Accountability gives people a reason to explain decisions, fulfil commitments, and respond when something goes wrong. It becomes difficult to trust when the standard changes after an outcome is known or applies differently depending on who made the decision.
A delivery lead may be asked to explain a missed date while the approved scope increases receive little attention. An employee may be criticised for failing to escalate a risk despite records showing that it was raised and accepted. A sponsor’s decision may be described as an unavoidable commercial judgement while a team member’s contribution is treated as a personal failure.
These patterns can turn accountability into a means of assigning blame selectively. Identifying them does not require assuming that every manager is acting deliberately or that the employee closest to the work has no responsibility. It requires examining how expectations were established, which decisions are included in the review, and whether the same standard is applied consistently.
When the Standard Is Reconstructed Afterwards
Consider a customer migration programme. The delivery lead is responsible for coordinating the work and reporting progress. A sponsor approves a fixed launch date, while product and commercial teams continue to request additions. Several changes are accepted without a corresponding adjustment to the date.
When the migration slips, the review concludes that the delivery lead should have controlled scope more firmly. That may be a legitimate question, but it cannot be answered without examining the role. Could the lead refuse additions, or only present their effects? Who approved the changes? What advice was provided, and how were the resulting commitments recorded?
A retrospective standard arises when the organisation assesses the lead as though they possessed authority that was never delegated. It can also arise when an expectation is introduced that was not part of the role and could not reasonably have been inferred from it.
Not every expectation has to be written down before it can be assessed. Professional roles include duties of care, communication, and judgement. The important distinction is whether the standard was reasonably applicable at the time. A review should be able to explain that basis without relying solely on the fact that the outcome was bad.
The migration lead may still have handled part of the work poorly. They might have failed to update a forecast or left an important dependency unexplained. Examining those actions is compatible with examining the sponsor’s approvals. Accountability becomes selective when only one part of that connected decision process is open to scrutiny.
When an Outcome Is Treated as a Complete Explanation
A named owner makes it easier to establish who should maintain a plan, raise a concern, or coordinate a response. The label does not, by itself, explain why an outcome occurred. Shared services, other teams, customers, and external events can all affect the result.
For the migration programme, the lead might own coordination while a platform owner controls an access change and the sponsor controls the launch commitment. If the access change is delayed, the review needs to examine what each participant was expected to do. Describing the lead as accountable for delivery should not erase those separate responsibilities.
Equally, distributed authority should not make responsibility disappear. The platform owner can explain the decision within their remit, the sponsor can explain the accepted trade-off, and the lead can explain how the dependency was managed. A useful review connects those accounts rather than choosing one person to represent the entire system.
Atlassian’s DACI framework provides one way to distinguish the person driving a decision from the approver, contributors, and people who need the result. Such distinctions can help establish expectations before work begins. They need to reflect actual authority and remain current when decisions or responsibilities change.
A successful result deserves the same care. If the migration completes because staff work unsustainable hours, the outcome does not establish that the planning was sound. Judging only the final date can reward a weak process when people manage to compensate and punish a similar process when they cannot.
When the Record Becomes Selective
Decision records help reconstruct what people knew and what they were authorised to do. They become misleading when a review includes the employee’s warning but omits the response, quotes an early estimate after scope has changed, or treats an informal suggestion as an approved commitment.
The relevant record should include the sequence. What was requested? What options and risks were presented? Who decided? What changed afterwards? A message saying that a date was achievable may have depended on conditions that were later removed. The review needs those conditions to interpret the statement fairly.
People should also have an opportunity to correct the factual account. A manager may not know about a discussion held elsewhere, and written records can themselves be incomplete. Treating the first coherent story as settled can give an unfair advantage to whoever presents it first.
Documentation does not need to become an exhaustive archive of every conversation. The useful record preserves consequential choices, assumptions, and changes. It should help people work together while the programme is active, which also makes it more useful if questions arise later.
When staff expect records to be used selectively, they may become cautious about what they write or seek unnecessary approval before acting. That is a possible response to the conditions, rather than proof of poor character. The organisation should investigate whether its review practices are making accurate information harder to obtain.
When Visibility Becomes Exposure
Transparency can help an organisation understand an incident or delay. Publishing the name of the person closest to the event does not necessarily provide that understanding. A complete explanation needs the relevant decisions, responsibilities, constraints, and corrective work.
In a programme review, a public demand for one employee to account for the result may make other participants reluctant to explain their own decisions. It can also expose the employee to judgement before the facts have been established. The apparent clarity of a single owner can conceal how the work was actually governed.
Review design should reflect its purpose. A learning discussion needs the people and evidence required to understand the event. A management concern about an individual’s conduct needs a fair process appropriate to that concern. Combining them without explaining the purpose can leave participants unsure how their account will be used.
This does not imply that a learning review can promise unlimited confidentiality or disregard serious conduct. It means that expectations should be clear and that findings about a person should follow an appropriate examination of the evidence. A team meeting should not quietly become a substitute for that process.
The specific practice of blameless postmortems described by Google SRE aims to examine contributing conditions and improve reliability. It is useful here as an example of designing a review around learning. It should not be interpreted as a claim that individuals never make mistakes or that conduct concerns cannot be addressed separately.
Apply Accountability Across the Decision Chain
A fair review asks comparable questions of people at different levels. What did each person control? What information was available? Which obligations applied? What options were feasible? How did the person respond when the situation changed?
For the migration sponsor, that includes the decision to retain the launch date after accepting additional scope. For the platform owner, it includes the agreed access work and any warning about its availability. For the delivery lead, it includes the quality of planning, coordination, and reporting within their remit.
The responses need not carry equal responsibility. Differences in authority, knowledge, and action matter. Consistency means using a defensible basis for those differences, rather than assuming that a senior person’s choice is strategic while a junior person’s choice is a performance issue.
The review should also identify what needs changing. The organisation may need a clearer route for approving scope, a more reliable dependency commitment, or support for a skill the lead lacks. Those responses can coexist. Finding a structural issue does not rule out an individual development need, and finding an individual error does not settle the structural questions.
This is the distinction explored in the cost of blame without authority: the assessment should reach the people and arrangements capable of changing the relevant conditions. Replacing a visible owner while leaving the same unresolved decisions in place is an incomplete response.
Check What Changes After the Review
Accountability should produce a clearer account of responsibilities and a credible response. An apology, reassignment, or warning may be part of a particular situation, but none establishes that the conditions leading to failure have improved.
For the migration programme, check whether new scope now receives an explicit decision about dates and capacity. Check whether dependency owners have agreed their commitments and whether the delivery lead can obtain a timely response when those commitments are threatened. Where training or support was agreed, establish whether it happened and whether it helped.
The organisation should also examine whether people can now raise concerns with enough specificity to support a decision. A process that produces fewer reported risks may be improving, or it may be suppressing useful information. The explanation needs evidence from the work.
Accountability becomes credible when people can understand the expectations beforehand, explain their decisions afterwards, and see that the resulting assessment includes the relevant authority and context. Applying that standard consistently gives ownership a practical meaning and makes it harder to use the language of responsibility merely to decide where blame will land.
