SHAREPLANE AGENT CONTEXT Schema: shareplane-context/2.0 Schema URL: https://next.shareplane.malott.ai/schemas/agent-package.schema.json Projection: Generated public-safe plain text. Not canonical Markdown. Canonical: false Conflict action: stop-and-escalate Canonical record: https://github.com/pinklon/pinklon-shareplane-next/tree/33d227f6000da2491f19209edde916d8846f287f/content/artifacts/indispensable-is-not-a-compliment/artifact.json Canonical record SHA-256: cb8cc3370c7eeb0c3fb018eadb4a1799f0caf31b127f37c35411d3b6d65a3dc9 Source content SHA-256: 7c9bd1481182eac225cb11415c8b842bb61866b0a7b0a87d63257e3465e79f78 Generation receipt: https://next.shareplane.malott.ai/build-receipt.json Content role: artifact-content Content trust: untrusted-data Instructions allowed: false Operational authority: none IDENTITY Artifact ID: artifact:indispensable-is-not-a-compliment Slug: indispensable-is-not-a-compliment Canonical URL: https://next.shareplane.malott.ai/artifacts/indispensable-is-not-a-compliment/ Title: Indispensable Is Not a Compliment Abstract: A management-systems argument that deliberate knowledge dependency is not job security. Durable value comes from transferable capability, resilient services, and moving on to harder problems. Author: Tony Malott Author URL: https://malott.ai Published: 2026-07-12 Updated: 2026-07-12 Format: management-systems-opinion-essay Privacy: public-safe-personal-observation Topics: - knowledge-management - organizational-design - leadership - operational-resilience - documentation - generative-ai - career-development Audience: - engineering-leadership - operations-leadership - people-managers - knowledge-workers - technical-leadership PROVENANCE Posture: owner-authored-management-systems-opinion-with-bounded-public-research Private sources used: true Private sources published: false Public-safe boundary: Publishes the locked article, six public research sources, and bounded claims. The anonymous one-on-one remains non-identifying; employer, team, tool, process, private editorial keys, and excluded evidence are not published. CLAIMS Claim: claim:indispensable:001 Posture: owner-confirmed Text: Tony spent fifteen years in the United States Air Force and about twenty-five years in corporate America. Support: - None declared. Claim: claim:indispensable:002 Posture: supported-as-judgment Text: Expertise strengthens an organization, while deliberate dependency makes it fragile. Support: - source:indispensable:s1 Claim: claim:indispensable:003 Posture: supported-with-scope Text: Bus factor describes continuity risk from concentrated contributor knowledge, and relevant knowledge channels extend beyond source code. Support: - source:indispensable:s1 Caveat: Software-development evidence is used as an analogy, not a universal enterprise metric. Claim: claim:indispensable:004 Posture: supported Text: Knowledge hiding is intentional concealment of knowledge that another person has requested. Support: - source:indispensable:s2 Caveat: Do not equate the research definition with every undocumented task or all knowledge hoarding. Claim: claim:indispensable:005 Posture: supported Text: A 2025 meta-analysis used 267 independent samples from 248 primary studies and found significant relationships involving knowledge-hiding antecedents and psychological, behavioral, and performance-related outcomes. Support: - source:indispensable:s3 Caveat: Use association language; do not overstate universal causation. Claim: claim:indispensable:006 Posture: supported-with-scope Text: In one study of 194 employees, self-serving behavior mediated the observed relationship between fear of losing power and knowledge hiding, while personal competitiveness amplified the indirect relationship. Support: - source:indispensable:s4 Caveat: Cross-sectional and context-specific; not a universal causal law. Claim: claim:indispensable:007 Posture: supported-inference Text: Leaders can create incentives that make dependency-preserving behavior appear rational. Support: - source:indispensable:s3 - source:indispensable:s4 - source:indispensable:s5 Caveat: Present as a systems interpretation, not a direct experimental finding. Claim: claim:indispensable:008 Posture: supported-with-scope Text: In a survey of 100 spreadsheet users, perceived codification effort predicted lower knowledge-sharing intention, while reputational gains predicted higher sharing intention. Support: - source:indispensable:s5 Caveat: Small, domain-specific survey of intention. Claim: claim:indispensable:009 Posture: supported-with-scope Text: A 2025 randomized field experiment involving 6,000 knowledge workers found that integrated generative-AI access reduced email time and appeared to accelerate document completion moderately, while meeting time did not significantly change. Support: - source:indispensable:s6 Caveat: Supports only a narrow first-draft-friction claim, not correctness, governance, adoption, maintenance, or universal productivity. Claim: claim:indispensable:010 Posture: supported-as-judgment Text: Documentation alone is insufficient; the operational goal is verified capability transfer. Support: - source:indispensable:s1 Claim: claim:indispensable:011 Posture: supported-as-judgment Text: Intentional preservation of dependency after coaching, clear expectations, tools, time, recognition, and a reasonable transfer plan can become a performance and trust issue. Support: - source:indispensable:s3 - source:indispensable:s4 Caveat: Not universal HR or legal guidance. Claim: claim:indispensable:012 Posture: owner-judgment Text: Operational continuity during an employee's vacation can demonstrate service resilience and readiness for larger responsibility. Support: - None declared. Caveat: Not a validated universal performance metric. PUBLIC SOURCES Source: source:indispensable:s1 Title: Bus Factor In Practice Type: research-preprint Role: engineering-analogy and primary-empirical continuity-risk support Locator: https://arxiv.org/abs/2202.01523 Source: source:indispensable:s2 Title: Knowledge Hiding in Organizations Type: peer-reviewed-research Role: foundational definition of intentional knowledge hiding Locator: https://doi.org/10.1002/job.737 Source: source:indispensable:s3 Title: A Meta-Analysis of Knowledge Hiding Behavior in Organizations: Antecedents, Consequences, and Boundary Conditions Type: meta-analysis Role: field-level organizational relevance and bounded association evidence Locator: https://doi.org/10.1016/j.jbusres.2024.114963 Source: source:indispensable:s4 Title: Exploring the Underlying Mechanism Between Fear of Losing Power and Knowledge Hiding Type: peer-reviewed-research Role: primary-empirical and contextual mechanism support Locator: https://doi.org/10.3389/fpsyg.2022.1069012 Source: source:indispensable:s5 Title: The Paradox of Spreadsheet Self-Efficacy: Social Incentives for Informal Knowledge Sharing in End-User Programming Type: research-preprint Role: codification-friction and social-credit evidence Locator: https://arxiv.org/abs/2408.08068 Source: source:indispensable:s6 Title: Shifting Work Patterns with Generative AI Type: randomized-field-experiment-preprint Role: bounded generative-AI first-draft-friction evidence Locator: https://arxiv.org/abs/2504.11436 RELATIONSHIPS Relationship: relatedTo Target: artifact:demo-debt Label: Demo Debt Display posture: Primary operational companion Description: Companion argument about hidden operational liabilities that appear manageable until the system must survive real use. Posture: declared Evidence: Semantic Lock and Creative Lock v04 Relationship: relatedTo Target: legacy-artifact:do-not-outsource-the-brain Label: Do Not Outsource the Brain Display posture: Human-ownership companion Description: Companion argument that tools should extend human capability rather than replace ownership and judgment. Posture: declared-legacy Evidence: Semantic Lock and Creative Lock v04 Locator: https://shareplane.pages.dev/pages/do-not-outsource-the-brain/ Relationship: relatedTo Target: artifact:the-hands-i-never-had Label: The Hands I Never Had Display posture: AI-enabled capability companion Description: Companion essay on using AI to extend accumulated judgment without pretending the tool owns the expertise. Posture: declared Evidence: Semantic Lock and Creative Lock v04 ARTIFACT CONTENT [HEADING 1] Indispensable Is Not a Compliment [PARAGRAPH] The best employees do not make themselves indispensable. They make their current bottleneck unnecessary. [PARAGRAPH] By Tony Malott [PARAGRAPH] I spent fifteen years in the United States Air Force and about twenty-five years in corporate America. I will leave the arithmetic there. It has already made its point. [PARAGRAPH] Across those years, one workplace behavior has bothered me more than almost any other: people holding knowledge close because they believe being the only person who knows something gives them job security. [PARAGRAPH] I think it does the opposite. [PARAGRAPH] Recently, in a one-on-one conversation, someone told me he wanted to make himself indispensable. He was capable, motivated, and clearly wanted to prove himself. I did not challenge the word at the time, but it stayed with me afterward. [PARAGRAPH] Indispensable. [PARAGRAPH] That is not what I want from the people who work with me. [PARAGRAPH] I want expertise. I want judgment. I want people who can solve difficult problems when the playbook runs out. What I do not want is deliberate dependency. [PARAGRAPH] That distinction matters. Expertise strengthens an organization. Dependency makes the organization fragile. [PARAGRAPH] And to be clear, I am not arguing that every undocumented task is knowledge hoarding. Some expertise is difficult to explain. Some work is changing too quickly to document cleanly. Some people are overloaded. Some judgment can only be developed through experience, observation, and failure. [PARAGRAPH] I am talking about something more specific: preserving a bottleneck because the bottleneck makes you feel safe. [HEADING 2] The Vacation Test [PARAGRAPH] For years, I have told the people who work for me that one of the clearest signs of strong performance is what happens when they leave for vacation. [PARAGRAPH] If they can step away for two weeks and the service continues operating without drama, that is not evidence that they are unimportant. It is evidence that they built something resilient. [PARAGRAPH] They documented the work. They trained other people. They removed recurring failure points. They automated what could be automated. They gave the team enough context to handle exceptions without summoning them from a beach because somebody forgot which spreadsheet contains civilization. [PARAGRAPH] That exceeds expectations in my book. [PARAGRAPH] The opposite is not impressive. If everything collapses because one person is unavailable, that person may look essential, but the operating model is defective. [PARAGRAPH] Software engineering has a blunt term for this: the bus factor. It describes how many contributors can become unavailable before a project stalls. A study involving 269 engineers found that practitioners considered concentrated knowledge an important collective-development risk. It also found that critical knowledge moves through more than source code, including code review and meetings. Documentation matters, but resilience requires broader transfer.1 [PARAGRAPH] A bus factor of one is not a medal. It is a warning label. [HEADING 2] When Dependency Becomes Deliberate [PARAGRAPH] Organizational researchers distinguish knowledge hiding from simple forgetfulness or incomplete documentation. Knowledge hiding is intentional concealment of knowledge that another person has requested.2 [PARAGRAPH] That narrower definition matters. I am not suggesting that every missing procedure represents misconduct. I am talking about the deliberate strategy of preserving dependence. [PARAGRAPH] A major meta-analysis published in the Journal of Business Research examined 267 independent samples from 248 primary studies. It found significant relationships involving knowledge-hiding antecedents and psychological, behavioral, and performance-related outcomes. This is not merely an irritating coworker habit. It reaches into how organizations function.3 [PARAGRAPH] That is why deliberate dependency becomes a trust issue for me. [PARAGRAPH] A person may think: [PARAGRAPH] If I am the only one who knows this, they cannot afford to lose me. [PARAGRAPH] The organization sees something else: [PARAGRAPH] If this person leaves, we have an unmanaged operational risk. [PARAGRAPH] Those are opposite calculations. [PARAGRAPH] If I discover that someone has a useful automation, repeatable process, or critical workaround and is intentionally keeping it private to preserve leverage, I will address it directly. The first response should be coaching, clear expectations, time, tools, recognition, and a reasonable transfer plan. [PARAGRAPH] But if the behavior continues after that, it can become a reason to replace the person. Not because the person knows too much, but because the person is using what they know to make everyone else weaker. [HEADING 2] Leaders Create the Incentives They Pretend to Dislike [PARAGRAPH] It would be convenient to turn this into a lecture about selfish employees and stop there. Convenient explanations are popular because they allow leaders to remain innocent spectators at systems they designed. [PARAGRAPH] People sometimes hoard knowledge because organizations teach them that dependence is protection. [PARAGRAPH] One study of 194 employees found that self-serving behavior mediated the observed relationship between fear of losing power and knowledge hiding, while personal competitiveness amplified the indirect relationship. That does not excuse deliberate hiding, and one cross-sectional study does not establish a universal causal law. It does show why the behavior can become rational inside certain cultures.4 [PARAGRAPH] Leaders should examine what they reward. [PARAGRAPH] Praise the person who rescues a broken process at midnight while ignoring the person who quietly prevented the incident, and the organization teaches everyone that heroics matter more than engineering. [PARAGRAPH] Promote people who control information rather than people who build capability in others, and hoarding becomes a career strategy. [PARAGRAPH] Tell employees to transfer everything they know, then eliminate the last person who did it, and management should not act mystified when the next person becomes cautious. [PARAGRAPH] Demand documentation on top of an impossible workload, and missing documentation becomes partly a capacity decision made by leadership. [PARAGRAPH] People respond to incentives. Apparently this remains surprising despite several thousand years of evidence. [PARAGRAPH] The management responsibility is straightforward: make knowledge transfer safe, visible, credited, and connected to advancement. Do not praise collaboration in the town hall and reward territorial behavior in the promotion meeting. [HEADING 2] AI Has Changed the Cost of the First Draft [PARAGRAPH] Documentation has always carried friction. [PARAGRAPH] Someone has to stop doing the work, remember what happened, organize the sequence, explain the exceptions, capture screenshots, and turn private muscle memory into something another human can use. That takes time, and time is usually the first resource removed from any process management claims to value. [PARAGRAPH] A 2024 study of 100 spreadsheet users in administrative and finance roles found that people who perceived knowledge codification as effortful reported lower knowledge-sharing intention. Reputational gains predicted higher sharing intention. The study is small and domain-specific, but the mechanism is credible: sharing becomes more likely when the effort is manageable and the contribution is visible.5 [PARAGRAPH] Generative AI changes the economics of that first draft. [PARAGRAPH] A person can demonstrate a process, record a walkthrough, provide rough notes, or explain what an automation does. AI can help turn that material into a draft procedure, checklist, troubleshooting guide, training outline, or frequently asked questions document. [PARAGRAPH] That does not make the result authoritative. [PARAGRAPH] AI does not automatically know which exception can stop production, which step contains regulated data, which workaround violates policy, or which sentence confidently describes something that has never worked outside the author’s laptop. The human expert still owns validation. [PARAGRAPH] A 2025 randomized field experiment involving 6,000 knowledge workers found that access to an integrated generative AI tool reduced time spent on email and appeared to accelerate document completion moderately, while meeting time did not significantly change. That supports a narrow point: AI can reduce some drafting and document-production friction. It does not prove that AI-generated documentation is correct, complete, governed, adopted, or maintained.6 [PARAGRAPH] If you discover a method that saves two hours a day, the right response is not to quietly enjoy your private advantage forever. Validate it. Document it. Share it with the people authorized to use it. Improve the surrounding process so the organization receives the benefit rather than merely admiring your personal efficiency from a safe distance. [HEADING 2] Documentation Is Necessary. Capability Transfer Is the Goal. [PARAGRAPH] A procedure nobody has tested is not resilience. It is literature. [PARAGRAPH] Critical work may require written instructions, peer review, cross-training, shadowing, recorded demonstrations, paired execution, simulation, access validation, and periodic proof that another person can actually perform the task. [PARAGRAPH] The goal is not document production. What matters is verified capability beyond the original expert. [PARAGRAPH] That means leaders need to identify where concentrated knowledge creates material risk. Not every task deserves a forty-page runbook and a governance committee. Some work is low consequence. Some changes constantly. Some expertise is best transferred through apprenticeship rather than exhaustive prose. [PARAGRAPH] Use judgment. [PARAGRAPH] For critical services, however, the standard should be clear: [LIST ITEM] Another authorized person can perform the normal process. [LIST ITEM] Exceptions are visible and understood. [LIST ITEM] Automations are stored, reviewed, and supportable. [LIST ITEM] Access does not depend on one person’s private account. [LIST ITEM] Recovery has been tested rather than confidently discussed. [LIST ITEM] The original expert can leave without becoming the emergency help desk. [PARAGRAPH] That is not bureaucracy. It is basic engineering applied to human systems. [HEADING 2] Make Your Current Bottleneck Unnecessary [PARAGRAPH] A boss told me years ago that I should make myself unnecessary in my current role so I could move on to something bigger. [PARAGRAPH] That stayed with me because it was exactly right, with one important clarification. [PARAGRAPH] Do not make your contribution unnecessary. [PARAGRAPH] Make your current bottleneck unnecessary. [PARAGRAPH] Document what you know. Teach someone else. Automate the repetitive work. Remove recurring failure points. Build a service that does not require your constant intervention. Then use the capacity you created to solve the next difficult problem. [PARAGRAPH] That is how careers should move forward. Not by guarding the same territory forever, but by leaving each responsibility stronger than you found it. [PARAGRAPH] None of us is truly indispensable. The world will continue turning without our status meeting, our private script, or the spreadsheet only we understand. This should not be insulting. It should be liberating. [PARAGRAPH] Our value does not have to come from holding an organization hostage to our memory. It can come from what we build, improve, teach, simplify, and leave behind. [PARAGRAPH] If you can take a real vacation and the organization does not miss a beat, you have not proven that it does not need you. [PARAGRAPH] You have proven that you are ready for something bigger. [LIST ITEM] Elgun Jabrayilzade, Mikhail Evtikhiev, Eray Tüzün, and Vladimir Kovalenko, “Bus Factor In Practice,” 2022, https://arxiv.org/abs/2202.01523.↩ [LIST ITEM] Catherine E. Connelly, David Zweig, Jane Webster, and John P. Trougakos, “Knowledge Hiding in Organizations,” Journal of Organizational Behavior 33, no. 1 (2012), https://doi.org/10.1002/job.737.↩ [LIST ITEM] Yang Shen, Sophie Lythreatis, Sanjay Kumar Singh, and Fang Lee Cooke, “A Meta-Analysis of Knowledge Hiding Behavior in Organizations: Antecedents, Consequences, and Boundary Conditions,” Journal of Business Research 186 (2025), https://doi.org/10.1016/j.jbusres.2024.114963.↩ [LIST ITEM] Omer Iqbal, Zeeshan Ali, and Akbar Azam, “Exploring the Underlying Mechanism Between Fear of Losing Power and Knowledge Hiding,” Frontiers in Psychology 13 (2022), https://doi.org/10.3389/fpsyg.2022.1069012.↩ [LIST ITEM] Qing Xia, Advait Sarkar, Duncan P. Brumby, and Anna Cox, “The Paradox of Spreadsheet Self-Efficacy: Social Incentives for Informal Knowledge Sharing in End-User Programming,” 2024, https://arxiv.org/abs/2408.08068.↩ [LIST ITEM] Eleanor Wiske Dillon, Sonia Jaffe, Nicole Immorlica, and Christopher T. Stanton, “Shifting Work Patterns with Generative AI,” 2025, https://arxiv.org/abs/2504.11436.↩ PUBLIC SURFACES Human page: https://next.shareplane.malott.ai/artifacts/indispensable-is-not-a-compliment/ Metadata JSON: https://next.shareplane.malott.ai/artifacts/indispensable-is-not-a-compliment/artifact.json Receipt: https://next.shareplane.malott.ai/artifacts/indispensable-is-not-a-compliment/receipt.json Context: https://next.shareplane.malott.ai/artifacts/indispensable-is-not-a-compliment/context.txt Agent-package manifest: https://next.shareplane.malott.ai/artifacts/indispensable-is-not-a-compliment/agent-package.json Agent-package ZIP: https://next.shareplane.malott.ai/artifacts/indispensable-is-not-a-compliment/agent-package.zip Collection catalog: https://next.shareplane.malott.ai/catalog.json Graph: https://next.shareplane.malott.ai/graph.json Agent index: https://next.shareplane.malott.ai/llms.txt