Employee experience has become the most expensive search term in human resources software. The keyword "employee experience" carries an average cost-per-click around $14.64, and competitive enterprise terms in the category hit max bids over $200. That kind of money does not chase a soft concept. It chases a market that buyers think will solve a real problem.
The problem is real. Gallup's 2025 State of the Global Workplace report found global engagement dropped to 21 percent, matching pandemic lows, and the financial cost of disengagement reached $8.9 trillion globally, equivalent to 9 percent of global GDP (Gallup, 2025). The employee experience management software market hit approximately $7.1 billion in 2026 and is projected to grow to $12.9 billion by 2035 (Global Market Insights, 2025).
So companies are spending money. The question is whether what they are buying is solving the actual problem.
The Textbook Definition
Two frameworks dominate the academic and consulting literature on employee experience. Both are useful starting points. Both also share a blind spot.
Josh Bersin: Six Elements of EX
Josh Bersin's research, based on a survey of 981 companies, defines employee experience through six elements: meaningful work, trust in leadership, a positive work environment, growth opportunity, health and well-being, and a sense of inclusion and belonging (Bersin, 2021). The factors with disproportionate impact on outcomes were trust in the organization, a mission and purpose that inspires people, leaders who communicate transparently, and a sense of belonging.
Bersin frames EX as a holistic system that includes 15 specific HR practices and a technology architecture for measuring and improving each element. The framework is comprehensive. Its strength is breadth. Its weakness is that "trust in leadership" and "sense of belonging" are not single decisions you can implement. They are emergent properties of thousands of small interactions.
Jacob Morgan: Three Environments
Jacob Morgan's research, published in The Employee Experience Advantage, analyzed over 250 organizations and identified three environments that shape employee experience: cultural, technological, and physical. Morgan's data assigned weights: cultural environment contributes 40 percent, technological 30 percent, physical 30 percent (Morgan, 2017).
Morgan also identified 17 specific attributes spread across the three environments. Companies that scored highest on his Employee Experience Index outperformed S&P 500 averages on profit, revenue, and stock performance. The cultural environment did the most work in explaining the gap between great EX companies and average ones.
The MIT Number
MIT CISR ran a complementary study and found that organizations in the top quartile of employee experience had net profit margins 8.4 percentage points higher than the bottom 75 percent, and revenue from new products and services more than double the bottom 75 percent (MIT CISR, 2017). EX is real. EX is measurable. EX produces hard financial outcomes when it works.
So the textbook answer is settled. Employee experience is the sum of every interaction across the lifecycle, weighted toward cultural environment, driven by trust and belonging, measurable through a defined set of attributes, and tied to real business outcomes. That is the part everyone agrees on.
The Layer Both Frameworks Underweight
Here is where the textbooks get quiet. Both Bersin and Morgan describe EX as a system. Both identify culture as the dominant variable. Neither framework specifies the unit of analysis where culture is actually produced.
Gallup answered that question with one number. 70 percent of the variance in team engagement comes from the manager (Gallup, 2025). That single finding has been replicated for two decades. It says the macro frameworks miss the actual point of leverage. Culture does not happen at the company level. It happens at the team level, and it is mostly determined by how one specific person, the direct manager, runs the daily work.
Manager engagement itself dropped from 30 percent to 27 percent between 2023 and 2024, the steepest decline of any worker group. Less than half (44 percent) of managers globally have received any management training (UNLEASH, 2025). The people who produce 70 percent of the EX outcome have themselves been deprioritized and undertrained.
This is the gap. The frameworks describe the cathedral. The actual stained glass is being installed by undertrained team leads in cross-functional meetings, hallway conversations, retro decisions, and stand-ups that nobody is measuring.
What Employee Experience Actually Is, Hour by Hour
If you stop describing EX as a strategy and describe it as a sequence of moments, the picture changes.
Your employee experience this Tuesday is whether your one-on-one happened or got moved for the third week in a row. Whether the engineer you handed work off to actually picked it up, or it sat in review for four days. Whether your team lead defended your scope in the planning meeting or threw you under the bus to keep the deadline. Whether the disagreement in your design review stayed productive or curdled into a passive-aggressive Slack thread three hours later. Whether the new hire on your pod feels safe enough to ask the dumb question.
None of those moments live inside an annual survey. None of them show up in a perks budget or a recognition platform. They are the substrate of employee experience, and they happen in the gaps between the things HR measures.
The Achievers research that gets cited everywhere makes this point sideways. Employees who receive meaningful recognition weekly are 9 times more likely to feel a strong sense of belonging and 2.6 times more likely to be highly productive (Achievers, 2026). Note what produces belonging. Recognition. From a specific person. About a specific behavior. Delivered at the right moment in the workflow. The platform is the delivery mechanism. The interaction is the ingredient.
The False Promise of "How Are You Feeling?"
The dominant EX measurement tools all start from the same primitive: ask employees how they feel and aggregate the answers. Pulse surveys, eNPS, sentiment dashboards, the entire listening category. The category is useful, and you should run it. It is also incomplete in a way the category vendors do not advertise.
Here is the problem. When you ask "how was this week?" you are asking about the output, not the input. The output is a feeling. The feeling depends on the inputs: whether decisions got made, whether handoffs were clean, whether the team backed each other up, whether the manager listened. None of those inputs are visible in the survey response. You see the score. You do not see the cause.
This is why companies that have run engagement surveys for ten years still get the same scores year over year. They are measuring the temperature without measuring the room. The room keeps producing the same temperature because nothing in the underlying interaction has changed.
Measuring engagement properly requires layering sentiment data on top of behavioral data from the actual interactions. Otherwise you are running diagnostics on the wrong system.
What "Improving Employee Experience" Actually Looks Like
If 70 percent of the variance comes from the manager and most of that comes from how the manager runs daily team interaction, then "improve employee experience" decomposes into a small set of specific behaviors:
- Decisions get made. Not deferred, not buried in a doc, not handed up the chain. The team has clear decision rights and uses them.
- Disagreement stays productive. Conflict surfaces in the open, gets resolved on the merits, and does not metastasize into politics.
- Handoffs are clean. Information transfers without loss. The next person knows what they are picking up and why.
- People back each other up. When something goes wrong, the team owns it together. When something goes right, credit gets distributed accurately.
- Recognition happens in the moment. Not at the annual review. Not in a quarterly nomination. In the moment, from someone who saw the thing happen.
- The new person can ask the dumb question. Edmondson's psychological safety research is the academic name for this. The practical name is "the team has your back even when you do not know what you are doing yet."
Notice what is not on the list. Catered lunch. A perks portal. A redesigned office. A new HRIS. Those things are fine. They are not the lever. The lever is the daily interaction layer, and it is the part of EX that conventional tooling cannot reach.
Why EX Tools Mostly Skip the Interaction Layer
If team interaction is where EX actually lives, why does the EX software category mostly ignore it? Three reasons.
Surveys are easier to ship than behavior change. A pulse tool is a form, a database, and a dashboard. A team behavior tool is harder to design, harder to sell, and harder to demonstrate ROI on in a 30-minute demo. The market took the easy path.
Surveillance became the default fallback when surveys ran thin. Once vendors realized sentiment data was weak, some pivoted to "behavioral insights" that turned out to be Slack message volume, calendar density, and screen time. Slack activity is not a signal. Counting messages tells you who types a lot. It does not tell you whether the team is collaborating well. The market correctly rejected surveillance, and then failed to find an alternative.
The real signal lives inside team behavior, and team behavior is hard to measure without disrupting it. The Heisenberg problem of HR analytics. The moment you tell people you are measuring how they collaborate, the collaboration changes shape. You stop measuring the thing you wanted to measure and start measuring the performance of the thing.
Measuring Team Behavior Through Designed Practice
The way to get a clean signal on team interaction without distorting it is the same way pilots, surgeons, and elite military units have measured team behavior for half a century: in a simulator. A scenario where the team practices the real work in a low-stakes environment, and the system observes the patterns of how they coordinate, defer, escalate, and recover.
Crew Resource Management training in commercial aviation is the canonical example. The same crew that flew the line all week spent simulator time on coordination exercises. The simulator was the measurement layer because it was the only environment where the team's actual interaction patterns could be observed without distortion. QuestWorks is the flight simulator for team dynamics, applied to knowledge work teams.
Teams play voice-controlled scenario challenges on QuestWorks' own cinematic platform. Each session generates behavioral data on how the team coordinated, who carried decisions, where handoffs got clean and where they got messy, who backed up whom in conflict, where the new hire spoke up and where they froze. QuestDash surfaces those patterns to leaders as aggregate trends and to players as strengths-based callouts. HeroGPT, the private AI coach that lives in Slack, helps individual players work on the behaviors the simulator surfaced. Coaching conversations never share upstream.
Participation is voluntary and never tied to performance reviews. The data comes from voluntary play, not monitoring. That is the difference between behavioral measurement and surveillance. Surveillance measures activity people did not choose to share. Practice measures behavior people demonstrated on purpose, in a designed scenario, knowing the system was observing.
The Definition That Actually Lets You Improve EX
So here is the working definition for the rest of 2026.
Employee experience is the cumulative effect of every interaction an employee has with their employer, weighted heavily toward the daily interactions inside their immediate team, mediated almost entirely by the quality of their direct manager, and measurable only when you combine sentiment data with behavioral data from real team practice.
The Bersin and Morgan frameworks describe the surface. The Gallup manager finding describes the lever. The team interaction layer is where the lever actually moves things. Every EX program that skips the team interaction layer is paying for diagnostics on the wrong system.
If you are evaluating EX software in 2026, the question to ask vendors is no longer "how do you measure sentiment?" Every vendor measures sentiment. The question is "how do you measure the team interaction patterns that produce the sentiment?" Very few vendors have an answer. The ones who do are the ones to watch.
QuestWorks starts at $20 per user per month with a 14-day free trial. It integrates with Slack for install, invites, and HeroGPT coaching, then runs on its own voice-controlled platform for the actual practice. Your team plays. The behavioral signal surfaces. You stop guessing what is producing your engagement scores and start seeing the daily interaction patterns underneath.
Try it free
