Choose three measures that reflect the work

A first-month scorecard should answer one question: can this specialist do the work to the standard, with the support the role was designed to have? Choose three measures, not ten. One should cover output, one should cover quality, and one should cover independence. A scorecard with too many measures tells the new person nothing about what matters.

Pick measures that already exist in the workflow, such as completed items, sample pass rate, and the number of questions that required the reviewer to supply missing context. Avoid inventing a productivity score from mouse movements or message counts. The measure should be explainable to the specialist and to the client in one sentence. A measure that cannot be explained that quickly is probably tracking activity rather than the work.

Separate learning speed from output quality

Separate learning speed from output quality during the first month. A new specialist may be slow because they are checking the SOP, which is the behavior you want. If you measure only volume, you teach them to skip the check. Report time-to-first-independent-case separately from steady-state throughput.

Record how long it takes the specialist to complete a familiar task without help. Compare that with the role's expected pace after training, not with the fastest person on the team. If the gap is large, ask whether the process or the training is the cause. Do not assume the new person is the problem.

Sample the first month's work

Sample the first month's work rather than relying on the specialist's own summary. Pull one ordinary case, one exception, and one case the specialist found difficult. Check the source record against the output and the SOP. The sample should show whether the work is correct and whether the reasoning is visible.

Have the reviewer write one sentence for each sample: what was correct, what was missing, and what the specialist should change next time. Keep the notes specific to the work. A sample that produces only a score cannot guide the next month. The written note is the actual training material.

Review with the specialist, not about them

Hold the review with the specialist, not about them. Ask them to walk through the sample and explain the decisions before you give your view. Listen for where the SOP was unclear or the tool got in the way. A new person often sees the process problem that the team has stopped noticing.

Agree on one improvement for month two and write it down. It might be a checklist change, a faster approval path, or a practice set on one exception type. Name who will make the change and when it will be tested. A review with five actions is usually a review with none.

Set the month-two target

Set the month-two target from the evidence in the sample and the specialist's current pace. The target should be achievable without skipping the quality check. If the first month showed clean work but slow handling, the target can focus on speed. If it showed fast work with errors, the target should protect accuracy.

Share the target with the client in plain language, including what will be measured and what will not. The client should know that the first month is a calibration period. Ask the client to flag any promise or priority that the scorecard misses. The target is a shared expectation, not a private quota.

Close the loop with the client

Close the loop by comparing the month-two result with the first-month baseline. If the change worked, keep it and update the onboarding checklist. If it did not, ask whether the measure was wrong or the support was missing. Do not simply repeat the same review with a firmer tone.

File the scorecard, the samples, and the notes with the onboarding record. The next new specialist should inherit a better checklist, not a blank page. The client should receive a short summary of what was learned. A scorecard is useful only if it improves the next start.

Watch for the signals that the role is not ready

A first month can look acceptable while the role is still fragile. Watch for repeated questions that the SOP should answer, work that passes only when the reviewer supplies context, and tasks that quietly return to the client. These signals mean the role design or the training is incomplete. They should change the next month's plan, not just the score.

Bring the signals to the client with the scorecard. A client may need to simplify the queue, provide better examples, or delay a new task type. That conversation is easier in month one than after a customer complaint. The scorecard should trigger a decision, not only a rating.

Carry the scorecard into the quarterly review

Keep the first-month baseline when the specialist moves to normal review. The same measures can show whether the role is stable after the training period ends. A sudden drop in quality may point to a changed queue or a removed support, not to the person. The baseline gives the quarterly review something concrete to compare.

Update the onboarding checklist with what the scorecard revealed. If a measure was never useful, remove it rather than carrying it forward. If a sample caught a problem the score missed, add that sample type. The scorecard is a tool for improving the role, not a permanent report card.

Keep the client summary to one page and name the one change the next month will test. Avoid turning it into a status report the client has to interpret. If the client wants more detail, offer the sample notes rather than a longer summary. The summary should help the client decide, not just inform them.

If the queue and decision rights are already defined but need a steady operator, outsourced operations support can turn approved examples, decision lines, and review rules into a draft your manager can check. The manager still owns exceptions and final approval.

For the next step, compare this workflow with Decide when a delegated task needs a second reviewer and Write a client glossary for a Philippines-based support team. For access and oversight, use the authoritative planning reference alongside your own policy.