Job Characteristics: S-Tier Behavioral Designer’s Guide

Every workplace gamification consultancy that sells you a “points and leaderboards” system in 2026 is selling a degraded version of a 1976 paper they probably haven’t read. The original is better. It also predicted, with disturbing precision, what would go wrong when we tried to motivate humans with the version we ended up with.

The paper is “Motivation Through the Design of Work” by Richard Hackman and Greg Oldham. The framework it built, the Job Characteristics Model, sat quietly in organizational behavior textbooks for fifty years while consultants made millions reinventing badly what it already named precisely. Then algorithmic management started stripping the same five characteristics out of millions of jobs at industrial scale, and the model became urgent again.

This guide is the version I wish every workplace gamification client had read before they hired anyone. It maps the five characteristics onto the eight Octalysis Core Drives. It names where the math behind the Motivating Potential Score is wrong and what to use instead. It tells you why Hackman and Oldham’s Growth Need Strength moderator was right in spirit and weak in measurement. And it walks through the Octalysis × JCM audit I run inside actual job-redesign engagements at The Octalysis Group.

Speed Run Notes

• Hackman and Oldham’s 1976 Job Characteristics Model names five characteristics (Skill Variety, Task Identity, Task Significance, Autonomy, Feedback). These five characteristics determine whether a job feels motivating from the inside, before anyone adds points, badges, or leaderboards on top.
• The Motivating Potential Score is a famous formula but its multiplicative form is wrong: Fried & Ferris’s 1987 meta-analysis and every replication since shows a simple additive composite predicts outcomes about as well. Use the characteristics list, drop the math.
• The Octalysis crosswalk: Skill Variety + Feedback activate Core Drive 2; Task Identity activates Core Drive 4; Task Significance activates Core Drive 1; Autonomy activates Core Drive 3 + Core Drive 4. This is the structural alignment workplace gamification keeps trying to bolt on instead of design in.
• Where JCM falls apart: objective characteristics matter less than perceptions (Wrzesniewski & Dutton’s job crafting), Growth Need Strength as a moderator barely replicates, and JCM ignores the social/relational layer that Adam Grant’s 2007 task-significance work added.
• The Elephant in the Room: algorithmic management on gig platforms is systematically lowering Task Identity, Autonomy, and Feedback for millions of workers. JCM warned us this would crash motivation 50 years ago. Read the empirical record on Uber driver turnover, Amazon warehouse churn, and platform-worker depression rates.
• The 6-step Octalysis × JCM Job Redesign Audit lives in the Apply section: score current job on V, I, S, A, F; map each weak characteristic to its Core Drive; redesign the structural surface (vertical loading, natural work units, client relationships, feedback channels); measure with the WDQ at six months.

Author Credibility: Yu-kai Chou

Yu-kai Chou created the Octalysis Framework after studying gamification since 2003, years before the term entered mainstream vocabulary. As a Human-Systems Architect and Behavioral Designer, his framework has been applied by LEGO, Microsoft, Porsche, Coca-Cola, Salesforce, and MrBeast, impacting over 1.5 Billion Users.

Chou has taught the Octalysis methodology at Harvard, Stanford, Yale, Tesla, Google, BCG, and IDEO.

His work has been cited by Harvard, Stanford, MIT, Forbes, Wall Street Journal, Wired, US Department of Energy, NIST, NSF, NCBI, US Department of Education, ClinicalTrials.gov, and Google Scholar — with 3,700+ more academic publications. Explore his books here.

What Is the Job Characteristics Model?

The Job Characteristics Model (JCM) is a structural theory of work motivation that traces every output we care about (internal motivation, work quality, satisfaction, retention) back to five characteristics of the job itself rather than to the worker’s personality or the manager’s pep talks. Richard Hackman and Greg Oldham published the canonical version in 1976 in Organizational Behavior and Human Performance, building on Hackman’s earlier work with Edward Lawler and on Frederick Herzberg’s two-factor distinction between motivators and hygiene factors.

The argument is structural: don’t ask whether your people are motivated. Ask whether their jobs, as designed, contain the structural features that produce motivation from the inside. If those features are absent, no amount of recognition theater, employee-of-the-month plaques, or quarterly engagement surveys will move the needle for long. If those features are present, motivation tends to emerge whether or not anyone manages it explicitly.

Hackman and Oldham named five such features. They mapped how those five produce three internal psychological states. They proposed a formula, the Motivating Potential Score, to combine the five into a single index. They built the Job Diagnostic Survey to measure all of it. The math has aged badly. The characteristics list has aged extraordinarily well.

The Five Core Job Characteristics

The five characteristics aren’t variables you tune individually like sliders on a mixing board. Each one carries a specific load in the motivation system, and missing any one creates a predictable failure mode the other four can’t compensate for.

Skill Variety

Skill Variety asks: does the job demand a range of different activities involving different talents and abilities, or the same narrow motion repeated? An ER nurse runs on high Skill Variety: she does triage assessment, IV placement, family communication, chart documentation, code-team coordination, sometimes pediatric and sometimes geriatric care, within a single shift. An assembly-line worker stamping one specific stamp on one specific component for eight hours runs on low Skill Variety.

Skill Variety isn’t the same as task complexity. A simple job can have variety if it cycles through multiple distinct activities, and a complex job can be low-variety if it grinds through one cognitive operation repeatedly. The construct Hackman and Oldham measured is the number of different demands the job makes on the worker’s repertoire.

Task Identity

Task Identity asks: does the worker complete a whole, identifiable piece of work from start to finish, or just one fragment in a long assembly line? A cabinet-maker who builds a cabinet from the lumber selection to the final lacquer has high Task Identity. A drywall installer who hangs one piece of drywall on one wall of one room has low Task Identity. The cabinet exists as a recognizable thing because of the cabinet-maker; the drywall exists as one component of a building no single worker can claim.

This is the most under-appreciated characteristic in modern job design. Fragmentation is the default, because fragmentation lets management interchange workers and standardize quality control. Every gain from fragmentation comes at the cost of Task Identity, and the costs show up not on the same balance sheet that records the gains.

Task Significance

Task Significance asks: does the work have substantial impact on the lives or wellbeing of other people, whether inside the organization or in the broader world? A surgeon’s task significance is obvious. A janitor cleaning an oncology ward also has high task significance, and it’s not condescending feel-good talk to say so. It’s a structural fact that staphylococcus infections kill immunocompromised patients and the janitor’s diligence prevents some of those deaths.

Task Significance is where Adam Grant’s 2007 paper “Relational Job Design and the Motivation to Make a Prosocial Difference” added important depth. Grant showed empirically that direct contact with beneficiaries (meeting the patient, donor, customer, or recipient your work helps) amplifies the motivational pull of Task Significance well beyond merely being told that the work matters.

Autonomy

Autonomy asks: does the worker have substantial freedom, independence, and discretion in scheduling the work and determining the procedures for carrying it out? A field service technician who decides which calls to take first, how to diagnose the problem, and which parts to bring runs on high Autonomy. A call-center agent following a scripted decision tree under per-call time pressure runs on low Autonomy.

Autonomy is the characteristic management is most reluctant to grant because Autonomy is the characteristic that visibly transfers control. The other four can be granted without anyone in the org chart losing power. Autonomy means someone above the worker has to give up the assumption that they know best and trust that someone closer to the work can decide better.

Feedback

Feedback in the JCM sense is specifically feedback from the job itself, not from a manager’s quarterly review or a colleague’s offhand comment. A salesperson sees the deal close or not close. A coder sees the test pass or fail. A teacher sees the student understand or not understand. The job gives the worker direct, immediate, unambiguous information about how well they’re performing without anyone needing to mediate that information.

Feedback from a manager is fine but slower, noisier, and politically loaded. Feedback from the job is fast, clean, and motivationally pure. Workplace gamification consultants who bolt point-systems onto jobs are reaching for the wrong kind of feedback. They’re trying to mediate a signal the job should be delivering directly. Real Feedback design redesigns the job so the signal flows naturally, not so a points dashboard wraps it in noise.

The Three Critical Psychological States

Hackman and Oldham’s actual theoretical contribution wasn’t the five characteristics. Frederick Herzberg, Louis Davis, and others had pieces of the list earlier. It was the mechanism: the five characteristics produce motivation only by passing through three intermediate psychological states, and naming those states makes the model predictive instead of correlational.

Experienced Meaningfulness

Skill Variety, Task Identity, and Task Significance combine to produce Experienced Meaningfulness: the worker’s sense that the work itself counts, beyond the paycheck. Meaningfulness is a felt state, not a logical conclusion. A worker can be told their work matters and not feel it; a worker can do work most observers would dismiss as menial and feel deeply that it matters. The model says the three input characteristics are how that felt sense actually arises.

Experienced Responsibility

Autonomy produces Experienced Responsibility for outcomes: the worker’s sense that the results of the work are personally attributable to their own effort, judgment, and choices rather than to forces beyond their control. A worker without Autonomy can be praised for good outcomes and not feel responsible for them, because the procedure was dictated. A worker with Autonomy who produces the same outcomes owns them in a way the praise alone cannot replicate.

Knowledge of Results

Feedback produces Knowledge of the Actual Results of the Work Activities: the worker’s understanding of how well they are performing as the work unfolds, not as a delayed quarterly summary. Without Knowledge of Results, the worker is flying blind. The other states can be present and motivation can still collapse because the worker has no signal to refine their effort against.

The model claims, and most empirical work has confirmed, that all three states must be at least minimally present for the model’s outcomes to emerge. Two strong states and one absent state doesn’t average out. Motivation is a conjunction, not a sum.

The Motivating Potential Score (and Why the Formula Is Wrong)

Hackman and Oldham proposed the Motivating Potential Score (MPS) to summarize all five characteristics in one number:

MPS = ((Skill Variety + Task Identity + Task Significance) / 3) × Autonomy × Feedback

The math has a specific theoretical claim baked into it. The three Meaningfulness inputs are averaged because they substitute for one another to some degree: high Task Significance can carry a job with modest Skill Variety. But Autonomy and Feedback are multiplicative because they’re irreplaceable: a job with zero Autonomy has zero MPS no matter how meaningful the work feels, and a job with zero Feedback has zero MPS no matter how much autonomy the worker has.

That’s an elegant theoretical claim. It’s also empirically wrong, or close enough to wrong that the formula stopped being taken seriously decades ago.

Yitzhak Fried and Gerald Ferris’s 1987 meta-analysis in Personnel Psychology tested the multiplicative MPS against a simple additive composite of the same five characteristics. The additive composite predicted job satisfaction, internal motivation, and performance about as well as the multiplicative MPS, and in some cases slightly better. The multiplicative structure adds complexity without adding predictive power.

Stephen Humphrey, Jennifer Nahrgang, and Frederick Morgeson’s 2007 integrative meta-analysis confirmed the finding with a much larger dataset and extended the critique: not only is the multiplicative form unsupported, but the five-characteristic structure itself is incomplete. A more comprehensive Work Design Questionnaire from Morgeson and Humphrey 2006 identifies 21 work characteristics across task, knowledge, social, and contextual categories. JCM’s five are the foundation, but they aren’t the whole house.

The practical implication: use the characteristics list as your design checklist. Don’t multiply, don’t divide. Don’t run the JDS, get a five-digit “score,” and present it to your CEO as if it means something precise. The score is folk math. The characteristics it draws from are real.

What Hackman and Oldham Got Right

Three insights from the 1976 paper remain load-bearing fifty years later, even after the formula has been retired and the moderators have been challenged.

First, work motivation is structural. The model takes the manager-pep-talk worldview (motivation is about how you communicate, recognize, and energize people) and replaces it with a design worldview. Motivation is about how the job is shaped before any worker shows up. If the job is shaped wrong, no amount of communication theater fixes it. If the job is shaped right, communication is a multiplier, not a substitute.

Second, the route from input to outcome passes through felt states, not through logical conclusions. A worker has to feel meaningfulness, feel responsibility, and have knowledge of how they’re doing. Telling them they should feel these things doesn’t produce them. The structural inputs do.

Third, the conjunction principle is real even where the multiplicative math is wrong. Strong Skill Variety with zero Autonomy doesn’t add up to medium motivation; it produces a specific failure mode (a skilled worker forced to follow scripts will burn out or leave). Strong Autonomy with zero Feedback doesn’t produce motivation either; it produces a worker spinning their wheels without ever knowing if the spinning matters. The characteristics work together because the psychological states they feed need all three to be present.

Where JCM Falls Apart

For all the model’s strengths, four serious limitations have accumulated in the empirical literature and one is worth highlighting separately.

The Growth Need Strength Moderator Barely Replicates

Hackman and Oldham predicted that the five characteristics would produce motivation more strongly for workers high in Growth Need Strength, the personality trait of wanting challenge, learning, and development from work. The moderator made theoretical sense and let the model account for individual differences. Empirically, it barely shows up. Karen Loher, Raymond Noe, Nancy Moeller, and Michael Fitzgerald’s 1985 meta-analysis found that Growth Need Strength moderated the characteristic-satisfaction relationship at about r = .10, present but small. Fried and Ferris 1987 reached a similar conclusion. The model often performs about equally well across high-growth-need and low-growth-need workers, suggesting either that the personality measure is too crude or that the moderation effect is real but tiny.

Workers Reshape Their Jobs, So Objective Characteristics Lie

The JCM treats the five characteristics as features of the job that the organization designs and the worker receives. Amy Wrzesniewski and Jane Dutton’s 2001 paper “Crafting a Job: Revisioning Employees as Active Crafters of Their Work” overturned that picture. Workers actively reshape their own jobs through task crafting (adding or dropping tasks), relational crafting (changing who they interact with and how), and cognitive crafting (re-framing what the job means). Two hospital cleaning staff with identical job descriptions can have wildly different Experienced Meaningfulness because one has crafted relationships with patients into the role while the other has stayed within the formal description. The model’s objective characteristics underestimate what’s actually happening.

JCM Was Built Before the Social Layer Was Visible

The 1976 model is at root individualistic. It treats the worker-job interaction as the unit of analysis and treats coworkers, customers, and beneficiaries as context. Grant’s 2007 work on Task Significance, the Morgeson-Humphrey Work Design Questionnaire, and the entire emerging literature on relational job design have shown that the social characteristics of work (direct contact with beneficiaries, social support from coworkers, interdependence with colleagues) are first-class motivational inputs, not contextual conditions. JCM needs an extension to capture them.

The Model Travels Less Well Across Cultures Than the Original Paper Implied

Christopher Robert, Tahira Probst, Joseph Martocchio, Fritz Drasgow, and John Lawler’s 2000 paper on cross-cultural variability in job design found that the same Autonomy intervention produced quite different motivation effects in collectivist versus individualist national cultures. Autonomy granted to a worker whose self-concept is interdependent can feel isolating rather than empowering. The five characteristics still matter cross-culturally, but the weight they carry, along with the cultural meaning of granting them, varies.

What’s Really Happening Inside the Brain

The neuroscience of work motivation has caught up to the JCM in ways Hackman and Oldham could not have predicted in 1976, and the convergence is striking.

Autonomy maps onto ventral striatal activation patterns documented by Lauren Leotti, Sheena Iyengar, and Kevin Ochsner in 2010. Choice itself, not just better outcomes from choice, activates reward circuitry. A worker given Autonomy gets a neural reward independent of the work outcome, and that reward shows up in the same circuits cocaine and food activate, just at lower amplitude and over longer time scales.

Task Significance, when amplified by direct beneficiary contact in the Grant 2007 design, recruits the mentalizing network identified by Rebecca Saxe and others: medial prefrontal cortex, temporo-parietal junction, precuneus. The worker is doing literal social cognition about the beneficiary, and that social cognition produces oxytocin-mediated motivational pull that purely abstract significance (“your work matters”) fails to produce.

Feedback from the job itself activates the dopaminergic prediction-error system Wolfram Schultz mapped in the 1990s and refined through the 2010s. Each completion-or-failure signal updates the brain’s predictive model of future reward, and that updating is what creates the felt sense of competence growing. Manager-mediated feedback delivered weeks after the work activates the same system more weakly. The temporal smearing degrades the dopaminergic signal.

The implication for design: the JCM didn’t just describe surface-level workplace phenomena. It accidentally described the design surface that engages the human reward, social, and prediction systems directly. That’s why it still works fifty years later. The brain hasn’t changed; the model touched something real about how the brain processes work.

JCM vs Other Theories

JCM vs Herzberg’s Two-Factor Theory

Frederick Herzberg’s two-factor theory distinguishes hygiene factors (pay, policies, working conditions) whose absence demotivates but whose presence merely prevents dissatisfaction, from motivators (achievement, recognition, the work itself, responsibility) whose presence actually drives motivation. JCM is best read as the operationalization of Herzberg’s motivators category. Where Herzberg said “the work itself” matters, Hackman and Oldham said what specifically about the work itself matters: the five characteristics. JCM is what Herzberg becomes when you stop philosophizing and start measuring.

JCM vs the Job Demands-Resources Model

Evangelia Demerouti, Arnold Bakker, Friedhelm Nachreiner, and Wilmar Schaufeli’s 2001 Job Demands-Resources (JD-R) model corrected JCM’s biggest blind spot: the demands side. JCM lists resources (the five characteristics) but is mostly silent on demands (workload, role conflict, emotional labor, time pressure) that drive burnout. JD-R pairs the two: demands deplete, resources replenish, and burnout emerges when the imbalance is sustained. The healthy reading is that JCM tells you how to design the resources side; JD-R tells you to size the demands side proportionately. Use both, not one.

JCM vs Self-Determination Theory

Edward Deci and Richard Ryan’s Self-Determination Theory (SDT) identifies three psychological needs whose satisfaction produces internal motivation: autonomy, competence, relatedness. JCM’s Autonomy maps directly onto SDT’s autonomy. JCM’s Feedback feeds SDT’s competence. JCM is mostly silent on relatedness while SDT puts it on equal footing with the other two. The clean way to read the relationship: SDT is the psychological theory; JCM is the corresponding work-design theory. SDT names the needs; JCM names the job characteristics that meet them. The two frameworks converge in their pragmatic recommendations far more often than they diverge.

JCM vs Equity Theory

J. Stacy Adams’s Equity Theory looks at motivation as a comparison problem: workers form a fairness perception by comparing their input-output ratio to a referent’s. JCM looks inward at the work itself, Equity Theory looks outward at the comparison. They aren’t competing; they’re addressing different layers. A perfectly designed JCM job can still produce demotivation if the worker discovers a peer with the same job has higher Outputs (pay, status, recognition). A perfectly equitable pay structure can still produce demotivation if the underlying work has zero Autonomy or Feedback. Both layers must be designed.

JCM in the Real World

Workplace Job Redesign

The original JCM-driven job redesign interventions Hackman and Oldham documented in Work Redesign (1980) used five concrete techniques: combining tasks (workers do longer task sequences instead of isolated steps), natural work units (assign work in coherent batches a worker can identify with), establishing client relationships (give the worker direct contact with the downstream beneficiary), vertical loading (push planning and quality decisions down to the worker doing the work), and opening feedback channels (let job-intrinsic feedback flow without manager mediation). Every one of these five remains a high-impact redesign move fifty years later. Modern “team-of-teams” and “self-managed squads” reorganizations are usually some combination of vertical loading and natural work units rebranded.

Knowledge Work and Software Development

Software development is one of the cleanest natural experiments in modern JCM. A well-run engineering org tends to have high Skill Variety (architectural decisions, debugging, code review, mentoring), high Task Identity (engineers own features from spec through ship), variable Task Significance depending on what the product does, high Autonomy (engineers choose their tools and decompose their tasks), and exceptionally high Feedback from the job itself (tests pass or fail in seconds, production telemetry shows whether the feature works for users in hours). GitHub-style PR review provides a feedback channel intrinsic to the work, not bolted on. This is why well-run engineering teams produce internal motivation at levels most corporate function teams envy. The work itself is doing the motivational labor JCM predicted it would.

Healthcare

Linda Aiken and colleagues’ 2014 Lancet study of European hospitals found that nurse-to-patient ratios and job design directly predict patient mortality and nurse burnout. The JCM reading: when hospitals fragment nursing work (each nurse covering more patients in less time with less continuity), Task Identity collapses, Feedback gets noisier, and Autonomy disappears. Nurses experience the predicted motivational and emotional consequences, patients die at higher rates, and the cost shows up downstream as turnover and litigation. Hospital systems that have rebuilt primary-nursing models (one nurse owns a defined patient panel through their stay) have re-established Task Identity and seen the predicted effects reverse. This is JCM running live in life-and-death conditions.

Education

Project-based learning works because it embeds all five characteristics in student work. A student designing and building a working solar-charged device exercises Skill Variety (research, design, fabrication, testing), Task Identity (the device exists or doesn’t because of the student), Task Significance (especially when paired with a community partner who’ll use the device), Autonomy (decisions about scope, approach, and timeline rest with the student), and Feedback (the device works or it doesn’t). The conventional alternative (worksheets in a textbook chapter) strips four of the five characteristics. JCM predicts which produces internal motivation. The empirical record confirms it.

How to Apply the Job Characteristics Model with the Octalysis Framework

The five Job Characteristics map onto the Octalysis Core Drives with a structural precision that surprised me the first time I worked through it. The mapping isn’t thematic; it’s load-bearing. Each characteristic activates specific Core Drives, and that activation explains why the characteristic produces motivation rather than just that it does.

Skill Variety → Core Drive 2 + Core Drive 3

Skill Variety activates Core Drive 2 (CD2): Development & Accomplishment because each different skill the job demands is a different competence the worker is exercising and growing. The mind treats variety as repeated mini-onboardings into new capability domains. It also activates Core Drive 3 (CD3): Empowerment of Creativity & Feedback because variety creates the space for the worker to choose how to bring their full repertoire to bear on different problems rather than executing one pre-decided procedure repeatedly.

Task Identity → Core Drive 4

Task Identity activates Core Drive 4 (CD4): Ownership & Possession because a worker who builds the whole cabinet owns the cabinet in a way the worker who hangs one piece of drywall on one wall does not. Octalysis recognizes that ownership is a felt psychological state, not just a legal one. The cabinet-maker doesn’t legally own the cabinet (the customer does), but the maker owns it psychologically, which is what carries the motivational load. Task Identity is what creates that psychological ownership in the design.

Task Significance → Core Drive 1

Task Significance activates Core Drive 1 (CD1): Epic Meaning & Calling because the worker locates themselves inside a story larger than the paycheck. The mechanism is direct: a worker who can name the lives their work touches has CD1 firing as part of the work, not adjacent to it. Adam Grant’s beneficiary-contact research is CD1 amplification at scale: direct contact with the person who benefits is the dosage form of CD1 that produces the strongest measurable behavior change.

Autonomy → Core Drive 3 + Core Drive 4

Autonomy is the only characteristic that activates two Core Drives heavily on its own. It activates Core Drive 3 (CD3): Empowerment of Creativity & Feedback because Autonomy is structurally the latitude to bring creativity to bear: choose the method, sequence the work, decide what counts as done. It also reinforces Core Drive 4 (CD4): Ownership & Possession because the decisions the worker makes under Autonomy are theirs, and the outcomes those decisions produce belong to them. Without Autonomy, CD3 has no space to operate and CD4’s psychological ownership is hollow.

Feedback → Core Drive 2 (and the Feedback Mechanic family)

Feedback from the job itself activates Core Drive 2 (CD2): Development & Accomplishment because each feedback signal updates the worker’s internal model of their own competence and trajectory. The faster and cleaner the feedback, the tighter the development loop. This is why the Feedback Mechanic family in Octalysis (Progress Bar, Boss Fight, Aura Effect) is so structurally tied to CD2. They are formalizations of what the JCM identifies as job-intrinsic feedback. Where workplace gamification consultancies fail is when they add a points dashboard instead of redesigning the work to deliver Feedback directly. The right design move is to make the work itself feedback-rich; the wrong move is to wrap badly-designed feedback-poor work in a points layer.

The Other Three Core Drives, as Moderators

The remaining Core Drives (Core Drive 5 (CD5): Social Influence & Relatedness, Core Drive 6 (CD6): Scarcity & Impatience, Core Drive 7 (CD7): Unpredictability & Curiosity, and Core Drive 8 (CD8): Loss & Avoidance) moderate the model rather than being its primary drivers. CD5 explains why Grant’s beneficiary-contact design works (relatedness amplifies Task Significance). CD8 is what burnout looks like when the four CDs above stay starved for too long. Workers shift to fear-based motivation that the demands side keeps demanding. CD6 and CD7 show up in well-designed work as the natural rhythm of deadlines and discovery, not as bolted-on artificial scarcity.

The 6-Step Octalysis × JCM Job Redesign Audit

Inside Octalysis Group engagements, I run a six-step audit when a client believes their workplace gamification system isn’t producing the motivation results they expected. The audit usually reveals that the underlying jobs are JCM-broken in ways no point system can fix.

1. Score the current job on V, I, S, A, F. For each role being redesigned, score each of the five characteristics 1-7 using the Job Diagnostic Survey or, for the more comprehensive read, Morgeson and Humphrey’s Work Design Questionnaire. Aim for an honest manager assessment plus an anonymous worker assessment. The gap between the two is itself a finding.
2. Identify the bottom two characteristics. The model is conjunctive: weak Autonomy with strong everything else produces a specific failure mode; weak Feedback with strong everything else produces a different specific failure mode. Name the two weakest and what failure mode each predicts in the actual role.
3. Map each weak characteristic to its Core Drive. Use the crosswalk above. Weak Skill Variety means CD2 + CD3 are under-activated. Weak Task Identity means CD4 is hollow. Weak Task Significance means CD1 is absent. Weak Autonomy means CD3 has no room. Weak Feedback means CD2 has no loop to close.
4. Design the structural surface that would activate the missing Core Drive. For weak Task Identity (hollow CD4): create natural work units, give the worker a defined panel they own through completion, attribute named ownership in handoffs. For weak Feedback (CD2 with no loop): build the feedback signal into the work itself rather than waiting for manager mediation; let the worker see the result the same day. For weak Autonomy (CD3 boxed out): vertical-load planning and quality decisions, replace scripts with principles, document outcomes wanted instead of procedures to follow.
5. Run the publicity test from Libertarian Paternalism. Would your job redesign survive being described accurately to the workers it will affect, including the trade-offs you’re making? If you’re transferring Autonomy down, you’re transferring control away from middle management. Say so. If you’re enriching one role by relocating tasks elsewhere, the elsewhere matters. Publicity test catches dishonest redesigns before they ship.
6. Re-measure at six months. Use the same Work Design Questionnaire baseline you scored in step 1. If V, I, S, A, F have moved in the directions you designed for, and Experienced Meaningfulness, Responsibility, and Knowledge of Results have moved with them, the redesign worked. If V, I, S, A, F moved but the psychological states didn’t, you have a moderator problem, usually context (pay, security, peer relationships) eating the gains. If neither moved, the design changes were theatrical, not structural. Either way, you have a finding that tells you what to do next.

Throughout: Calibrating CD6 and CD8

Core Drive 6 (CD6): Scarcity & Impatience is the design pressure that, used correctly, gives the work natural deadlines and forcing functions that sharpen attention. Used incorrectly, CD6 becomes the always-on urgency that crushes the Autonomy gains you just designed for. Core Drive 8 (CD8): Loss & Avoidance calibrates whether the workers see the design as “we are removing things you used to have” (CD8 firing the wrong direction, resistance follows) or “we are removing things that were getting in your way” (CD8 calibrated, adoption follows). Communication about the redesign isn’t a soft skill on top of the design; it’s part of the design. Communicate poorly and CD8 sinks the change you spent three months engineering.

The Elephant in the Room

Algorithmic management is the largest natural experiment in JCM-violating job design that has ever been run, and every prediction the model makes is showing up in the data.

Ride-share drivers complete a four-minute fragment of a stranger’s commute, with no ongoing relationship to that passenger, no choice of the next ride, no view of the pricing logic that determined their pay, and almost no Feedback channel that isn’t an algorithmic 5-star rating delivered after the fact. Task Identity is shredded. Autonomy is theatrical (you can “go offline” but the platform knows where you are and the algorithm penalizes refusals). Feedback is performative (the 5-star system mediates feedback through a metric that doesn’t tell the driver what they could do better). The model predicts low internal motivation, high turnover, and elevated depression rates. The empirical record on rideshare driver welfare confirms all three predictions.

Amazon warehouse work has the same shape but with the demands side cranked up. Pickers execute a fragment of a fulfillment workflow, route-optimized by an algorithm they can’t override, with seconds-level metrics measuring their pace and termination thresholds applied automatically. JCM-style characteristics are systematically minimized. The annual turnover rate exceeds 150% in many facilities according to leaked internal documents. JD-R would predict burnout (demands large, resources small); JCM would predict zero internal motivation and rapid exit. Both predictions are confirmed.

The discourse around these jobs tends to focus on pay, scheduling, and benefits (which matter) but misses that the underlying work is structurally designed to extract effort while denying the psychological states JCM showed are required for sustainable motivation. You cannot fix algorithmic management with better pay. You also cannot fix it with workplace gamification points overlays, which are how some platforms have tried. The model says: redesign the work itself, or accept that the labor pool will rotate constantly because no individual can sustain motivation in the structural conditions you’ve engineered.

Run the publicity test on the platform side: would you describe the actual algorithmic management design accurately to the workers it governs, including which decisions are made by which algorithm and why? Most platforms will not, and the unwillingness is the diagnostic. JCM’s deeper lesson, that work design is an ethical surface and not just a productivity one, is what the Elephant section in every V6 and V7 pillar has been trying to point at. Workplace gamification done well makes work better. Workplace gamification done as a thin layer over JCM-violating job design is camouflage for harm.

Practical Steps for Applying JCM This Quarter

1. Pick one role. Not “the org chart.” One role with one job description. JCM works at the job level, not the system level.
2. Get the Work Design Questionnaire from Morgeson & Humphrey 2006. Score the role with an anonymous worker sample of at least five people. Score it yourself as the manager. Note where the gaps are largest.
3. Identify the bottom two characteristics. Resist the urge to address all five. Two changes, well-designed, beat five half-changes.
4. Map each weak characteristic to its Core Drive and design one structural intervention. Vertical loading, natural work units, client relationships, combined tasks, or feedback channels. Pick the one that addresses the specific Core Drive deficit you named.
5. Run the publicity test before announcing. Tell the workers exactly what you’re changing and why. If you cannot describe it accurately, the design isn’t ready.
6. Re-measure at six months. Same instrument, same workers. If the characteristics moved but the psychological states didn’t, look at moderators. If neither moved, the change was theatrical.
7. Hold the line on the multiplicative MPS. Don’t compute it. Don’t report it. The characteristics list is the value; the formula is folklore.

The Job Characteristics Model Was the Beginning, Not the End

Hackman and Oldham gave us a foundation. The fifty years since have layered on Wrzesniewski and Dutton’s job crafting, Grant’s relational job design, the Morgeson-Humphrey Work Design Questionnaire’s broader characteristics list, and the JD-R model’s demands-resources symmetry. None of these replace JCM; they extend it. The five characteristics remain the core because they remain measurable, designable, and predictive.

The firm-scale extension is Mark Huselid’s high-performance work systems research: a survey of nearly a thousand firms showing that a coherent system of people practices, work design among them, was worth tens of thousands of dollars more per employee.

Where JCM still has the most to offer in 2026 is exactly where its critics are loudest: as the design checklist for jobs that algorithmic management is trying to render obsolete. When platforms tell you their workers are “independent contractors” who set their own schedules, the JCM checklist is how you measure whether the design actually delivers what the rhetoric promises. Usually it does not. The model’s value is that it gives you the words to say why.

If you redesign one role this quarter using the six-step audit above and the workers report higher Experienced Meaningfulness, higher Experienced Responsibility, and higher Knowledge of Results at six months, the model worked for you the way it worked for the original 1976 case studies. You didn’t add badges. You didn’t run an engagement survey. You redesigned the job. The motivation showed up because the structural conditions for motivation now exist where they didn’t before. That’s what JCM was always about, and it’s why the model is still teaching us how to do this fifty years later.

Frequently Asked Questions

What is the Job Characteristics Model in simple terms?

The Job Characteristics Model says that five features of a job (how varied the skills are, whether you complete whole pieces of work, whether the work matters to other people, how much freedom you have, and whether the job itself tells you how you’re doing) determine whether the work feels motivating from the inside. Designed correctly, the work motivates without needing external recognition systems on top of it.

Who created the Job Characteristics Model?

Richard Hackman and Greg Oldham developed the model in the early 1970s and published the canonical version in 1976 in Organizational Behavior and Human Performance. The book-length treatment, Work Redesign, followed in 1980. The model built on Hackman’s earlier work with Edward Lawler and on Frederick Herzberg’s distinction between motivators and hygiene factors.

What are the five core job characteristics?

Skill Variety (how many different skills the job uses), Task Identity (whether you complete whole pieces of work), Task Significance (how much the work affects other people), Autonomy (how much freedom you have over scheduling and method), and Feedback (whether the job itself tells you how you’re doing).

What is the Motivating Potential Score formula?

MPS = ((Skill Variety + Task Identity + Task Significance) / 3) × Autonomy × Feedback. The formula was designed to capture how all five characteristics combine. Fried & Ferris’s 1987 meta-analysis showed a simple additive composite predicts outcomes about as well as the multiplicative MPS, so the math is largely retired in modern practice while the characteristics list remains valuable.

What are the three critical psychological states in the JCM?

Experienced Meaningfulness (produced by Skill Variety + Task Identity + Task Significance), Experienced Responsibility for Outcomes (produced by Autonomy), and Knowledge of the Actual Results of Work Activities (produced by Feedback). The five characteristics produce motivation by passing through these three felt states, not by directly producing motivation themselves.

How does the Job Characteristics Model relate to the Octalysis Framework?

The mapping is structural. Skill Variety activates Core Drive 2 (Development & Accomplishment) + Core Drive 3 (Empowerment of Creativity & Feedback). Task Identity activates Core Drive 4 (Ownership & Possession). Task Significance activates Core Drive 1 (Epic Meaning & Calling). Autonomy activates Core Drive 3 + Core Drive 4. Feedback activates Core Drive 2 and the Feedback Mechanic family of Game Techniques. The Octalysis × JCM crosswalk lets job-redesign practitioners predict which Core Drive will move when each job characteristic is enriched.

Why doesn’t the Growth Need Strength moderator work well?

Loher and colleagues’ 1985 meta-analysis found Growth Need Strength moderated the characteristic-satisfaction relationship at r = .10, present but small. Fried and Ferris 1987 reached a similar conclusion. The most likely explanations: the personality measure is too crude to capture meaningful individual differences, or the model’s effects are large enough that they show up even for workers who say they don’t value growth.

What is the difference between the Job Diagnostic Survey and the Work Design Questionnaire?

The Job Diagnostic Survey (JDS) is Hackman and Oldham’s original instrument measuring the five JCM characteristics. The Work Design Questionnaire (WDQ), from Morgeson and Humphrey 2006, expands the measurement to 21 work characteristics across task, knowledge, social, and contextual categories. The WDQ is the more comprehensive instrument and is generally preferred in modern research, though the JDS remains useful when you want a focused JCM-only baseline.

Can the Job Characteristics Model be applied to remote or hybrid work?

Yes, and more cleanly than to office work. Remote work tends to enrich Autonomy (workers control schedule and method) and Task Identity (workers tend to own larger chunks of work without manager hovering), but can starve Task Significance and Feedback because beneficiary contact gets mediated and feedback channels narrow. The redesign moves are: build in deliberate beneficiary-contact rituals (Adam Grant’s research), and design job-intrinsic feedback into the work itself rather than relying on synchronous manager check-ins.

What is job crafting and how does it relate to JCM?

Job crafting, from Amy Wrzesniewski and Jane Dutton’s 2001 paper, is the way workers actively reshape their own jobs through task crafting (adding or dropping tasks), relational crafting (changing who they interact with and how), and cognitive crafting (re-framing what the work means). It extends JCM by showing that the five characteristics aren’t fixed features the organization grants. They’re partly negotiated by the worker. Two workers with identical job descriptions can experience quite different motivation profiles depending on how they craft.

Is the Job Characteristics Model still relevant in the age of AI and algorithmic management?

More than ever. Algorithmic management on gig and warehouse platforms systematically strips Task Identity, Autonomy, and Feedback from millions of jobs at scale. JCM gives you the diagnostic vocabulary to name what’s being stripped and predict the consequences (high turnover, low internal motivation, elevated mental-health costs), which the empirical record on platform work has confirmed. The model is fifty years old; the relevance is current.

References

• Hackman, J. R., & Lawler, E. E. (1971). Employee reactions to job characteristics. Journal of Applied Psychology, 55(3), 259-286.
• Hackman, J. R., & Oldham, G. R. (1975). Development of the Job Diagnostic Survey. Journal of Applied Psychology, 60(2), 159-170.
• Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16(2), 250-279.
• Hackman, J. R., & Oldham, G. R. (1980). Work Redesign. Addison-Wesley.
• Fried, Y., & Ferris, G. R. (1987). The validity of the Job Characteristics Model: A review and meta-analysis. Personnel Psychology, 40(2), 287-322.
• Loher, B. T., Noe, R. A., Moeller, N. L., & Fitzgerald, M. P. (1985). A meta-analysis of the relation of job characteristics to job satisfaction. Journal of Applied Psychology, 70(2), 280-289.
• Humphrey, S. E., Nahrgang, J. D., & Morgeson, F. P. (2007). Integrating motivational, social, and contextual work design features: A meta-analytic summary and theoretical extension of the work design literature. Journal of Applied Psychology, 92(5), 1332-1356.
• Morgeson, F. P., & Humphrey, S. E. (2006). The Work Design Questionnaire (WDQ): Developing and validating a comprehensive measure for assessing job design and the nature of work. Journal of Applied Psychology, 91(6), 1321-1339.
• Parker, S. K., Morgeson, F. P., & Johns, G. (2017). One hundred years of work design research: Looking back and looking forward. Journal of Applied Psychology, 102(3), 403-420.
• Grant, A. M. (2007). Relational job design and the motivation to make a prosocial difference. Academy of Management Review, 32(2), 393-417.
• Wrzesniewski, A., & Dutton, J. E. (2001). Crafting a job: Revisioning employees as active crafters of their work. Academy of Management Review, 26(2), 179-201.
• Pierce, J. L., Jussila, I., & Cummings, A. (2009). Psychological ownership within the job design context: Revision of the job characteristics model. Journal of Organizational Behavior, 30(4), 477-496.
• Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands-resources model of burnout. Journal of Applied Psychology, 86(3), 499-512.
• Herzberg, F. (1968). One more time: How do you motivate employees? Harvard Business Review, 46(1), 53-62.
• Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68-78.
• Adams, J. S. (1963). Toward an understanding of inequity. Journal of Abnormal and Social Psychology, 67(5), 422-436.
• Robert, C., Probst, T. M., Martocchio, J. J., Drasgow, F., & Lawler, J. J. (2000). Empowerment and continuous improvement in the United States, Mexico, Poland, and India: Predicting fit on the basis of the dimensions of power distance and individualism. Journal of Applied Psychology, 85(5), 643-658.
• Leotti, L. A., Iyengar, S. S., & Ochsner, K. N. (2010). Born to choose: The origins and value of the need for control. Trends in Cognitive Sciences, 14(10), 457-463.
• Aiken, L. H., Sloane, D. M., Bruyneel, L., et al. (2014). Nurse staffing and education and hospital mortality in nine European countries: A retrospective observational study. The Lancet, 383(9931), 1824-1830.
• Saxe, R., & Kanwisher, N. (2003). People thinking about thinking people: The role of the temporo-parietal junction in “theory of mind”. NeuroImage, 19(4), 1835-1842.
• Schultz, W. (2016). Dopamine reward prediction-error signalling: A two-component response. Nature Reviews Neuroscience, 17(3), 183-195.
• Chou, Y. K. (2015). Actionable Gamification: Beyond Points, Badges, and Leaderboards. Octalysis Media.

Related Reading

• Employee Engagement: Kahn’s Three Conditions and the Gallup Q12
• Burnout: Maslach and Leiter’s Exhaustion, Cynicism, Inefficacy Model
• Equity Theory: Adams’s Input/Output Ratio Comparison
• Situational Leadership: Hersey & Blanchard’s Telling, Selling, Participating, Delegating
• Authentic Leadership: Luthans & Avolio’s Four-Component Model
• The Complete Octalysis Framework

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?