The short answer
Resistance to a process change should not be treated automatically as a problem to overcome, nor as evidence that the proposed change is wrong. It is information that needs to be interpreted.
An employee may be protecting a control that management has overlooked, knowledge that has never been documented, a workable relationship between teams, a manageable workload, professional standing or a job. In other cases, the objection may reflect familiarity with the current process, incomplete information or a preference that cannot reasonably determine the design.
The manager's task is to establish what the objection is about, decide whether it changes the case for the proposed process, and then implement the decision clearly.
Find out what the existing step does
A process map records a step's formal purpose. It may miss functions that have developed around it. An accounts clerk's spreadsheet can duplicate the ERP while also holding informal payment arrangements. A reporting meeting may be the only regular point at which sales and production coordinate. An approval can create delay while protecting against a genuine pricing risk.
Removing the activity requires a decision about those functions. Some should disappear; others need to survive through a better mechanism. Speak to the people performing the work and those using its output. Ask what would go wrong if the step stopped, who relies on it, which failures it has caught and what information exists nowhere else.
The history matters too. A report or control may have been a sensible response to conditions that have since changed. Understanding its origin allows the organisation to preserve its purpose without preserving its method.
Where a low-risk step has no clear function, suspending it for a defined period can reveal dependencies. Do not use that experiment for legal, financial or safety controls whose importance may become visible only after a serious failure. Before removing an output, check any contractual, regulatory or audit obligation and agree a replacement with external users who still need the information.
Why people react differently to the same change
Organisational research does not support a simple distinction between employees who embrace change and employees who resist it.
A review of 79 quantitative studies by Oreg, Vakola and Armenakis found that reactions to organisational change are associated with several categories of factors, including individual characteristics, the existing organisational context, the way the change is implemented, the content of the change and whether employees expect it to benefit or harm them. A 2026 review by Oreg and Sverdlik reaches a similarly cautious conclusion: responses to change are multidimensional, can be ambivalent and can evolve over time.
This matters for management because the same objection can have different causes, and different objections require different responses.
A concern may reflect a real loss
Process changes can alter more than the sequence of tasks. They can affect autonomy, workload, access to information, contact with colleagues or clients, opportunities to use a particular skill, professional status and, in some cases, employment itself.
Those consequences do not determine whether the change should proceed. They do affect how an employee is likely to evaluate it.
A person whose work becomes easier may welcome the same redesign that another person experiences as a loss of responsibility. A manager who describes both reactions as attitudes towards "change" misses the more useful question: what does each person expect to gain or lose?
This distinction is particularly important when technology is involved. If a system removes half of someone's current work, reassurance about "higher-value tasks" is not enough unless those tasks have been identified and there is a credible plan for the role.
Familiarity can influence judgement
The current process also benefits from being familiar. In their 1988 study of status quo bias, Samuelson and Zeckhauser found that people can disproportionately favour an existing option over an alternative.
The concept is relevant, but it is easy to misuse. Familiarity is not the only reason employees may prefer the current process. They also know its exceptions, its informal workarounds and the people to contact when it fails. Those are real operational advantages, even if they are not documented.
Status quo bias is therefore a reason to examine whether familiarity is influencing a judgement. It is not a reason to dismiss the judgement.
People can become attached to work they have shaped
Organisational psychology uses the term psychological ownership to describe the feeling that a job, project, process or organisation is "mine" or "ours" even without legal ownership.
A meta-analysis of 141 studies by Zhang and colleagues found that psychological ownership is associated with antecedents including control, intimate knowledge of the object and personal investment in it. These conditions can develop naturally when somebody has designed, improved or operated a process for years.
This does not mean that attachment makes an objection invalid. It helps explain why a technical judgement about a workflow can also be experienced as a judgement about past work or current competence.
The practical response is to separate the process from the capability behind it. A manual workaround may no longer be needed while the person's knowledge of exceptions, customers or failure modes remains valuable.
Be clear about what is genuinely open to discussion
Participation is useful only if the scope of that participation is clear.
Sometimes the decision itself is still open. Management may be testing whether an approval is necessary or comparing several possible workflows.
In other cases, the direction has already been decided. A new system may have been purchased, a regulatory requirement may leave little discretion, or management may have concluded that a particular activity must stop. The remaining questions concern implementation rather than the decision itself.
Both situations are legitimate. Confusing them is not.
Asking employees for views on a decision that has already been taken creates the appearance of influence without the substance. A more credible approach is to state what has been decided, what remains open and what evidence could still change the design.
This is also consistent with the broader organisational-justice literature. A meta-analysis of 183 studies by Colquitt and colleagues found distinct relationships between perceived distributive, procedural, interpersonal and informational justice and outcomes including trust, organisational commitment, evaluations of authority, withdrawal and performance. The way decisions are reached and explained matters alongside the outcome itself; a fair process need not make an unwelcome result popular.
For a manager, the implication is straightforward: where an outcome is difficult, there is little value in adding uncertainty about whether the process used to reach it was genuine.
Use participation to improve the design
The strongest reason to involve employees is not to increase enthusiasm. It is to improve the quality of the change.
People who perform a process know where exceptions occur, which information is missing from formal systems, which handoffs depend on informal relationships and which controls exist because of failures that may no longer be visible to management.
Ask where the proposed process is likely to fail, which exceptions are missing and what training or authority people would need. Compare those answers with real cases rather than treating them as a vote for or against change.
Management remains responsible for choosing between competing interests. Participation gives people influence where their knowledge can improve the decision; it does not transfer every design choice to them.
Keep a record of what objections change
Listening is useful only if the information reaches the decision. A short objection record helps distinguish a design problem from an unresolved personal consequence or a disagreement management has considered and declined to follow.
| Objection | Evidence to examine | A possible response |
|---|---|---|
| "The spreadsheet contains information the ERP does not." | Compare the records and identify who uses the additional information | Preserve the missing information in an agreed system before retiring the spreadsheet |
| "This saves sales time but gives us more work." | Follow representative cases through both teams, including exceptions | Change the handover or allocation of work; measure the effect across both functions |
| "I will lose the part of my role I am good at." | Specify the future role, required skills and actual work available | Make a concrete role and training decision; acknowledge losses that cannot be avoided |
| "The old method is more reliable." | Agree a quality measure and compare the two methods on the same cases | Retain the old route if evidence supports it, or explain why the new one is sufficiently reliable |
For each material concern, record who will investigate it, the evidence considered, the decision and the reason. State whether the concern changed the design, requires a later review or has been considered without changing the decision. Give unresolved items an owner and a date.
This makes consultation inspectable. It also prevents the same objection from remaining indefinitely open because nobody has answered it, while preserving a route for new evidence to reopen a decision.
Be precise about personal consequences
Broad reassurance is particularly risky when the change affects the content of a role.
If someone currently spends half of the week preparing a report that will disappear, saying that the person will have more time for valuable work does not answer the obvious questions. What work will replace it? Does the employee have the necessary skills? Will responsibilities increase or decrease? Is there enough work to fill the released capacity? Could the role itself change?
Where the answers are known, they should be stated. Where they are not, uncertainty should be identified as uncertainty.
A useful distinction is between:
- what will definitely change;
- what remains undecided;
- who will take the next decision, on what basis and when.
This is especially important where jobs may be affected. Management should not promise employment security that it cannot guarantee. If a role is known to be disappearing, the person concerned should hear that directly rather than infer it from a wider announcement about efficiency.
The human element of process redesign is not separate from the operational one. If the work changes materially, the role may need to change with it.
Make the final decision explainable
After the relevant information has been collected, management still has to decide.
A defensible explanation should address four points:
- What problem is being solved?
- What evidence supports the chosen change?
- Which concerns raised during consultation affected the design?
- Why were the main alternatives not chosen?
Consider an approval that adds two days to a quotation process. If the review shows that the approval rarely changes the commercial decision but does catch unusual pricing risks, removing it without replacement would be careless. Keeping it unchanged would preserve the delay.
A better decision may be to remove the general approval while introducing a targeted review for quotations with defined risk characteristics.
The explanation should make that reasoning visible. Employees do not need to agree with the conclusion to understand why it was reached.
This also provides management with a useful discipline. A decision that cannot be explained without relying on vague claims about efficiency, transformation or resistance may not yet be sufficiently developed.
Translate the decision into roles
For each affected role, specify what stops, what starts, what still requires judgement and which exceptions now reach that person. Establish the authority to act, the training required and the person who will resolve questions during transition.
When automation removes routine cases, the remaining work can become harder. A role that handled one hundred cases may now handle twenty because those twenty contain the ambiguity or risk the system cannot resolve. Staffing and training should reflect that change in content, rather than assume that fewer cases mean proportionately less demanding work.
End the parallel process deliberately
Running old and new processes in parallel can be appropriate for a limited period where failure would be costly. Without a defined end, however, temporary duplication tends to become permanent.
The organisation then carries two sources of truth, duplicate controls and employees choosing between procedures case by case.
A clean transition normally requires:
- a cutover date;
- a named owner for the new process;
- a documented fallback if the new process fails;
- a review after enough real cases have passed through it.
The old route should then be retired deliberately. Forms can be archived, folders closed, reports discontinued and procedures updated.
If employees continue to rely on the old process, the reason should be investigated before assuming unwillingness to change. The behaviour may reflect habit, but it may also reveal information, functionality or confidence that the new process has not yet replaced.
Distinguish disagreement from non-compliance
Disagreement should remain possible throughout the design and early implementation of a change.
Employees close to the work may identify risks that management has missed, and a process that suppresses criticism loses information. Management should also be prepared to revise a change if evidence shows that the new process performs worse than expected.
The position changes once the decision has been made, the reasons have been explained, responsibilities are clear and a reasonable transition has been provided.
An employee may still disagree with the decision. Continuing to use a retired procedure, withholding information required by the new one or deliberately undermining the agreed process is different. At that point, the issue is no longer whether the employee supports the change. It is whether the agreed way of working is being followed.
The distinction matters because labelling all disagreement as resistance is poor management, while allowing disagreement to become a permanent individual veto makes a process impossible to govern.
A narrow Swiss employment-law point
If process redesign may result in dismissals, the employment-law consequences should be treated as a separate workstream rather than as part of ordinary change management.
Under Art. 335d of the Swiss Code of Obligations, the statutory collective-redundancy rules apply to employer-initiated dismissals within 30 days for reasons unrelated to the individual employees when the number affected reaches:
- at least 10 employees in an establishment normally employing more than 20 but fewer than 100 people;
- at least 10% of employees in an establishment normally employing at least 100 but fewer than 300 people;
- at least 30 employees in an establishment normally employing at least 300 people.
Where those rules apply, Art. 335f requires consultation with employee representatives or, where there are none, the employees themselves. They must have an opportunity to propose ways to avoid or reduce the dismissals or mitigate their consequences. Art. 335g also requires notification to the competent cantonal employment office.
Establishments with 20 or fewer employees fall outside these specific collective-redundancy thresholds, but other employment-law obligations remain applicable.
Complete the transition and keep a route for new evidence
A change is established when the new process works, responsibilities are understood and the old route no longer needs to survive beside it. Review actual cases after cutover: whether the intended result improved, whether work moved to another team and whether the fallback is usable.
New evidence should be able to change the design. A concern that has already been examined and answered should not remain indefinitely unresolved. Keeping both principles visible allows an organisation to learn from objections while still making and implementing decisions.
Method note. The workplace examples are hypothetical. Psychological findings describe patterns across groups and settings, not the motives of an individual employee. The often repeated claim that 70% of organisational changes fail is not a reliable universal failure rate; Hughes's review found no valid empirical basis for that specific claim. The employment-law section provides general information, not legal advice; advice on the actual circumstances is needed where dismissals are being considered.