Resistance to change: what the objection is telling you

“Resistance to change” is usually a verdict reached by the people leading a change about the people receiving it. Sometimes it is accurate. Often it bundles together different things: a flaw in the plan, a real loss, too much work for the time available, distrust left over from the last initiative, an obstacle nobody on the project has seen, or simply mixed feelings. Each needs a different response, and the label cannot tell you which.

The practical answer is to treat an objection as information before treating it as a problem. Sort what people say and do into kinds of concern, check your reading with them, then choose a response that fits: change the plan, negotiate, provide support or time, explain honestly, involve people in the design, or decide openly and accept that some disagreement will remain.

That does not make every objection right or every change negotiable. The aim is to know which situation you are in, and to respond without pathologising people or steering them with tactics you would not want to explain. Throughout, we follow a fictional hotel whose housekeeping team is asked to report room status on tablets instead of paper sheets and radio calls.

Why “resistance” is a label, not a diagnosis

The phrase has a history. Eric Dent and Susan Galloway Goldberg trace the term back to Kurt Lewin, who used it as a systems concept: a force affecting managers and employees alike. Later usage, they argue, dropped that context and turned it into a psychological concept that personalised the issue as employees versus managers. They proposed letting go of the term altogether.

Jeffrey Ford, Laurie Ford and Angelo D’Amelio made a related argument in 2008. The prevailing story, they wrote, favours change agents by presenting resistance as an irrational, dysfunctional reaction located “over there” in the recipients. Their “rest of the story” proposes that change agents contribute to resistance through their own actions and inactions, and that resistance can be a resource for change.

Even the classic management text leaves room for this. Kotter and Schlesinger’s “Choosing strategies for change”, first published in 1979, counts a belief that the change does not make sense for the organisation among four common reasons for resistance. Sometimes the people closest to the work have analysed it better.

So treat “the team is resistant” as a hypothesis about other people’s motives, formed by someone with a stake in the outcome. A useful stakeholder map records the concern itself, such as loss of discretion or extra workload, instead of filing a group under “resistant”.

Mixed feelings are not opposition

Sandy Kristin Piderit argued in 2000 that responses to change are better understood as multidimensional attitudes than as a single position between support and resistance, and she put ambivalence at the centre. A person can think a change is sensible, feel anxious about it and still intend to go along with it. Another can like the goal and distrust the method. Calling both “resistant” discards what they are telling you.

A review by Shaul Oreg, Maria Vakola and Achilles Armenakis examined 79 quantitative studies of change recipients’ reactions over six decades. It treats explicit reactions as three-dimensional attitudes and groups their antecedents into conditions before the change, such as recipients’ characteristics and internal context, and features of the change itself: its process, perceived benefit or harm, and content. Reactions are not only a property of the person; they respond to how the change is done and what it costs.

In the hotel, one attendant might welcome the end of crackling radio calls, dislike the idea of her times being visible and doubt the app will work on her floor. That is ambivalence, not opposition, and people voice it more readily where speaking up carries no penalty, the subject of our guide to psychological safety.

A fictional hotel changes how rooms are reported

Consider a fictional city hotel. Each room attendant receives a paper sheet at the morning briefing, ticks rooms as they finish and radios the housekeeping supervisor when a room is ready for inspection. The supervisor radios the front desk. On busy mornings the channel is crowded, and nobody can tell guests waiting for an early check-in when their room will be ready.

The operations manager chose a tablet-based room-status app. Each trolley gets a tablet, and attendants move rooms from “dirty” to “in progress” to “clean”. The supervisor marks rooms “inspected”, and the front desk sees availability as it changes.

Two weeks after launch, the front desk still radios housekeeping. Statuses change in bursts, often long after rooms were finished. When the supervisor raises it at the briefing, an attendant says, “We don’t have time for tablets.” Others nod. The operations manager concludes that the team is resisting. He proposes daily reminders and a back-office screen ranking attendants by how quickly they update their rooms.

He has an interpretation and a remedy, but no information about what is happening on the floors.

Sort what you hear into kinds of concern

Before responding, sort the objection. These seven kinds are an editorial working aid, not a validated typology. They overlap, and one person can hold several, but each points to a different response.

• The change has a flaw. It will not work as intended, or causes a problem elsewhere. Change the plan.
• Loss of something valued, such as discretion, status, income or a respected skill. Negotiate where the loss can be offset; acknowledge it honestly where it cannot.
• Workload and capacity. The new way adds work or arrives with no slack. Provide time or support, or remove the tasks it replaces.
• History and trust. Earlier changes were abandoned or promises broken. Explain honestly; make small commitments and keep them.
• Practical obstacles: equipment, access, physical conditions, handoffs. Fix them with the people who meet them.
• Values and professional identity. The change clashes with how people understand good work. Involve them in design; where the clash remains, decide openly.
• Ambivalence. Mixed feelings, not opposition. Keep the conversation open rather than forcing a side.

In the table below, the “check” column matters most: an interpretation is useful only once the people concerned have confirmed or corrected it.

What you hear or see	Possible concern	How to check it	A fitting response
“This won’t work on our floor.”	A flaw or practical obstacle	Try the task together during a real shift	Change the plan or fix the obstacle
“So now they can see everything we do.”	Loss of discretion; fear of monitoring	Ask what they think the data is for; compare with the plan	Agree and publish limits on data use
“We don’t have time for this.”	Workload, or an obstacle slowing the task	Walk through old and new steps side by side	Remove duplicate work; give time or support
“We did this with the last system too.”	History and trust	Ask what happened and what was promised	Explain honestly; keep small promises
Updates arrive in batches	A workaround, perhaps also self-protection	Ask people to show you their routine	Redesign the step with them
“Guests will notice if we rush.”	Values and professional identity	Ask what good work looks like to them	Involve them; state the standard openly
Agreement in meetings, little change after	Ambivalence, or a concern unsafe to raise	Offer a private route	Keep listening; separate the concerns
“Fine, but I think it’s a mistake.”	Considered disagreement	Ask for their reasoning and the risk they see	Decide openly; record dissent and a review date

When the table points to conditions rather than attitudes, a readiness assessment helps separate support for a change from the means to carry it out.

Listen until the complaint becomes specific

“We don’t have time for tablets” is a summary, not an explanation. The supervisor asks to spend part of a shift on the fourth floor with two attendants. She says she wants to see where the app gets in the way, not to check anyone’s speed, and promises only what she can keep: her notes will not go into anyone’s file.

In our invented version, she begins: “Can you show me what happened with the last room?” One attendant holds up her hands. She wears disposable gloves, and the screen ignores them. To update a room she must remove a glove, tap and put on a fresh pair. “And when my hands are wet from the bathroom, the screen jumps about.”

“What happens after you tap?” “Nothing, near the service stairs. It just spins.” They try it together. On the landing and in the service lift lobby, where trolleys wait, there is no signal.

“Anything else about it that bothers you?” A pause. Then the second attendant asks who can see the times. She has heard about the ranking screen. “If a guest leaves a room in a state, it takes me twice as long. On a screen that just looks slow.”

The apparent reluctance contains three separate things: an equipment problem, a gap in wireless coverage and a well-founded question about data use. Reminders would solve none of them, and the questions that surfaced them were plain. Systemic questions offer more ways to explore a pattern without judging the person; the core discipline is to let each answer change your next question.

Before a conversation like this, agree with the sponsor what is actually open to change. Go where the work happens, say how notes will be used, and do not promise confidentiality or influence you cannot deliver.

What the workaround tells you

The linen room holds the other half of the story. Several attendants have stopped updating room by room. They tape a strip of the old paper sheet to the trolley handle, tick rooms with a pen and enter them in a batch in the linen room, where the signal works. That is why rooms turn “clean” in clusters, long after they were ready.

A workaround is a design review nobody asked for. This one reveals two needs. Updates must work with gloved or wet hands and in dead zones. And batch entry blurs individual timestamps, which may be part of its appeal: it shields attendants from the ranking they fear.

Debono and colleagues’ scoping review of nurses’ workarounds, covering 58 studies in acute care, found that a mismatch between newly introduced technology or policies and existing workflow was one of the most common causes. It describes workarounds as enabling care while potentially compromising it, and reports claims that they hide actual practice and opportunities for improvement. A ward is not a hotel corridor, but the lesson transfers: a workaround solves a local problem while hiding its cause.

So neither ban the workaround before you understand it nor simply bless it: batch updates leave the front desk blind. The workaround tells the manager what to redesign, not whether the attendants are willing. In Kotter’s eight-step change model, this is the barrier-removal step: finding what makes the new way harder than the old.

Choose a response that fits the concern

Kotter and Schlesinger set out six approaches: education and communication, participation and involvement, facilitation and support, negotiation and agreement, manipulation and co-optation, and explicit and implicit coercion. Their core advice is to diagnose the resistance and match the method to the situation. In the hotel:

Change the plan. Attendants will update each room at the door as they leave, when they change gloves anyway. Glove-compatible styluses are trialled on two trolleys. Facilities is asked to extend coverage to the service landings, and the vendor is asked whether updates can be queued offline.

Negotiate. The ranking screen is dropped. The manager publishes what the timestamps are for: room readiness and the inspection queue, not comparisons of individual speed. Where UK data protection law applies, the Information Commissioner’s Office guidance on monitoring workers says employers must tell workers about monitoring and should seek their views when planning it. It also warns against re-using data collected for one purpose for performance management without a new lawful basis.

Provide support and time. Until both fixes work, the paper sheet remains the official fallback. An attendant who understands the app but still struggles gets a separate conversation; ADKAR helps separate a gap in knowledge from one in ability.

Involve people in the design. Two attendants join the vendor configuration session. They ask for a “guest declined service” status and a note field for rooms needing extra time, so a slow room comes with a reason.

Explain honestly. The manager explains why early check-ins matter and what he will and will not look at.

Decide and accept disagreement. The app stays, and paper retires once the fixes are in place. One senior attendant still prefers the radio and says so. Her view is recorded with a review date. Nobody pretends she has been won over.

The fifth approach deserves direct criticism. In Kotter and Schlesinger’s article, co-optation means giving someone a desirable role in a change to secure their endorsement rather than their advice, and the authors warn that people who feel tricked may respond very negatively. The hotel version would be naming the most vocal attendant “tablet champion” to quieten her, with no intention of changing the design. We do not recommend it. It spends the trust the change depends on and turns involvement into theatre. A simple test: would you be comfortable if the person could see exactly how, and why, you were trying to influence them? The same goes for implicit coercion: if a decision is binding, say so plainly.

When the objection is self-interest, or the change must go ahead

Listening is not agreeing. Some objections protect a narrow interest: imagine a supervisor on another shift who dislikes the app because it ends her informal control over who gets the easier floors. That is a real loss, and may still be the right one. Name it and decide.

Some changes are not negotiable: a legal requirement, a safety rule, a contract or a decision the owner has already taken. Say so at the start; presenting a settled decision as open consultation is itself manipulation. What usually remains open is implementation: timing, support, sequence and safeguards.

“Resistance” also blurs three different things:

• Disagreement is saying openly, through normal channels, that the change is wrong. It stays legitimate when the decision goes the other way.
• Obstruction is deliberately undermining an agreed decision, such as entering false statuses after expectations are clear and obstacles fixed. That is a performance or conduct matter.
• Misconduct, such as harassing colleagues who use the app, belongs in the disciplinary process, not a change conversation.

Finally, do not psychologise. “Fear of change” or “negative attitude” is hard to observe and harder to act on. Kotter and Schlesinger list low tolerance for change among their reasons, but even where it fits, the useful response is support and time, not a diagnosis. Describe what people say and do, and ask about it.

Close the loop so people can see what happened

Tell people what changed because of their input, what did not and why. At the next briefing, the supervisor reports that styluses are on trial, landing coverage is booked, the ranking screen will not go ahead and paper retires once both fixes work. A change communication plan with named owners and a feedback route makes that routine rather than a one-off.

If the same objection returns after the practical fixes, the rule behind the change may need examining, not just its execution; our comparison of single- and double-loop learning explains the difference. If “we did this with the last system too” keeps surfacing, look at the unwritten rules earlier changes left behind. And before the hotel extends the app to maintenance requests, a premortem with the engineers is a cheap way to hear objections before launch.

Limitations and evidence

The seven kinds of concern, the table and the hotel are original editorial illustrations, not a validated instrument, and the example reports no real outcomes.

The research behind the reframing is largely conceptual: Dent and Goldberg, Ford and colleagues, and Piderit do not test an intervention. Kotter and Schlesinger’s article is a 1979 practitioner framework, not a controlled study. Oreg and colleagues reviewed quantitative studies of reactions, and a 2022 systematic review by Khaw and colleagues notes that much of this research asks workers to report both their reactions and the causes. Measuring both at once can inflate associations, and the authors call for longitudinal, multi-source designs. They also caution against generalising findings across sectors and countries.

The workaround evidence comes from nursing; applying it to hotels is our interpretation. The ICO guidance is UK-specific, under review following the Data (Use and Access) Act, and not legal advice.

What this approach can do is make a disagreement specific enough to act on. It cannot make everyone agree, and it should not try to.