Accountability is one of those business terms that can sound unnecessarily severe. In practice, it should mean something fairly straightforward: an important outcome has a clearly identified owner, the owner has suitable authority, and progress can be discussed using information that both parties understand.
As a company grows, informal assumptions about responsibility become less reliable. Two managers may both believe the other is handling a customer issue. A recurring cost may have several people monitoring it but nobody able to decide what should change. More reports and meetings are not automatically the answer.
Start with outcomes, not activities
It is possible for every person to be busy whilst an important customer commitment remains unfulfilled. That is often because individual tasks are visible but overall accountability is not.
A sales manager may own a proposal, an operations manager may own delivery and finance may own invoicing. The customer experiences one continuous relationship. Where a contract passes between functions, the business needs a clear owner for the resulting promise and a reliable way to resolve conflicts between departments.
This does not require every outcome to have its own manager. It means that for the important services and commitments, the organisation can explain who is answerable for their performance, even when several people contribute.
Give authority a defined shape
Accountability without authority is frustrating and frequently ineffective. A manager expected to improve delivery performance needs some influence over priorities, resources and how exceptions are handled. If those decisions must always be approved elsewhere, the business should acknowledge that limitation.
Decision boundaries can be simple. Define what can be agreed independently, what requires consultation and which circumstances demand escalation. Financial thresholds are one example, but operational commitments, staffing decisions, customer remedies and supplier variations can be equally important.
An illustrative professional services firm might give project leads authority to rearrange work within an agreed budget and delivery window. A material change to scope, profitability or contractual risk would be escalated. The intent is to allow normal decisions to move while keeping exceptional commitments visible.
Make measures worth discussing
A KPI can create clarity, but an impressive dashboard does not necessarily improve performance. A measure becomes useful when it answers a decision question and has a credible definition.
For delivery, that might mean the proportion of agreed milestones achieved, the age of unresolved exceptions or the amount of work waiting for customer input. For cash management, it could mean overdue debt, billing delays and forecast shortfalls. These are illustrations, not a universal set of measures. The British Business Bank’s 2026 report provides wider context on the smaller-business finance environment. [2]
It is worth asking where the data comes from, who checks it and what would happen if the figure moved materially. A metric with no owner and no decision consequence can become reporting for its own sake.
Keep management routines short and purposeful
The most valuable management review is not always the longest. A weekly operational meeting can work well with a modest agenda: material commitments, exceptions requiring a decision, actions from the last meeting and the few indicators that help interpret the position.
Where work is stable, a fortnightly or monthly cadence may be adequate. A rapidly changing business may need shorter cycles. The correct frequency is determined by the consequences of delayed decisions, not by a standard template.
A decision log is often more useful than extensive minutes. It records what was decided, why it matters, the responsible person, any deadline and the evidence required to close the action. This provides continuity without obliging the organisation to document every conversation.
Avoid the performance theatre
Targets can produce unintended behaviour when they are applied without context. A team rewarded solely for speed may compromise quality; one measured only on utilisation may avoid the capacity needed to address urgent exceptions. The number itself is not the complete story.
I would use management information as the starting point for enquiry. If performance has changed, ask what lies behind it, whether the measure is reliable and what decision might improve the outcome. A good review should be capable of discovering that the original target was badly designed.
There is also a distinction between accountability and blame. Clear ownership should allow concerns to surface earlier. If every adverse result triggers public criticism, problems are more likely to be concealed until they are difficult to resolve.
When a more formal approach is justified
Regulated obligations, contractual commitments, multiple locations and larger teams may require additional controls. Segregation of duties, independent checks and documented approvals can be sensible where the consequences of error are material.
However, a modest business should not adopt those arrangements simply to resemble a larger organisation. Controls should have a clear purpose, be owned by someone and be revisited when circumstances change. The objective is dependable decision-making, not an impressive collection of policies. The UK government’s SME plan provides wider policy context for the growth of smaller businesses; it does not determine which controls a particular company needs. [1]
For a founder-led company, this kind of discipline can also support delegation. When a manager knows the outcome, authority and review method, the founder gains visibility without becoming the default approval point for ordinary work.
Make room for disagreements
Clear accountability does not mean that every decision will be obvious. A sales leader may want to accept an urgent order that the delivery team believes is unrealistic. Finance may be concerned about credit exposure while the founder sees a strategically important customer. These can be legitimate competing interests rather than failures of cooperation.
The operating model should specify how the conflict is resolved. Which role has authority to accept the trade-off? What information is necessary? Is there a threshold beyond which the decision belongs with the founder or board? Making the route explicit can protect working relationships because people no longer need to rely on informal influence to settle recurring disputes.
The decision and its rationale should be recorded where it is material. Not every conversation requires minutes, but consequential exceptions are worth remembering. They provide a reference for future cases and help reveal whether the agreed policy still matches reality.
Close the loop on improvement
Action registers are easy to create and surprisingly easy to neglect. An action can remain marked 'in progress' for months while the underlying problem persists. I would distinguish an activity from its outcome: introducing a checklist is one task, demonstrating that the handover now works is another.
A useful closure test asks whether the agreed change was implemented, whether it was actually used and whether the intended result was observed. The evidence may be a sample of transactions, a resolved recurring exception or confirmation that an authorised person can now make a particular decision. Where the evidence is inconclusive, a short follow-up period may be appropriate.
This is where management discipline should earn its keep. The process is justified only when it helps the business make or verify a better decision. If reporting becomes the objective, it is time to simplify again.
Sources and further reading
[1] UK Department for Business and Trade, Backing your business: plan for SMEs Read online
[2] British Business Bank, Small Business Finance Markets Report 2026 Read online
Sources accessed 6 November 2026. All examples explicitly described as illustrative are hypothetical.
