The Worker Journey Has the Same Data Richness as the Customer Journey. Organizations Have Built the Infrastructure for One and Almost Nothing for the Other.

Reframe.work founder and workplace humanist Stela Lupushor on why the four W’s framework reframes every workforce technology decision, what managing for endurance actually requires from HR platform operators, and why treating AI adoption as an IT project is the single most expensive mistake organizations are currently making. Consumer-facing organizations have spent decades building the infrastructure…

·

21 min read

Reframe.work founder and workplace humanist Stela Lupushor on why the four W’s framework reframes every workforce technology decision, what managing for endurance actually requires from HR platform operators, and why treating AI adoption as an IT project is the single most expensive mistake organizations are currently making.

Consumer-facing organizations have spent decades building the infrastructure to understand every moment of the customer journey: what attracted the customer, how they moved through discovery and consideration, where friction caused drop-off, what kept them engaged, and what the relationship looks like after the transaction. They have analytics dashboards, journey maps, signal collection at every touchpoint, and continuous optimization cycles that connect behavioral data to business outcomes with a precision that HR leaders can only observe from a distance and wish existed for their own workforce. The irony is that the worker journey generates at least as much behavioral data as the customer journey, and the outcomes it determines, retention, performance, engagement, productivity, are at least as important to business results. The infrastructure to capture and act on that data has simply never been built with the same rigor.

The workforce technology market has produced a set of tools that address the administrative transactions of employment: the ATS for applications, the HRIS for records, the payroll system for compensation, the performance management platform for annual reviews. What it has not produced is the continuous signal collection architecture that connects every touchpoint in the worker journey, from the moment someone first encounters the employer brand through the full arc of their tenure and even their post-departure relationship, to the behavioral intelligence that would allow organizations to improve that journey with the same systematic discipline they apply to their customer experience. Every moment in the worker journey is a data opportunity. Most organizations have built the infrastructure to capture almost none of it.

Stela Lupushor has spent her career at precisely this intersection: the technical discipline required to build measurement systems that connect human behavior to business outcomes, and the humanist conviction that every data point in those systems represents a person whose experience deserves to be designed with the same care that consumer brands apply to their best customers. As the founder of reframe.work and co-author of Humanizing Human Capital and Humans at Work, she brings a framework for workforce intelligence that predates the current AI conversation and is now more urgently relevant than ever. The five arguments that follow are what HR tech platform operators and employers need to hear about the gap between the workforce data they have and the workforce intelligence they could build.

About the Guest

Stela Lupushor is the founder of reframe.work, a consultancy that brings together human-centered design, workforce technology, and analytics to improve the worker experience at scale. She is co-author of Humanizing Human Capital and Humans at Work, serves on the faculty at NYU, and advises enterprises, startups, and governments on human-centered work transitions. Her career spans workforce transformation roles at Fidelity Investments, TIAA, IBM, and PwC, giving her a practitioner’s grounding in both the technical infrastructure of workforce systems and the human dimensions of what those systems are supposed to support.

The Four W’s Reframe Every Workforce Technology Decision That Organizations Are Currently Getting Wrong.

The most practically useful diagnostic framework in the conversation is the one Stela introduced as the four W’s, and it is worth examining in detail because it exposes the structural misalignment at the center of most workforce technology investments. The first W is work: not the jobs an organization is hiring for, but the actual essence of what needs to get done. Jobs, in Stela’s framing, are containers designed to filter out unqualified candidates. They are administrative artifacts, not descriptions of organizational need. An organization that starts its workforce planning with job descriptions is starting in the wrong place, because job descriptions describe how the work was done last time rather than what the work actually requires and who or what is best positioned to deliver it.

The second W is worker, and the argument here is the one that most disrupts current assumptions about workforce management. The optimal worker for any given piece of work is not a binary choice between an employee and a contractor. It is a spectrum that includes teams, alliances, partnerships, digital labor, and automation, and the right answer depends on the nature of the work and the conditions in which it needs to be performed. Most workforce technology platforms were built on the assumption that workers are employees, and that assumption is embedded in the architecture of every HRIS, performance management tool, and compensation system in the market. As the workforce spectrum expands to include significant digital labor components, those architectures will require either fundamental redesign or the kind of overlay infrastructure that connects the full workforce spectrum to a unified data environment.

The third W is workplace, the context and environment constructed for the match between work and worker to be effective. The pandemic forced every organization to confront the fact that their assumptions about what workplace meant were not design choices but defaults inherited from a model that no one had explicitly chosen. The fourth W is worth: the value exchange, which Stela identifies as the most dynamic of the four because what people come to work for changes across life stages, career stages, and environmental conditions. Stability matters differently to a 28-year-old than to a 52-year-old; the value of professional development looks different in an expanding economy than in a contracting one. The organizations that are designing their total rewards and culture programs without a real-time read on what their workforce currently values are designing against a static model of worker motivation that may not describe anyone actually on their team.

“What is the work that needs to get done, not the jobs you’re hiring for. Those are just containers to filter out unqualified candidates. What is the essence of what needs to get done? Who is the optimal worker to do that work? That’s all part of your workforce spectrum.”

The behavioral intelligence connection to the four W’s framework is in where the signal lives that would allow organizations to make better decisions across all four dimensions. The work dimension requires data about what tasks are actually being performed, not just what the job description says. The worker dimension requires continuous behavioral signal about how different people are performing across different work types and conditions. The workplace dimension requires signal about how environmental conditions are affecting behavioral patterns. The worth dimension requires continuous signal about what employees are currently valuing and how that is shifting. All of this signal runs through the post-hire period, in the behavioral patterns that transactional platforms were never designed to capture, and it is precisely what MustardHub’s behavioral intelligence infrastructure is built to surface and route to the decision-makers who need it.

For platform operators, the four W’s framework creates a specific product audit opportunity. Evaluate your platform against each dimension: does it help organizations understand the actual work being done, not just the jobs being filled; does it capture behavioral data across the full workforce spectrum including contingent workers and digital labor; does it surface the environmental and contextual signals that determine whether the workplace is enabling or inhibiting performance; and does it provide a real-time read on what specific employee populations currently value? Most platforms address parts of one or two of these dimensions. The infrastructure investment that connects all four is what the market is building toward, whether it has named the destination yet or not.

The Worker Journey Is as Data-Rich as the Customer Journey. Organizations Have Simply Never Built the Infrastructure to Capture It.

Stela’s consumer journey analogy is one of the clearest arguments available for why the workforce technology market has consistently underbuilt the infrastructure for behavioral signal collection across the employment lifecycle. A consumer-facing brand tracks every touchpoint from first brand awareness through purchase decision through loyalty and advocacy, building a behavioral profile that becomes more accurate and actionable with every interaction. The signal that worker generates across an equivalent journey arc, from employer brand discovery through candidacy, onboarding, development, performance, and post-departure relationship, is at least as rich and at least as consequential to business outcomes as the consumer signal. The tools to capture and act on it have simply not been built with the same systematic investment.

The specific touchpoints Stela mapped through the worker journey, the hiring experience, the onboarding connection to team and manager, the timing and availability of developmental opportunities, the nature of the departure experience, and the post-departure relationship, are each moments where behavioral data would allow an organization to identify friction, improve the experience, and connect the improvement to measurable business outcomes. The hiring experience that leaves candidates feeling ghosted is losing talent that the organization spent real money to attract. The onboarding experience that fails to connect new hires to their team and manager is generating the attrition risk in the first ninety days that produces the most expensive turnover in the employment lifecycle. The departure experience that burns bridges rather than building alumni relationships is destroying a talent pipeline and a brand advocacy network simultaneously.

The measurement infrastructure for the consumer journey was built because consumer brands understood the financial return on understanding and improving every touchpoint. The measurement infrastructure for the worker journey has not been built at equivalent scale because the financial return on understanding and improving every worker touchpoint has not been modeled with equivalent rigor. Stela’s observation that you can create a very direct value chain between the decisions made at each touchpoint in the worker journey and the business outcome those decisions affect, using the same analytical discipline that consumer brands apply to their customer experience, is the argument that translates workforce intelligence from a human resources concern into a CFO conversation.

“I’m trying to bring the same mindset to the world of HR, to think of the worker in the same way. All of that creates lots of interaction and opportunity for data to capture. And then you can think through ways of improving that experience, reducing the friction for the worker, and then trying to create those interventions from HR perspective that will have an impact on the business outcome.”

The post-hire blind spot that MustardHub addresses is the specific segment of the worker journey that generates the highest concentration of behavioral signal with the lowest current rate of capture: the period between hire and the next recorded HR event, where the onboarding experience, the initial team and manager relationship, the early development trajectory, and the first indicators of role fit and engagement are all running continuously without any infrastructure to surface them to the people who could act on them. Building the signal collection infrastructure for this segment of the worker journey is not a different project from building the consumer journey infrastructure that Stela describes. It is the same project applied to the workforce rather than the customer base, using the same analytical discipline and producing the same quality of actionable intelligence.

For platform operators, the consumer journey parallel creates a specific sales conversation opportunity. Every organization that has invested seriously in customer experience measurement infrastructure understands the financial return on that investment. The argument that an equivalent investment in worker journey measurement infrastructure would produce equivalent returns, because the decisions made at each worker touchpoint have direct and measurable connections to the revenue and productivity outcomes the organization cares about, is the argument that moves workforce intelligence from the HR budget to the business case conversation. Building your product narrative around the worker journey as a data asset with quantifiable business value, rather than as an HR program with soft ROI, will reach the buyers who have already learned to think in those terms from their customer experience investments.

How does your workforce measure up against operators like you?

Live insights on your workforce in two minutes.

Estimate my turnover cost

Managing for Endurance Is Not Resilience with a Different Name. It Is a Fundamentally Different Organizational Capability.

The distinction Stela drew between resilience and endurance is one of the most practically important arguments in the conversation for organizational leaders trying to understand what the current environment actually requires. Resilience, in its conventional HR application, is the capacity to absorb a discrete stressor and return to baseline: a difficult quarter, a reorganization, a market disruption that produces a temporary period of elevated pressure. The intervention model for resilience is point-in-time: a wellness program, a leadership training, a period of intensive support followed by a return to normal operating conditions. The assumption embedded in the resilience model is that there is a baseline to return to, and that the stressor is temporary.

The environment Stela describes, relentless volatility, geopolitical instability, accelerating technological change, and the absence of any shared reality against which organizational decisions can be calibrated, is not a resilience problem. It is an endurance problem: the capacity to sustain high-quality performance and judgment in an environment where the pressure is permanent, the uncertainty does not resolve, and the return to baseline is not a realistic planning scenario. Building endurance requires different interventions than building resilience: not point-in-time support programs but the continuous development of the exposure, the skill sets, and the psychological frameworks required to function effectively in conditions of sustained uncertainty. The organizations that are treating this as a resilience problem will consistently under-invest in the endurance infrastructure that the current environment actually requires.

The leadership behaviors Stela identified as most effective during the pandemic, the closest recent analog to the kind of sustained uncertainty the current environment requires, are instructive precisely because they represent a departure from the conventional leadership model. Showing up vulnerably. Making it okay to not be okay. Sharing the decisions being made with the information available rather than projecting false certainty. Creating space for course correction rather than penalizing the course corrections required by honest engagement with a rapidly changing environment. These are not soft leadership behaviors in the pejorative sense. They are the specific organizational conditions that allow a workforce to sustain honest behavioral engagement with their work and their environment rather than retreating into the self-protective patterns that produce the managed performance that makes behavioral signal collection unreliable.

“When there’s stress and changes and volatility, it’s relentless. It’s day in and day out. You need to strap on for the long haul. You need to build resilience. You need to give exposure to team members in a safe way. You need to have empathy when things fail and fall apart. You need to celebrate failures because that’s the only way they’ll learn to take risks again.”

The behavioral intelligence connection is in what endurance-oriented leadership produces as a data environment. A workforce that has been given the psychological safety to engage honestly with uncertainty, to report difficulties without penalty, and to course-correct publicly without career consequence, generates behavioral patterns that accurately reflect organizational conditions. The stability pattern data that MustardHub surfaces through continuous signal collection is most accurate and most actionable in organizations where the endurance infrastructure has been built: where employees are not managing their visibility out of self-protection, and where the behavioral signal running through the post-hire period is an authentic representation of how people are actually experiencing their work rather than a performance for the benefit of the measuring system.

For platform operators, the endurance argument creates a specific product design requirement: the organizational capability to build endurance in a workforce requires real-time visibility into the behavioral conditions that indicate whether the workforce is sustaining authentic engagement or retreating into managed performance. An organization that is building endurance infrastructure needs to know which teams and populations are showing the behavioral patterns of sustained engagement and which are showing the early indicators of burnout or disengagement, at a cadence and granularity that annual surveys and quarterly reviews cannot provide. Building the continuous signal collection architecture that gives leadership this visibility is the platform investment that makes endurance-oriented leadership actionable rather than aspirational.

AI Adoption Treated as an IT Project Will Produce the Surgery Is Successful, Patient Dead Outcome.

The German medical analogy Stela quoted is the most precise description available of the failure mode that most organizations are currently building toward in their AI adoption programs. Surgery is successful, patient dead: every technical objective has been met, every efficiency metric has been hit, the deployment is on schedule and on budget, and the organizational system that the surgery was supposed to improve is in worse condition than before the procedure began. The instrument left inside. The artery cut in the process. The infection introduced because the post-operative recovery plan was not included in the original surgical planning. The long tail of the transition that was not embedded in the business case because the business case was built by the people who were good at the surgery and had no model for what the recovery would require.

The specific organizational conditions that produce this outcome are the ones Stela identified as characteristic of AI adoption treated as an IT project: a cross-functional governance committee setting rules for what can and cannot be deployed, a project plan with milestones and delivery dates, a leaderboard tracking token utilization as a proxy for adoption success, and a set of marching orders that optimize for the deployment metric without any model for whether the deployment is producing the organizational capability it was supposed to produce. The organizations that are measuring AI adoption by token utilization are producing the same dynamic that Stela identified as the universal principle of organizational behavior: if you put a system in place that rewards certain behaviors, people will game it. Token utilization is a behavior that can be gamed. Organizational capability is not.

The alternative Stela proposes is to treat AI as the general purpose technology that it is, comparable in its foundational impact to the mechanical loom or electricity, and to design the organizational transition with the same deliberate long-term planning that the introduction of a genuinely foundational technology requires. The mechanical loom did not just make weaving faster. It restructured the entire textile industry, created new roles, eliminated others, changed the geographic distribution of production, and produced new economic relationships that took decades to fully work out. An organization planning its AI adoption as a twelve-month technology deployment is not planning for the equivalent of a faster loom. It is planning for the equivalent of not having thought through what electricity meant for the factory floor until it was already running.

“You need to look at AI as a similar type of general purpose technology that creates space to reinvent a lot of things. That transition to the future requires a lot of deliberate planning, not only at the organizational level, how you help people navigate through this transition and adopt it and change the work, but also at the governmental level.”

The workforce behavioral intelligence connection is in what the foundational technology transition requires as an organizational sensing capability. An organization navigating a genuinely foundational technology shift needs continuous visibility into how the workforce is actually experiencing the transition, where the behavioral patterns of adaptation and capability building are developing and where the patterns of resistance, anxiety, and disengagement are accumulating. That visibility cannot come from an annual engagement survey or a quarterly performance review. It requires the continuous signal collection that MustardHub’s infrastructure layer provides through the post-hire period, connected to the organizational conditions generating the behavioral patterns it surfaces. The organizations navigating this transition well are not the ones with the most sophisticated AI tools. They are the ones with the most accurate real-time read on how their workforce is experiencing the change.

For platform operators, the general purpose technology framing creates a specific product durability argument. The platforms that are building the behavioral intelligence infrastructure required to navigate a foundational technology transition are not building a feature for the current AI adoption cycle. They are building the organizational sensing capability that will be required for every major technology transition that follows, because the fundamental challenge of helping a workforce adapt to a genuinely new technological paradigm will recur regardless of which specific technology is producing the disruption. The infrastructure investment in continuous behavioral signal collection compounds in value across every technology transition rather than depreciating with the current cycle.

Human-Centered Design Is Not a Soft Alternative to Engineering. It Is What Happens When the Systems That Affect 100 Percent of the Workforce Are Designed by the 100 Percent, Not the 80.

Stela’s argument about the shift from design by engineers to design by designers is not an aesthetic preference. It is a functional critique of what happens when the systems that determine how work gets done, how people are hired, how performance is evaluated, how benefits are accessed, and how the full worker journey is experienced, are built by people who are optimizing for the conditions that work for the majority while treating the conditions that work for everyone as edge cases. An engineer designing a hiring system assumes that the applicant has full vision, reliable internet access, the ability to navigate a multi-step asynchronous process without accommodations, and the kind of credentials that the system was built to recognize. The system works for the 80 percent of the population that shares those assumptions. The 20 percent for whom the assumptions do not hold are not edge cases. They are people.

The specific example Stela offered, the 75 percent unemployment rate among people with disabilities, is both a moral indictment and a practical one. The talent that exists in the population of people who have spent their lives navigating environments that were not designed for them, developing cognitive flexibility, adaptive problem-solving, and resilience capabilities that the standard hiring process was never designed to recognize, is a talent resource that the credential-matching hiring architecture is systematically excluding. The organizations that have built the accessible, inclusive hiring and work experience infrastructure that allows this population to participate are not just doing the right thing. They are accessing a capability pipeline that their competitors have written off as an edge case.

The broader design principle Stela is making is that every system that affects the worker journey should be designed with the same discipline that consumer experience design applies to the customer journey: starting with a deep understanding of the full range of people who will interact with the system, mapping the exceptions and edge cases as explicitly as the standard flow, and designing for the full population rather than the optimal case. This is not a more expensive or more complicated design process. It is a more rigorous one, and the organizations that apply it to their workforce experience infrastructure will consistently produce better outcomes than those building for the 80 percent and hoping the remaining 20 percent will find a workaround.

“Human-centered design brings a different lens. You look at the different personas that are engaging with whatever you’re building. You look at what are the exceptions and the edge cases that you need to account for. How might you come up with solutions that include everybody, not just what’s optimal and the cheapest way to do it.”

The behavioral intelligence connection is in what human-centered system design produces as a data environment. A hiring process designed for the full population of potential candidates generates a candidate pool that includes the people the credential-matching process would have excluded. An onboarding process designed for the full range of learning styles, accessibility needs, and social connection preferences generates a post-hire behavioral pattern in which the full diversity of the workforce is genuinely included in the organizational experience rather than managing their visibility out of a system that was not designed for them. The stability pattern data that MustardHub surfaces through the post-hire period is most valuable in organizations that have built the inclusive infrastructure that allows all members of the workforce to show up authentically, because the behavioral signal from those organizations reflects actual organizational conditions rather than the conditions of the 80 percent for whom the system was designed.

For platform operators, the human-centered design argument creates a specific product quality standard: evaluate every feature of your platform against whether it was designed for the full range of workers who will use it or for the 80 percent case. A performance management tool that assumes all employees have the same relationship to feedback, goal-setting, and recognition is not a universal tool. It is a tool that works well for the majority and produces inaccurate data for everyone else. Building the personalization and adaptation capability that Stela identifies as the design standard for truly human-centered systems, connecting it to the behavioral signal data that reveals which workers are being well-served by the current design and which are not, is the product investment that closes the gap between the workforce intelligence the platform claims to provide and the workforce intelligence it is actually capable of delivering.

What HR Tech Operators Should Do With This

1. Audit Your Platform Against All Four W’s, Not Just the Worker Dimension

Most workforce technology platforms were built around the worker dimension of the four W’s and treat the work, workplace, and worth dimensions as secondary considerations or separate products. Audit your platform against all four: does it help organizations understand what work actually needs to get done beyond the job descriptions, does it capture behavioral signal across the full workforce spectrum, does it surface environmental and contextual signals that affect performance, and does it provide a real-time read on what specific populations currently value. MustardHub’s continuous behavioral signal collection operates across all four dimensions of the worker journey, providing the infrastructure that connects work, worker, workplace, and worth into a unified intelligence architecture.

2. Build the Worker Journey Data Architecture With the Same Rigor as the Customer Journey

Every touchpoint in the worker journey from employer brand discovery through post-departure relationship generates behavioral data that is as consequential to business outcomes as the consumer data organizations invest heavily in capturing. Map the worker journey touchpoints that your platform supports and identify the signal collection gaps between recorded events, particularly in the post-hire period where the most expensive attrition risk develops without any transactional trigger. MustardHub’s post-hire blind spot infrastructure closes the most significant gap in the worker journey data architecture, providing continuous signal from the period that determines whether the hire investment produces a return.

3. Design Your Endurance Infrastructure Before the Volatility Requires It

The organizational endurance capability that Stela describes as the requirement for sustained performance in permanent volatility needs to be built before the conditions that require it arrive at maximum intensity. Design your platform’s organizational health architecture to surface the behavioral leading indicators of burnout and disengagement, specifically the early stability pattern shifts that precede the performance decline and attrition that make the endurance deficit visible too late. MustardHub’s stability pattern data provides the continuous signal that makes endurance monitoring possible before the point of crisis rather than as a response to it.

4. Position AI Adoption Support as a Long-Cycle Infrastructure Investment, Not a Deployment Project

The organizations treating AI adoption as a twelve-month technology deployment will produce the surgery is successful, patient dead outcome because they have not built the organizational sensing capability to monitor the transition’s behavioral consequences continuously. Build your product narrative around the long-cycle organizational intelligence infrastructure that a foundational technology transition requires, specifically the continuous behavioral signal collection that tells leadership how the workforce is actually experiencing the change. MustardHub’s infrastructure layer embeds within the platforms operators already use, providing the real-time workforce sensing capability that makes a long-cycle technology transition navigable rather than reactive.

5. Evaluate Every Platform Feature for Whether It Was Designed for the Full Workforce or the Eighty Percent

A performance management tool, a feedback mechanism, or a benefits communication architecture that works well for the majority and produces inaccurate or inaccessible data for the remaining population is not a universal tool; it is a tool with a specific population embedded in its design assumptions. Audit every feature against the human-centered design standard Stela describes: who is this designed for, who does it exclude by default, and what adaptation or personalization capability would extend it to the full population. Connect the behavioral signal data from MustardHub’s post-hire infrastructure to your platform’s design review process to identify which worker populations are generating behavioral patterns that indicate the system is not working for them, and use those patterns to drive the inclusive design improvements that close the gap.


Stela Lupushor is the founder of reframe.work and co-author of Humanizing Human Capital and Humans at Work. Follow her monthly research and commentary at humanizinghumancapital.com and connect with her work at reframe.work.

If this conversation surfaced questions about the behavioral signal your platform is currently missing, MustardHub provides the infrastructure layer that embeds behavioral workforce intelligence inside the platforms HR tech operators already run. Visit mustardhub.com to learn how we surface attrition risk, financial impact, and stability patterns from the post-hire blind spot no transactional system was designed to measure.

Get customer-journey-grade visibility into your workforce.

Employers save $3,000+ per employee by catching preventable exits early.

Estimate my turnover cost

About the author

Estimate your turnover cost

Live insights on your workforce in two minutes. 5 questions, no credit card.

See my $ risk
Stay in the signal.
Original research and fireside chats on behavioral workforce intelligence, in your inbox.
Subscribe