd steps is the only person who can tell you which of those steps is now unnecessary. They are not resisting the change. They are the specification.

Most redesigned steps come out shorter. That is not a compromise — that is the point of asking.

How do you give each new step one owner and keep it short?

Write one line per step, in one place, with a name against it and a measured time.

Three rules that follow from it.

• No name against a step means the step will not happen. Ownership is the only part of this that is not optional.
• If a step takes longer than the old way, fix the step, not the people. Cutting it in half is a design decision, not a performance problem.
• If a step serves nobody's day but yours, cut it or make it optional. Keep the list short enough that you would still believe it in six months.

What do you do when the team is right that the process is bad?

Take it seriously, and then prove you took it seriously.

• Measure the step as it exists, not as it should exist. Time it for a week.
• If it is genuinely slower and serves no visible purpose, delete it. Say publicly that you deleted it because the people doing it raised it.
• If it is necessary, redesign it with them and change the workload around it — a new step of ninety seconds a job means something else has to go.
• Write down what changed because somebody said so, and who said so.

That last part is the compounding effect. The first time somebody's objection changes the process, you get an ally for the next change. The first time you override them and they were right, you lose the argument for a long time — and no amount of mandating will get it back.

How do you tell whether the people are avoiding it or the step is broken?

Watch. Sit with one person for ten minutes while they attempt the task on real work, and say nothing.

Note every point where they pause, hesitate, look at the old way, or ask whether they are doing it right. Each pause is a design fault, and the location of the pauses is your answer. A person who hesitates at the same step every single time has told you where the problem is without saying a word.

Three people, fifteen minutes each, beats any survey you could send.

And be honest about the diagnosis: most failed rollouts were not failed tools. They were bad process wearing a new tool.

Where do you start this week?

1. Write down every new step introduced in the last year, with one line each on why it exists.
2. Put a name against each one. Anything without a name is not running today.
3. Sit with the two people who do the most steps, for fifteen minutes each, and let them cut one.
4. Cut it. Say out loud that you cut it and why.
5. Add the benefit to each remaining step in the words of the person who does it, or delete that step too.

Not sure where to start? Book a free 30-minute strategy session and we will name the one gap worth fixing first: https://agentcy.co.za/#book-audit