Post
JA EN

What to Design Before You Say "Take Ownership"

Posted
By
21 min read
What to Design Before You Say "Take Ownership"
  • Target audience: Executives and managers who want their people to take more ownership (or to think like owners), and anyone working on organizational design
  • Prerequisites: None (this is the design piece in a three-part series; the anatomy of the cause lives in a companion article)
  • Reading time: about 15 minutes

Overview

There are days when “I wish everyone here would take more ownership” is the only sentence that fits. People do what they were told and nothing else. Nobody moves until asked. A problem surfaces two desks over and everyone walks past it. Wanting a wider field of view in that situation is a reasonable thing to want.

The companion piece, “Take Ownership” Becomes a Different Sentence the Moment You Name the Scope, sets out the condition under which that sentence flips. The short version: a request with the scope named draws a boundary and grows ownership, while a request that leaves the scope unnamed erases the boundary and widens responsibility alone. The destination itself is sound. Psychological ownership, understanding of the strategy, and a broadly defined role have all been measured as desirable states. They materialize when information and autonomy get handed over, not when they get demanded.

This article breaks “handing over” into tasks. Decide which numbers to disclose. Draw the line on delegated scope. Write down how the results get split. Each one is something you either did or didn’t do, and none of them is scored on a gradient of effort.

The skeleton of the argument, stated up front: three things get handed over, and the fourth cannot be. Decision rights, information, and stake are designed and transferred. Accountability arises as a consequence once the other three are in place, and it cannot be moved on its own. Trying to move it first is exactly what happens when ownership gets requested with no scope attached.

Empirical work adds a second constraint. Carving out the stake and handing that over alone works poorly. A meta-regression covering 56 studies and 355 estimates finds a positive average association between profit sharing and productivity, and reports the effect as larger when profit sharing is combined with capital investment and with employee participation in decision making1. These are not separable levers. They have to be designed as a set.

1. Four items, three of them tasks

Start with the whole picture. Four things are on the design table.

  1. Decision rights so that a stated scope belongs to the person
  2. Information sufficient to judge inside that scope
  3. Stake so the upside and the downside both come back
  4. Accountability, which appears once the first three are in place

Where authority should sit has a classical grounding in organizational economics. Jensen and Meckling argued that decision rights should move to wherever specific knowledge, the kind that is expensive to transfer, already sits. They then added a condition: moving decision rights creates a control problem, so the allocation has to be designed together with performance measurement and a system of rewards and punishments2. Distributing authority is not a step that completes by itself.

flowchart TB
    A["Hand over<br>decision rights"] --> D["Accountability<br>arises"]
    B["Hand over<br>information"] --> D
    C["Hand over<br>stake"] --> D
    D --> E["Ownership over<br>their own scope"]

Writing those three down is the second step of naming the scope, the contents in the companion piece’s terms. The rest of this article works through them one at a time.

2. Handing over decision rights

2.1 “It’s yours” is not a boundary

The most common failure in delegation is handing something over without stating its extent. “I’m leaving this to you” tells the recipient nothing about how far their own judgment reaches. They either come back for confirmation every time, or they proceed without asking and get corrected afterward.

Argyris pointed out that empowerment had been preached for decades while spreading remarkably little in practice, and that executives themselves often undercut it without noticing3. Saying one thing and doing another is a design failure rather than a motivational one.

Drawing a boundary means writing at least these three things down.

  • The scope you may decide within, stated in units someone can check: budget, timeframe, blast radius, which systems
  • The conditions that require escalation: what threshold sends the decision upward
  • The conditions under which the company overrides you: when the decision gets reversed

The third one is the one that goes missing. After a couple of experiences where a delegated call gets overturned later, people stop making calls. If reversal is possible, writing the reversal conditions in advance widens the real discretion rather than narrowing it.

2.2 One person decides

The other common failure is spreading decision rights thinly across everyone. Consensus processes can get in the way of execution. Rogers and Blenko proposed separating decision roles explicitly (who recommends, who agrees, who performs, who provides input, who decides) and recommended narrowing the decider to a single person4.

“Everyone should think like an owner” quietly does the opposite. When everyone judges everything, the location of judgment disappears and nobody decides. What should be handed over is not judgment over the whole business. It is decision rights over a specified scope.

2.3 It is fine to start small

Nothing requires a large transfer on the first attempt. A review of the effect sizes for participative decision making reports the effect on productivity as statistically significant and small5. That mainly argues for lowering expectations. It also argues that if dramatic improvement is not on the table, betting the whole company on a single rollout is worse value than practicing the writing on one project.

Take one project, or one technical call, and use it to get the organization used to writing boundaries at all.

3. Handing over information

3.1 Check whether the strategy actually arrived

The first item on the list of things people need in order to judge is the strategy itself. What Boswell called line of sight is the degree to which employees understand the organization’s strategic objectives and grasp how they contribute to them, and employees who score higher on it show better job attitudes and lower scores on turnover-related measures6.

The thing to verify is not whether it was communicated but whether it landed. Saying it at an all-hands and putting it in the deck are records of transmission, not of understanding. Ask a few people to name this quarter’s top three priorities. If the answers don’t match, it didn’t land.

3.2 Disclosure is not surveillance

Talk about handing over information often produces initiatives that run the other way: activity dashboards, finer-grained status reporting, constant monitoring. Those take information rather than giving it, and the effect can invert.

Bernstein combined participant observation and a field experiment at the world’s second-largest mobile phone factory and showed that increasing observability leads the observed to hide their activity, which can lower performance. A slight increase in group-level privacy, by contrast, produced a sustained and significant improvement in line performance7. The transparency paradox.

What belongs on the other side of the table is material for judgment, not visibility into behavior. Cost structure, room in the budget, what other departments are dealing with, the terms of the customer contract. Those are needed to decide. A log of who was doing what at 3pm is not.

3.3 Refusals need reasons

Not everything can be disclosed. Unannounced personnel moves, M&A, individual salaries. Some things are genuinely closed.

What works there is saying that they are closed. “I can’t give you that number, because it falls under disclosure rules, so use this indicator instead for the range you need to judge” at least tells the person what to reason from. Silence gets read as unwillingness to hand anything over. A refusal with a reason attached does not.

4. Handing over the stake

4.1 It works poorly on its own

Designing the stake is not limited to money. Grade, role, and the right to choose the next assignment are part of it too. The monetary part is the part with more empirical work behind it.

The meta-regression across 56 studies and 355 estimates finds a positive average association between profit sharing and productivity after controlling for publication bias and misspecification. The part that matters here is the conditions: the association is stronger where union density is high, and the effect is larger when profit sharing is combined with capital investment and with employee participation in decision making1. The stake does not work alone. It works when it meshes with the other conditions.

Which is this article’s skeleton restated. The four conditions work as a set.

Expectations about the monetary piece should stay modest in any case. Employee ownership is a different arrangement from profit sharing, and a meta-analysis of 102 samples covering 56,984 firms puts its correlation with firm performance at r = 0.048. What that measures is organization-level performance rather than the psychology of an individual employee, so the two questions are distinct. Still, the assumption that installing a scheme moves the numbers is not one to hold.

4.2 What you measure determines what gets abandoned

Designing a stake means measuring outcomes. That is where the trap is.

Holmström and Milgrom formalized what happens when an agent handles several tasks and only some are measurable: attaching strong incentives to the measurable part pulls effort away from the rest9. Measure cost and short-term revenue, and people withdraw from quality, safety, documentation, paying down technical debt, and helping colleagues.

So when you settle on a metric, write down in advance what gets sacrificed if someone maximizes it. If the sacrifice exceeds what you can live with, change the metric rather than the person.

4.3 Specific and difficult goals beat vague ones, conditionally

Decades of research have repeatedly confirmed that specific, difficult goals produce higher performance than “do your best”10. On that point the evidence lines up with the prescription here: a concrete goal beats an inspirational one, and “take ownership” on its own is the inspirational kind.

The moderators goal-setting theory names, though, are ability, commitment, feedback, and task complexity, not discretion. Discretion enters when the demand-control model is laid on top. Sharpen the goal while leaving discretion where it was and you move toward the quadrant that model flags as the worst case, high demands with low control11. Put the two findings side by side and the conclusion is that specifying goals and delegating authority have to move together.

5. Accountability is not something you hand over

That covers three. On the fourth there is one thing worth saying from a design standpoint.

Accountability cannot be transferred by itself. More precisely, what transfers on its own is the burden of it, and functioning accountability never arrives. The burden becomes accountability when decision rights, information, and stake are all present. Demanding it while withholding the three is what “take ownership” amounts to when no scope comes with it.

The same holds for ownership. The routes by which a person comes to feel that something is theirs are being able to control it, knowing it intimately, and having invested themselves in it12, and a meta-analysis synthesizing 141 studies adds safety-related antecedents such as organizational justice, trust, and organizational support13. Being ordered to is not on either list.

So the substitute for “show more ownership” is to hand over control (decision rights), hand over the material for knowing the work deeply (information), and build a path by which what someone put in comes back to them (stake). All three are the design items this article has been working through.

6. What has to be true first, and how far down it goes

Two things remain: what makes this design work possible at all, and how far into the organization it has to reach.

6.1 Not being able to write it down means having no criterion for what to drop

Section 8 asks you to write five items for a single project. If your hand stops there, the cause is usually not a gap in design technique. It is that what to prioritize and what to abandon has not been settled, so there is no criterion for where the line goes.

As Porter argued, the essence of strategy is choosing what not to do14, and a boundary is that choice given a shape. For the designer the implication is plain: decide what you are betting on and what you are dropping this quarter, first. Once that exists, the numbers to disclose, the scope to delegate, and the rules for splitting results are all just writing. Attempt the four conditions with that space blank and every project looks equally important, so no scope can be cut.

None of which means management decides in isolation. Involving the people doing the work in strategy formation is compatible with this, since realized strategy contains elements that emerge from action on the ground15. What separates that from dumping the problem on people is whether management retains the judgment framework, the information, and final accountability, not whether the floor participates.

6.2 Every layer gets the same four conditions

Drawing boundaries is not an executive-only job. Management’s tactics are the division’s strategy, and the nesting continues below that. Each layer owes the work to the layer under it.

Which means middle managers need the same four conditions handed to them. A manager told to “delegate to your team” without having received anything is being asked to distribute what they do not have. This is the part that goes unnoticed most often in practice. Organizations where the design is solid from the executive floor down to the division head, and then stops at a verbal “it’s yours,” are not rare.

Part of what Japanese engineering managers are dealing with is exactly this structure (Why Japan’s Engineering Managers Burn Out).

7. Common failures

Having laid out the design work, here is the same material as a list of ways it goes wrong.

Sharpening the goals and leaving the authority alone. Numerical targets get precise while the approval chain stays put. Demands rise, discretion doesn’t, and the load is all that changes11.

Asking for ownership without disclosing anything. The “knowing it intimately” route stays blocked while the output is requested anyway12.

Treating monitoring as disclosure. Activity surveillance supplies no material for judgment and can lower performance7.

Trying to hand everything to everyone at once. The effect size of participation was never large5. Starting with one project beats a company-wide rollout.

Skipping middle management. The design stops at the division level, and managers end up distributing conditions they never received.

Reversing decisions after the fact. Delegation without written reversal conditions stops working after the second or third override.

Ignoring readiness on the receiving end. As Argyris noted, employees are not automatically prepared to take on new responsibility either3. How much you hand over has to be considered alongside the recipient’s experience and the support available.

8. The smallest possible start

No company-wide program needs rebuilding. Take the next project and write the four conditions out as answers to five questions. That is the minimum experiment.

  • How far does this person’s own decision reach on this project (budget, timeframe, blast radius)?
  • Which thresholds require escalation?
  • Under what circumstances does the company override the decision?
  • Of the numbers needed for that judgment, which can be disclosed? For the rest, why not?
  • If this goes well, what comes back to the person, and when?

Any item you cannot write is an item that has not been designed. If you are asking for ownership while one of them is blank, making it writable is the prior task.

And keep what you wrote. When the two sides read the arrangement differently later, an absent record means the judgment belonged to nobody after all.

Summary

When the urge to ask for more ownership arrives, the work available instead looks like this.

Three things get handed over and the fourth cannot be. Design and transfer decision rights, information, and stake, and accountability follows. Reversing that order is what asking for ownership without naming a scope does.

Handing over decision rights means writing three things: the scope someone decides within, the thresholds that require escalation, and the conditions under which the company overrides them. Narrow the decider to one person. When everyone judges everything, the location of judgment disappears.

Handing over information is a different operation from making behavior visible. There is experimental evidence that raising observability lowers performance. What belongs on the other side is material for judgment, and what cannot be disclosed should be refused with a reason.

The stake works poorly in isolation. The profit-sharing meta-regression reports larger effects when it is combined with participation in decision making. And when you choose a metric, write down first what gets sacrificed if someone maximizes it.

Underneath all of it sits the choice of what you are betting on and what you are dropping. Without a criterion for dropping things, no boundary can be drawn, priorities stay unset, and the whole pile lands on the floor.

Then distribute the four conditions layer by layer. Organizations that design carefully down to the division head and stop at a verbal “it’s yours” are common. Nobody can distribute what they were never given.

The smallest experiment is five written answers about one upcoming project. Whatever you cannot write is the part that has not been designed yet.

If you want the view from the receiving end What an employee can do when the authority has not been handed over is covered in the companion piece What You Can Actually Do When You’re Told to “Take Ownership”. Knowing what the person being asked is looking at sharpens the design too.

If you want to know why the phrase misses The anatomy of the structure is in “Take Ownership” Becomes a Different Sentence the Moment You Name the Scope.

References

Footnote numbers correspond to the in-text citations, in order.

  1. Is Profit Sharing Productive? A Meta-Regression Analysis — Doucouliagos, H., Laroche, P., Kruse, D. L., & Stanley, T. D. (2020). British Journal of Industrial Relations, 58(2), 364-395. DOI: 10.1111/bjir.12483. Meta-regression over 56 studies and 355 estimates. After controlling for publication bias and misspecification, profit sharing shows a positive average association with productivity. The association is stronger where union density is high, and the effect is larger when combined with capital investment and employee participation in decision making. [Reliability: High] ↩︎ ↩︎2

  2. Specific and General Knowledge, and Organizational Structure — Jensen, M. C., & Meckling, W. H. (1992). In L. Werin & H. Wijkander (Eds.), Contract Economics, pp. 251-274. Basil Blackwell. The framework holding that the allocation of decision rights must be designed together with performance measurement and a reward-and-punishment system. [Reliability: High (classic academic work)] ↩︎

  3. Empowerment: The Emperor’s New Clothes — Argyris, C. (1998). Harvard Business Review, 76(3), 98-105. Identifies both executive behavior that undercuts empowerment and employees’ own lack of readiness as reasons the idea has been preached for decades without spreading. Introduces the distinction between external and internal commitment. [Reliability: Medium-High (practitioner journal; the author is a leading figure in organizational learning)] ↩︎ ↩︎2

  4. Who Has the D? How Clear Decision Roles Enhance Organizational Performance — Rogers, P., & Blenko, M. (2006). Harvard Business Review, 84(1), 52-61. Proposes separating decision roles (recommend, agree, perform, input, decide) and narrowing the decider to a single person. [Reliability: Medium-High (practitioner journal, based on consulting practice)] ↩︎

  5. Participation’s effects on performance and satisfaction: A reconsideration of research evidence — Wagner, J. A. III (1994). Academy of Management Review, 19(2), 312-330. DOI: 10.5465/amr.1994.9410210753. Finds the effect of participative decision making on productivity statistically significant but small. [Reliability: Medium-High] ↩︎ ↩︎2

  6. Aligning employees with the organization’s strategic objectives: out of ‘line of sight’, out of mind — Boswell, W. R. (2006). The International Journal of Human Resource Management, 17(9), 1489-1511. DOI: 10.1080/09585190600878071. Empirical evidence that understanding of strategic objectives and of one’s own path of contribution (line of sight) relates to job attitudes and turnover-related measures. [Reliability: Medium-High (single field study)] ↩︎

  7. The Transparency Paradox: A Role for Privacy in Organizational Learning and Operational Control — Bernstein, E. S. (2012). Administrative Science Quarterly, 57(2), 181-216. DOI: 10.1177/0001839212453028. Participant observation and a field experiment at a large mobile phone factory in China. Maintaining observability led workers to hide activity and could lower performance, while a slight increase in group-level privacy produced a sustained and significant improvement in line performance. [Reliability: High] ↩︎ ↩︎2

  8. Employee ownership and firm performance: a meta-analysis — O’Boyle, E. H., Patel, P. C., & Gonzalez-Mulé, E. (2016). Human Resource Management Journal, 26(4), 425-448. DOI: 10.1111/1748-8583.12115. 102 samples covering 56,984 firms. The correlation between employee ownership and firm performance is r = 0.04, positive but small. [Reliability: High] ↩︎

  9. Multitask Principal-Agent Analyses: Incentive Contracts, Asset Ownership, and Job Design — Holmström, B., & Milgrom, P. (1991). The Journal of Law, Economics, & Organization, 7(special issue), 24-52. Formalizes how attaching incentives only to measurable dimensions pulls effort away from dimensions that are hard to measure. [Reliability: High] ↩︎

  10. Building a practically useful theory of goal setting and task motivation: A 35-year odyssey — Locke, E. A., & Latham, G. P. (2002). American Psychologist, 57(9), 705-717. DOI: 10.1037/0003-066X.57.9.705. A summary of 35 years of empirical work, repeatedly confirming that specific and difficult goals produce higher performance than “do your best.” [Reliability: High] ↩︎

  11. Job Demands, Job Decision Latitude, and Mental Strain: Implications for Job Redesign — Karasek, R. A. (1979). Administrative Science Quarterly, 24(2), 285-308. DOI: 10.2307/2392498. The original demand-control paper. High demands combined with low decision latitude are most strongly linked to mental strain. [Reliability: High] ↩︎ ↩︎2

  12. Toward a Theory of Psychological Ownership in Organizations — Pierce, J. L., Kostova, T., & Dirks, K. T. (2001). Academy of Management Review, 26(2), 298-310. DOI: 10.5465/amr.2001.4378028. Sets out the three routes through which psychological ownership arises (control, intimate knowledge, self-investment). [Reliability: High] ↩︎ ↩︎2

  13. Psychological Ownership: A Meta-Analysis and Comparison of Multiple Forms of Attachment in the Workplace — Zhang, Y., Liu, G., Zhang, L., Xu, S., & Cheung, M. W.-L. (2021). Journal of Management, 47(3), 745-770. DOI: 10.1177/0149206320917195. Meta-analysis of 141 studies. Antecedents include control, familiarity, and investment plus safety-related factors (organizational justice, trust, organizational support, closeness of relationships). [Reliability: High] ↩︎

  14. What Is Strategy? — Porter, M. E. (1996). Harvard Business Review, 74(6), 61-78. The essence of strategy is choosing what not to do, and a sustainable position presupposes trade-offs. [Reliability: High (classic practitioner and academic literature)] ↩︎

  15. Of strategies, deliberate and emergent — Mintzberg, H., & Waters, J. A. (1985). Strategic Management Journal, 6(3), 257-272. DOI: 10.1002/smj.4250060306. Sets out how realized strategy includes emergent elements arising from action on the ground, alongside deliberate plans. [Reliability: High] ↩︎

Management・Organization
Organization Ownership Delegation Management Organizational Design Japanese Companies Decision Making
This post is licensed under CC BY 4.0 by the author.