When Change Begins in Mischief: Trojan Mice, the Horse & Organizational Shift

Beware Trojan Mice (And Set Them Free Anyway)

I live on a farm. Which means I know something about mice. They are small. They are quick. They are… everywhere.

No matter what you do, mice will find their way into places you didn’t expect. They can be destructive. They can proliferate. They can feel unstoppable.

They’re a nuisance on the farm, but in nature, mice are indispensable. They’re food for countless predators, spread seeds, and even aerate the soil. Take mice out of the ecosystem, and everything else starts to unravel.

This makes them a brilliant metaphor for change in complex organizations.

The Metaphor

A recent LinkedIn post by Joss Colchester reminded me of the Trojan Mice metaphor. His graphic brilliantly captures the core idea: instead of relying on one big “Trojan Horse” change program, we allow for many “Trojan Mice.” Trojan Mice are low-risk experiments carried out by many people, essentially scurrying about, testing the terrain, finding solutions, and adjusting. Over time, those Trojan Mice have organizational-wide impact.

As I understand it, Peter Fryer originally coined the phrase Trojan Mice at (archived) trojanmice.com. It was picked up and popularized by Euan Semple in his book Organizations Don’t Tweet, People Do (2012), and later connected to complexity science by Harold Jarche in his blog post On Trojan Mice (2012). Each emphasized the same point: small, well-focused, safe-to-fail experiments are more likely to result in widely accepted change than grand, sweeping plans ever will.That Trojan Mice way of generating large transformations is analogous to what we see in complex systems. That’s why the metaphor resonates so strongly with those working in complexity, whether through Cynefin’sProbe–Sense–Respond pattern, Complexity Adaptive Systems, or the messy realities of healthcare organizations.

A Real-World Example: VA’s EHR Story

The Department of Veterans Affairs vividly illustrates Trojan Mice versus Trojan Horses.

• Trojan Mice (CPRS/VistA): The VA’s Computerized Patient Record System (CPRS), built on the decades-old VistA platform, emerged through a “Trojan Mice” approach. Instead of a single master plan, it grew from numerous local experiments by clinicians and programmers who solved immediate problems and shared their open-source code (A 40-year ‘conspiracy’ at the VA). These grassroots innovations often advanced without formal central approval, as local staff sidestepped slow bureaucracy to create systems that worked for them. Although the first graphical interface for CPRS launched in 1997, the bottom-up development of VistA had begun decades earlier. By the mid-2000s, CPRS was in nationwide use, resulting from long-term decentralized innovation with development costs absorbed into local facilities’ operating budgets..

• Trojan Horse (Cerner Millennium/Oracle): In contrast, the VA is now attempting to replace CPRS with a vendor-based system (Cerner Millennium, now Oracle Cerner). This initiative reflects the opposite approach: a centrally directed, multi-billion-dollar program intended to modernize the EHR in a single stroke. Significant cost overruns have marked the project—estimates rising well beyond the initial $10 billion budget—as well as repeated delays and site pauses. Early rollouts have triggered reports of workflow disruptions, safety concerns, and clinician dissatisfaction, with many users describing the system as less intuitive and less responsive to their needs (Veterans Affairs’ Ongoing Struggle to Modernize Its Electronic Health Record System). Unlike the bottom-up evolution of CPRS, Cerner represents a top-down imposition whose challenges underscore the risks of trying to engineer transformation through a single “Trojan Horse.”

The Catch

As fun and catchy as “Trojan Mice” is, the metaphor has a habit of getting co-opted by the very mindset it was meant to challenge.

Take CPRS itself. It started as a swarm of “mice.” That is, many local innovations spreading and stabilizing into a national system. But over time, leaders in the VA Central Office (VACO) attempted to control its direction. Efforts to centralize CPRS updates and upgrades led to cries to standardize CPRS across the country, as those actions often broke local systems. Specifically, because the original mice hadn’t been deployed en bloc, the system emerged locally. The more leaders at VACO tried to manage it like a single Trojan Horse, the more brittle things became.

That brittleness eventually became one of the justifications for abandoning CPRS in favor of Cerner. The irony? A system built by mice, loved by clinicians, was displaced by a horse because its very “mouse-ness” didn’t fit the dominant discourse of control and uniformity.

This is the deeper risk with metaphors like Trojan Mice:

• Leaders start talking about “deploying” or “rolling out” mice.
• Small, emergent acts get reframed as centrally managed tools.
• And before long, the mice aren’t mice anymore. Now they’re being harnessed to pull a horse.

Let the Mischieve Be

So here’s the invitation: Enjoy the metaphor. Laugh with it. Picture those scrappy little rodents running around your organization, nibbling on old habits and gnawing through stale routines. Let the image remind you that small, local, experimental acts matter.

But hold it lightly. Don’t reify (make it a real thing) the Trojan mice into a management tool. Don’t let it become a line item in your change-management plan.

Because the real “Trojan Mice” are nothing more (and nothing less) than our everyday acts of relating to one another as we try something new in a conversation, test a fresh idea in a meeting, and tinker with tools and resources. These small acts, choices, and behavior changes occurring naturally and outside central control are metaphorical Trojan mice.

As a fun aside: a large group of mice is a mischief? A mischief of mice is an ideal metaphor. Because, like real mice, once small acts of change start moving, they rarely stay tidy or predictable, and you can seldom eliminate them entirely

So let them run. Notice the mischief. Cultivate the most useful ones. Just don’t start thinking you’re in charge of the “mischief.”

Set your mice free — but don’t start thinking you’re in charge of the mischief.