A recurring mistake is more likely to be a process problem when different people make it at the same stage of work. It may be an individual performance issue when it remains isolated to one person across comparable situations.
Neither pattern proves the cause. Before deciding, check where the error began, what information was available, how the work was handed over, who was responsible and whether the person could get the guidance they needed.
Follow the mistake to where it began
The place where a mistake becomes visible is not always where it started.
An incorrect invoice may appear to be an invoicing error. The actual cause could be an incorrect entry during order processing or an approved price that was omitted during the sales handover.
Start with one recent case and trace the work backwards. Identify the first point where the actual work differed from the expected result.
Use observable facts. For example, “Three orders were entered without a delivery date” describes what happened. “The employee does not care about details” assigns a motive without evidence.
Also check whether the error is linked to a particular customer, source document or exception. The standard process may work correctly while unusual cases remain undefined. If there is no clear pattern, collect more examples before changing the process or assessing an individual.
Check five conditions before deciding
Use the same five checks for each case. This gives you a consistent basis for comparison.
- Repetition: Who else has made the same mistake? Does it occur at one stage, with one type of work or only with one person?
- Inputs: Did the person receive complete, correct and consistent information? Missing customer details or an outdated price list can create errors later in the workflow.
- Handoffs: Was it clear what was being passed on, in what state, to whom and by when? Repeated errors between roles often indicate a weak handover.
- Responsibility: Was someone explicitly responsible for completing the work, checking it and handling exceptions? Shared or unspoken responsibility can leave decisions unowned.
- Access to information: Could the person find the current procedure, use the required data and get an answer in time? Guidance may exist but still be outdated, difficult to find or unavailable when needed.
Only move to an individual performance assessment after checking these conditions. The key question is whether the person had a reasonable opportunity to complete the work correctly.
Compare like with like
A fair comparison requires similar conditions. Two employees may appear to be doing the same task while working with different deadlines, workloads, source information or exceptions.
Before comparing their results, ask:
- Did they receive the same type and quality of input?
- Were the instructions current and equally clear?
- Did they have similar deadlines and workloads?
- Could they contact the same people for help?
- Were they dealing with standard cases or unusual exceptions?
Differences do not automatically excuse a mistake. They do affect what the comparison can tell you.
Relevant differences can also include accessibility needs, experience with the task or a misunderstanding of the expected result. Check these before concluding that an isolated pattern is a performance problem.
Keep a short record of comparable cases
Memory tends to preserve unusual incidents while losing the conditions around them. A short record keeps the discussion focused on evidence.
| What to record | Question to answer |
|---|---|
| Result | What was wrong, late, missing or incomplete? |
| Point of origin | Where did the work first differ from the expected result? |
| Inputs | What information was available at that point? |
| Comparison | Who completed the same step under similar conditions? |
| Responsibility | Who owned the decision, check or exception? |
| Information and support | Could the person find current guidance and get help in time? |
Use forms, messages, approvals and system records where available. These usually provide a more reliable account than recollections collected after the event.
The record does not need to become a large reporting exercise. It only needs enough detail to show whether the mistake follows a person, a process step or a particular type of case.
Test the diagnosis against contrasting cases
Suppose customer orders repeatedly reach operations without a confirmed delivery date. Several coordinators handled the affected orders, but every order arrived from sales before the delivery requirement had been confirmed.
The checklist requires delivery details, yet it does not say who must confirm them or what should happen when an order is incomplete.
The handover is the right place to begin. A suitable response would define the required fields, assign responsibility for confirmation and specify when an incomplete order must be returned. Retraining one coordinator would not close that gap.
Now suppose four employees process the same type of order. They use the same current instructions, receive comparable source information and can consult the same supervisor. One employee repeatedly omits the delivery date even when it is clearly provided.
The manager should still check for relevant differences, such as workload, accessibility needs or a misunderstanding. If the conditions remain comparable and the problem is isolated, an individual performance conversation is appropriate.
That conversation should cover the expected result, specific examples and the employee’s method for completing the step. Any agreed support should be followed by a review of later work against the same expectation.
Choose the response that matches the evidence
The diagnosis usually leads to one of three next steps:
- The error follows a process step: Investigate the inputs, handover, instructions, responsibility and treatment of exceptions at that point.
- The error follows one person across comparable cases: Discuss the specific performance gap, confirm the expectation and agree on appropriate support and review.
- The pattern is unclear: Record more cases before redesigning the process or drawing conclusions about an individual.
A process issue and an individual performance issue can exist at the same time. Unclear instructions may increase errors across a team while one employee still performs less consistently than colleagues under the same conditions.
Treat these as separate findings. Clarify the process for everyone, then address any remaining individual pattern using comparable evidence.
Frequently asked questions
How many examples are needed before deciding?
There is no fixed number that proves the cause. Look for repeated cases completed under comparable conditions. If incidents involve different inputs, deadlines or unusual exceptions, gather more evidence before drawing a conclusion.
Can training solve a process problem?
Training can help when people do not understand a clear and workable process. It will not correct missing inputs, contradictory instructions or undefined responsibility. If several trained employees continue to make the same mistake at the same stage, investigate how that part of the work is organised.
What if the employee and manager disagree about where the mistake started?
Trace the work through available forms, messages, approvals and system records. Identify the first point where the actual work differed from the expected process. This keeps the discussion focused on evidence rather than competing memories.
What should a manager check first?
Start with one recent example and identify where the error began. Then check who performed the same step under similar conditions, what inputs were available, how the work was handed over, who owned the decision and whether current information and timely help were available.