Digital Wayfinders: The Employees Who Already Know How to Make AI Work

The Productivity Gap Nobody Talks About

When Fast Company surveyed 1,000 process decision-makers and 5,000 employees at large organizations earlier this month, the headline figure was striking: 89% of leaders said AI makes their teams more productive. Only 64% of employees agreed. That 25-point gap is not a rounding error. It reflects two genuinely different experiences of what AI is doing to daily work.

The divide runs deeper than perception. According to the same survey, 69% of employees reported that their employer never fully consulted them about how AI would support their day-to-day work. Tools were rolled out, dashboards were updated, AI features were toggled on, often without any conversation about whether the people expected to use them had been involved in choosing them or redesigning the tasks around them.

This is not a minor implementation oversight. McKinsey research cited in the Fast Company piece found that AI high performers, companies where AI is genuinely driving measurable gains, are significantly more likely to have fundamentally redesigned workflows because of AI. They did not simply add a new tool to an existing process. They rebuilt the process. That distinction matters enormously.

The financial reality reinforces this. Many organizations allocate 10% or less of their AI budget to process redesign and organizational change, even when their own leaders acknowledge that transformation efforts will fail without such investment. The tools are getting the money. The humans using them are not.

How Executive Behavior Can Derail AI Transformation

Harvard Business Review‘s recent analysis of executive behavior adds another layer to the problem. The piece identifies two failure modes that frequently stall AI transformations at the top. The first: a senior leader who centralizes all AI decisions, treating the initiative as a personal domain and limiting input from other functions. The second: a leader who disengages from any AI discussion they perceive as outside their area of expertise, leaving critical decisions to drift.

The HBR piece offers a telling vignette. A CIO presents a technically sound AI roadmap to the executive committee. Within minutes, HR raises questions about workforce impact and reskilling that the plan had not addressed. The roadmap stalls, not because the technology was wrong, but because the organizational dimension had been treated as someone else’s problem.

This pattern is common enough to qualify as a structural issue. When AI strategy is designed as a technology project rather than an organizational one, it tends to concentrate decision-making power in IT or in a small leadership group, while the people closest to actual work, those who know which processes are broken and which proposed fixes would make things worse, are left out of the design process entirely.

Who the Digital Wayfinders Are

Harvard Business Review’s October 28 article introduces a concept that reframes this challenge. The authors describe “digital wayfinders” as employees who are already experimenting with emerging technologies on their own initiative, often without any formal mandate. They are not necessarily the most senior people in the room. They tend to be curious and self-directed, motivated by solving real problems rather than following a deployment plan.

What makes this concept immediately useful is its practicality. In most large organizations, these people already exist. They are the analyst who built a workflow using a new AI tool before the company had a policy on it, the operations manager who quietly automated a weekly report, the customer service representative who figured out a prompt that makes the AI actually answer the question being asked. They are not waiting for a roadmap. They are already moving.

HBR’s argument is that leaders systematically overlook these employees when designing transformation strategies. The people given formal authority over AI initiatives tend to be those with seniority or technical credentials, not those with proximity to the work and a track record of figuring things out under pressure. The wayfinders, meanwhile, are treated as interesting exceptions rather than as sources of strategic intelligence.

Empowering them does not require handing over governance or abandoning oversight. It means creating the conditions where their experiments can be recognized, evaluated, and built upon, and accepting that the most valuable AI knowledge in an organization is often distributed, informal, and waiting to be found.

What Ivalua Built by Listening First

The clearest example of what this approach looks like in practice comes from Ivalua, a software company profiled by Fast Company. Ivalua began experimenting with AI years before the technology became a boardroom priority. Rather than launching a top-down transformation program, the company’s founder and CEO took on the role of chief AI officer and set a deliberate tone: AI was going to be a company-wide movement, not an IT project.

The company trained managers to build their own agents. It established a cross-department network of AI champions who could surface ideas from within their teams. Over time, Ivalua ran experiments with 800 agents across every function, gradually revealing which use cases actually worked and where AI was creating confusion rather than value.

Eventually the company moved toward a cleaner model: one intelligent agent governed by the same rules that apply to the people using it. That simplification was only possible because years of bottom-up experimentation had clarified what the organization actually needed, from the inside out.

The result was not a faster rollout of pre-approved tools. It was an organization with a genuinely shared understanding of what AI could do, built by people who had been invited to find out for themselves.

The Real Measure of an AI Transformation

Organizations that measure AI adoption, how many employees have access to a tool, how often it is opened, how many licenses are active, will keep finding that leaders are satisfied and employees are unconvinced. These metrics optimize for deployment. They say nothing about value.

The alternative is to measure what actually changes: cycle time, error rates, decision quality, the share of time employees spend on work they find meaningful rather than work they find repetitive. These indicators are harder to track and slower to move, but they are the ones that reveal whether AI is genuinely improving work or simply adding another layer to it.

Digital wayfinders, if leaders choose to listen to them, are often the fastest route to understanding which of those measures is actually shifting. They are not a substitute for strategy. They are the feedback loop that keeps strategy honest.

For executives serious about closing the gap between what they believe AI is doing and what their teams actually experience, the first move may not be a new platform or a revised roadmap. It may be a conversation with the person two floors down who already figured out the answer, and has been waiting for someone to ask.