The standard story about why improvement initiatives fail goes like this. The team designs the change. The team plans the rollout. The team communicates the change. And then people resist. They drag their feet. They quietly work around the new process. They show up to training and don't apply what they learned. The diagnosis: resistance to change.

That diagnosis is usually wrong, or at least incomplete. Most people who appear to be resisting change aren't resisting it. They're at an earlier stage of their own change process than the team rolling out the initiative assumed.

This was the central argument of a KaiNexus webinar with Paola Torres, Senior Performance Improvement Manager at Healthfirst in New York City. Paola is a Lean Six Sigma Master Black Belt with 17 years of experience in healthcare and over 40 cross-functional kaizen events under her belt. She came to motivational interviewing the hard way, after being assigned to work with a department that hadn't requested her help and didn't want it. The methodology she initially dismissed turned out to be the methodology she needed.

The full session is worth watching for the depth, but the core argument is worth understanding even if you never get to the recording.

Change is a process, not a switch

The framework Paola opens with comes from psychologists James Prochaska and Carlo DiClemente, who studied how people actually change health behaviors like smoking and drinking. Their finding: change doesn't happen in a single moment. People move through stages of change.

The five stages, simplified:

1. not yet aware that change is needed (pre-contemplation),

2. aware but ambivalent (contemplation),

3. committed to act and getting ready (preparation),

4. implementing the change (action), and

5. sustaining it over time (maintenance).

Most improvement initiatives expect action from people who are still in pre-contemplation or contemplation. A primary care provider tells a patient to change their diet. The patient says their whole family ate red meat and lived to 90. The provider is treating this as a conversation about action. The patient hasn't even reached contemplation. From the provider's perspective, the patient is resisting. From the patient's perspective, nothing is wrong.

The same dynamic happens in continuous improvement work all the time. A manager creates a new standard operating procedure and complains that staff aren't following it. The manager went through their own change process to identify the need for the standard -- pre-contemplation to contemplation to preparation to action. The staff didn't go through that process. Some of them are still in pre-contemplation. The manager expects action; the staff isn't ready.

The conversations we have with people in different stages need to look different. Telling someone in pre-contemplation what to do doesn't move them forward. It usually moves them backward.

The righting reflex

The second piece of the framework explains why telling people what to do backfires so consistently. People versed in the motivational interviewing approach call it the "righting reflex."

When someone tells us about a problem, our natural response is to want to fix it. We see what they should do. We offer the obvious solution. This impulse is well-intentioned, especially when you're a consultant, a clinical practitioner, or a CI leader. The problem is that it almost always produces the opposite of what we want.

Paola illustrates with a small story. During the pandemic, she wanted to start running in the morning. She mentioned this to a friend, who immediately offered a sensible suggestion: set an alarm for 7:30 that says "run." As her friend talked, Paola found herself getting uncomfortable, then questioning whether she even needed to exercise in the morning. The friend's good-faith advice had activated the other voice in her head -- the voice that said this whole project wasn't necessary.

This is the pattern. People considering change carry two voices simultaneously. The yes voice (change talk in MI terminology) and the no voice (sustain talk). When someone else takes the change side of the argument, the person needing to change takes the no-change side. People are more likely to act on reasons they hear themselves say than on reasons someone else supplies.

This has direct implications for how CI leaders run change conversations. The instinct to articulate the case for change -- to explain why the new process is better, why the metrics matter, why everyone should get on board -- is the instinct that activates resistance. The more articulately the change agent argues for the change, the more clearly the person hears themselves arguing against it.

What motivational interviewing does instead

Motivational interviewing inverts the standard change conversation. Instead of the change agent making the case for change, the change agent's job is to create conditions where the person makes the case for change in their own words. The person doing the talking is the person doing the persuading.

The methodology organizes the conversation into four sequential phases.

1. Engaging (building the relationship).

2. Focusing (narrowing the scope of what the conversation is about).

3. Evoking (drawing out the person's own reasons for change).

4. Planning (translating commitment into action).

Each phase calls for different work. The engaging phase requires patience and reflection -- building enough relationship to support a real conversation, without rushing into solution mode. The focusing phase requires shared agreement on what the conversation is actually about. The evoking phase is the heart of MI -- this is where the change agent uses open-ended questions, affirmations, and reflections to amplify the person's change talk. The planning phase is the familiar action-plan work that's straightforward when the earlier phases have been done well.

The error most improvement teams make: they skip directly to planning. They build the action plan, communicate it, and expect adoption. The conversations that should have happened before the action plan -- the engaging, focusing, and evoking work that produces genuine commitment -- got skipped, and the commitment they would have produced doesn't exist.

Three handles practitioners can use immediately

The full methodology takes practice -- Paola has been working at it for over two years and still meets with a coach. But three specific moves from the session are useful enough to apply right away.

Diagnose where the person actually is. Before you start the change conversation, ask yourself what stage of change the other person is in. If they're in pre-contemplation, your job is to help them recognize the issue, not to convince them to act. If they're in contemplation, your job is to help them resolve ambivalence, not to push them into action. The mismatch between where you assume they are and where they actually are is the source of most failed change conversations.

Use elicit-provide-elicit when sharing expertise. When someone needs your input, the pattern that preserves their autonomy is: elicit permission first ("Would it be okay if I shared what I've seen work?"), provide the perspective, then elicit their response ("What do you think about that approach?"). The structure prevents the common consultant failure mode of offering unsolicited advice, which triggers resistance even when the advice is good.

Watch for the righting reflex in yourself. When someone tells you about a problem, notice the impulse to immediately offer the solution. The impulse comes from a good place. It usually backfires. The MI move is to ask questions that help the person think through their own situation before you start prescribing. Listen for change talk -- the moments when they articulate their own reasons for doing something different -- and reflect those moments back to them.

Why this matters for continuous improvement work

Most CI programs treat change management as something separate from improvement methodology. The improvement methodology is rigorous -- A3 thinking, value stream mapping, PDSA cycles, root cause analysis. Change management is whatever the CI leader figured out through trial and error.

The cost of this gap shows up in sustainment. The improvements are designed and implemented. Then they erode. The team blames a lack of leadership support, competing priorities, or insufficient training. Sometimes those explanations are right. Often they're not. Often, what eroded was the underlying commitment of the people who were supposed to be carrying the change forward. The commitment eroded because it was never really there. The action plan got built without the evoking work that would have produced genuine buy-in.

Motivational interviewing isn't a separate skill set bolted onto CI work. It's the conversation discipline that determines whether the improvements you design get adopted, and whether the adoption holds. The Institute for Healthcare Improvement has named it as a key piece of managing improvement implementation. Paola's case is that more CI practitioners should deliberately develop the skill with the same rigor they bring to their other tools.

The methodology is worth the practice it takes to develop. Most CI leaders have spent years getting good at the technical tools. The conversation skill is where the next round of leverage lives.

The full session covers the methodology in more depth, including three case studies from Paola's own practice that show what MI looks like in real improvement work -- a supply chain engagement that struggled because the leaders hadn't asked for help, an accounts payable team that was ready to engage and needed focusing help, and an ER kaizen that needed evoking work to move from "we would like to do this, but..." to genuine commitment. The session also covers the DARN-CATs framework for amplifying change talk, the OARS interviewing skills, and the specific traps to avoid in each phase.

Watch the recording →