More Tech, More Problems: Why Your AI Implementation Needs Better Change Management

Armando Cortez
5 days ago
7 min read

Introduction

The Promise and the Problem

By 2030, Artificial Intelligence (AI) is expected to automate only 15% of healthcare working hours—far less than the hype suggests, according to McKinsey's 2020 analysis of digital automation in healthcare. This figure will change, yet healthcare organizations are rushing to implement AI-powered diagnostics, telemedicine platforms, and intelligent supply chain systems. The technology works. The problem? People don't change as fast as software updates.

In my two+ decades working with healthcare organizations across New England and parts of the US, I've watched this pattern repeat: decent technology, poor adoption, wasted investment.

The Real Challenge

Here's the uncomfortable truth: successful AI implementation isn't just about algorithms—it's also about people. Research from multiple studies, including Harrison's comprehensive 2021 analysis of healthcare transformations, shows that 70% of organizational change initiatives fail, and the primary culprit isn't necessarily bad technology. It's inadequate change management. When healthcare organizations introduce AI without addressing employee resistance, poor communication, and cultural inertia, even the most sophisticated systems may gather digital dust.

I've seen such approaches firsthand when working with Federally Qualified Health Centers (FQHC) and Boston-area hospitals that invested significant amount of dollars on digital tools—only to find end-users reverting to old workflows or adding their own workarounds within months.

The Solution

This article explores two battle-tested change management frameworks—Lewin's Three-Step Model and Kotter's Eight-Step Process—and reveals which approach gives healthcare organizations the best chance of successful AI adoption.

Understanding the Stakes: Why Healthcare Can't Afford to Get This Wrong

The Unique Complexity of Healthcare Transformation

Healthcare isn't like other industries. While tech companies can pivot overnight, health centers and hospitals operate in an environment where:

Regulatory compliance is non-negotiable
Patient safety cannot be compromised during transitions
Staff members range from digital natives to technology skeptics
Every change affects fiscal outcomes for institutions and health outcomes for patients

The Four Forces Driving Change:

Technological advancement - AI powered tools are redefining care delivery
Regulatory evolution - Guidelines will need updates, requiring organizational agility
Shifting patient expectations - Aging and young diverse population demands personalized care
Continuous medical innovation - New research creates pressure to adopt emerging solutions

The Two Frameworks That Actually Work

Lewin's Three-Step Model: Simple, Human-Centered, Proven

How It Works:

Kurt Lewin's 1947 model breaks change into three digestible phases:

Unfreeze - Create awareness that change is necessary. Challenge the status quo and prepare employees psychologically for transition.
Change - Implement the new system with active employee involvement. This is where new tools actually get deployed, training happens, and new workflows emerge.
Refreeze - Stabilize the change so new behaviors become the norm. Prevent backsliding to old habits.

Why Healthcare Loves It:

Lewin's model emphasizes the human element—critical in an industry where compassion and patient interaction define success. It's frequently used in hospitals and clinics because it acknowledges that people need time to process, adapt, and accept new ways of working.

The Limitation:

While elegant in its simplicity, Lewin's model lacks specific guidance for complex, multi-phase transformations. It doesn't address how to maintain momentum through the messy middle of implementation.

Kotter's Eight-Step Process: Comprehensive, Dynamic, Results-Driven

How It Works:

John Kotter's 1995 framework provides granular steps organized into three phases:

Phase 1: Creating a Climate for Change

Create urgency around the initiative
Build a guiding coalition of influential leaders
Form a strategic vision that shows where you're headed

Phase 2: Engaging and Enabling the Organization

4. Communicate the vision relentlessly (through multiple channels)

5. Enable action by removing obstacles

6. Generate short-term wins to build momentum

Phase 3: Implementing and Sustaining Change

7. Sustain acceleration—don't declare victory too early

8. Institute change by embedding new practices into culture

Why It's Superior for AI Implementation:

Kotter's model acknowledges that organizational change is nonlinear and complex. It provides specific actions for overcoming resistance, addresses the critical role of communication at every stage, and recognizes that sustainable change requires embedding new behaviors into organizational DNA.

Research shows that Kotter's and Lewin's models are the most frequently used frameworks in healthcare transformation (Harrison, 2021; Leuba & Piricz, 2024)—but Kotter's eight-step approach offers more tactical guidance for navigating the complexity of AI adoption.

The Verdict: Choose Kotter (But Don't Ignore Lewin)

Why Kotter Wins for AI Implementation

Comprehensive roadmap: Kotter's eight steps provide specific actions at each stage—critical when implementing technology that affects clinical workflows, data management, and patient outcomes. During the height of the Covid-19 Pandemic, I used Kotter’s framework to successfully deploy and adjust supply chain optimization by creating custom labels for storage, achieving cost reductions during vaccination administration.

Built for complexity: Healthcare AI projects involve multiple stakeholders (IT, clinical staff, administrators, compliance officers). Kotter's emphasis on coalition-building and multi-channel communication addresses this reality. As noted in research by Leuba and Piricz (2024), healthcare's inherent complexity demands more sophisticated change frameworks than other industries.

Addresses resistance proactively: By generating short-term wins, Kotter's model helps convert skeptics into advocates before major resistance derails the initiative.

Prevents premature celebration: Healthcare organizations often declare victory after go-live, only to watch adoption rates plummet. Kotter's final steps prevent this failure pattern.

When to Blend Both Approaches

The most sophisticated healthcare organizations combine elements of both:

Use Lewin's framework for the psychological journey—helping staff move from fear to acceptance
Apply Kotter's process for the operational execution—ensuring nothing falls through the cracks

Practical Steps to Prevent AI Implementation Failure

Before You Start

Conduct readiness assessment:

Survey staff attitudes toward AI
Identify potential champions and resisters
Evaluate technical infrastructure gaps

Build your coalition:

Include clinical leaders, IT experts, and frontline staff
Ensure diverse representation across departments
Empower your coalition with decision-making authority

During Implementation

Communicate obsessively:

Share "why" the change matters (connect to patient outcomes)
Use multiple formats—town halls, emails, one-on-ones, visuals and dashboards
Create feedback loops so communication flows both ways

Celebrate early wins:

Identify quick wins within 3-6 months
Publicize success stories widely
Connect wins directly to the strategic vision

Provide continuous training:

Offer multiple learning formats (hands-on, videos, one-on-one coaching)
Recognize that learning is ongoing, not a one-time event
Create "super users" who can support colleagues

After Launch

Monitor and adapt:

Track adoption metrics rigorously
Address resistance patterns immediately
Adjust workflows based on real-world feedback

Embed the change:

Update job descriptions and performance metrics
Incorporate new processes into onboarding
Share success metrics regularly to reinforce progress

Conclusion

AI + Change Management will transform your organization. The technology is only as good as your ability to help people adopt it. While Lewin's model offers elegant simplicity and psychological insight, Kotter's eight-step process provides the comprehensive roadmap healthcare organizations need to navigate the complexity of AI implementation.

Your Action Plan:

Choose Kotter's framework as your primary methodology
Incorporate Lewin's human-centered principles where staff resistance is high
Start with a small coalition of change champions
Communicate relentlessly—more than feels comfortable
Generate visible wins within six months
Measure adoption rates, not just implementation dates

The question isn't whether AI will reshape healthcare—it's whether your organization will successfully make the transition. With the right change management approach, you won't just implement new technology. You'll build a culture capable of continuous adaptation in an industry that never stops evolving.

References

Armenakis, A. A., Harris, S. G., & Mossholder, K. W. (1993). Creating readiness for organizational change. Human Relations, 46(6), 681-703. https://doi.org/10.1177/001872679304600601

Badamo, M. (2024). Why change management is critical for successful AI adoption. Amplience Blog. https://amplience.com/blog/navigating-change-management-successful-ai-adoption/

Bucciarelli, L. (2015). A review of innovation and change management: Stage model and power influences. Universal Journal of Management, 3(1), 36-42. https://doi.org/10.13189/ujm.2015.030106

Caldwell, R. (2012). Systems thinking, organizational change and agency: A practice theory critique of Senge's learning organization. Journal of Change Management, 12(2), 145-164. https://doi.org/10.1080/14697017.2011.647923

Campbell, R. J. (2020). Change management in health care. The Health Care Manager, 39(2), 50-65. https://doi.org/10.1097/HCM.0000000000000290

Creasey, T. (2024). Change vs. change management. Prosci Blog. https://www.prosci.com/blog/change-vs-change-management

Enz, C. (2012). Strategies for the implementation of service innovations. Cornell Hospitality Quarterly, 53(3), 187-195. https://doi.org/10.1177/1938965512448176

Graves, L., Dalgarno, N., Van Hoorn, R., Stokes, M., Pawlowska, L., & Bourgeois, A. (2023). Creating change: Kotter's change management model in action. Canadian Medical Education Journal, 14(3), 136-139. https://doi.org/10.36834/cmej.76680

Harrison, R., Fischer, S., & Walpola, R. (2021). Where do models of change management, improvement and implementation meet? A systematic review of the applications of change management models in healthcare. Journal of Healthcare Leadership, 13, 85-108. https://doi.org/10.2147/JHL.S289176

Khan, A. U., Ali, Y., Pamucar, D., & Vasa, L. (2022). Risk management for cold supply chain: Case of a developing country. Acta Polytechnica Hungarica, 19(8), 161-185. https://doi.org/10.12700/APH.19.8.2022.8.10

Kotter, J. P. (1995). Leading change: Why transformation efforts fail. Harvard Business Review, 73(2), 59-67.

Leuba, V., & Piricz, N. (2024). AI adoption in healthcare: Addressing challenges and change management. In 2024 IEEE 24th International Symposium on Computational Intelligence and Informatics (CINTI) (pp. 283-288). IEEE. https://doi.org/10.1109/CINTI63048.2024.10830852

Lewin, K. (1947). Frontiers in group dynamics: Concept, method and reality in social science. Human Relations, 1(1), 5-41. https://doi.org/10.1177/001872674700100103

Luo, J. S., Hilty, D. M., Worley, L. L., & Yager, J. (2006). Considerations in change management related to technology. Academic Psychiatry, 30(6), 465-469. https://doi.org/10.1176/appi.ap.30.6.465

McKinsey & Company. (2020). The future of work after COVID-19. McKinsey Global Institute.

Oakland, J., & Tanner, S. (2007). Successful change management. Total Quality Management & Business Excellence, 18(1-2), 1-19. https://doi.org/10.1080/14783360601042890

Senge, P. M. (2006). The fifth discipline: The art and practice of the learning organization (Revised ed.). Currency Doubleday.

Sung, W., & Changil, K. (2021). A study on the effect of change management on organizational innovation: Focusing on the mediating effect of members' innovative behavior. Sustainability, 13(4), 2079. https://doi.org/10.3390/su13042079

Syed Talib, H. T. (2018). Kurt Lewin's change model: A critical review of the role of leadership and employee involvement in organizational change. Journal of Innovation & Knowledge, 3(3), 123-127. https://doi.org/10.1016/j.jik.2016.07.002

Tick, A. (2023). Industry 4.0 narratives through the eyes of SMEs in V4 countries, Serbia and Bulgaria. Acta Polytechnica Hungarica, 20(2), 83-104. https://doi.org/10.12700/APH.20.2.2023.2.5

Whittlestone, J., Nyrup, R., Alexandrova, A., Dihal, K., & Cave, S. (2019). Ethical and societal implications of algorithms, data, and artificial intelligence: A roadmap for research. Nuffield Foundation. https://www.nuffieldfoundation.org/sites/default/files/files/Ethical-and-Societal-Implications-of-Data-and-AI-report-Nuffield-Foundat.pdf