This lesson is for mentoring and coaching teams that need to understand development over several sessions. You will build a goal timeline and a follow-up plan. It does not assume that every participant should improve at the same rate or that a fixed number of sessions proves a problem.

1. Agree the goal and what relevant evidence would look like.
2. Connect dated sessions to the correct person and program episode.
3. Separate reported, observed and missing evidence.
4. Review the trajectory and changes in context.
5. Follow unresolved actions using an appropriate review rule.
6. Prepare an accurate summary with the participant’s perspective.

What goal are you following?

Write the goal in terms the participant and mentor understand. “Become more confident” may be a useful starting conversation, but a review needs to know what the person hopes to do differently. For example, a participant may want to prepare and lead part of a team meeting. Confidence, preparation and observed participation are related but distinct evidence.

Choose the review evidence with the participant where appropriate. A brief self-report, a draft agenda and an account of leading a meeting each answer a different question. Collect only what the decision requires. Not every development goal needs a numeric scale.

How should the timeline be structured?

Keep the person identifier, goal, session date, evidence source and next action together. If a person joins another program episode, retain that distinction. Record the actual observation date separately from the date a note was entered.

The postponed meeting is not evidence of stalled ability. The later account is not an independent observation of the meeting. The timeline is useful because those distinctions remain visible.

What does no apparent change mean?

It may mean stability, limited opportunity, an unsuitable measure, missing evidence or a goal needing more time. Review those possibilities before describing someone as stuck. A missing note is a documentation gap, not a measurement of the person.

If the same text is copied into several notes, check whether it represents a new observation or a carried-forward summary. Repetition is not automatically evidence that the same barrier persisted. Preserve dated observations and distinguish them from unchanged background context.

How should unresolved actions be reviewed?

Use a review point suited to the action and circumstances. A promised resource might need checking by its due date; a developmental goal may need a later opportunity to practice. Do not apply an arbitrary “three sessions means risk” rule to every program.

For each open action, ask who owns it, when it was agreed, whether it is still relevant and what happened since. Record a revised date or changed plan rather than silently resetting the action. An urgent concern follows the organization’s established process rather than waiting for a session-count threshold.

Can different locations use different session forms?

Yes. Agree the small common core needed for continuity and any shared reporting question. Local prompts can fit local practice. A data dictionary can map compatible fields such as action status or goal category while preserving the actual local evidence.

Do not average local progress ratings unless the criteria, opportunities and populations make that comparison meaningful. A rating of “3” from one program may mean something different elsewhere. Keep the participant-level narrative useful even when a cross-program number is not defensible.

What belongs in the review summary?

State the goal, period, observed or reported changes, evidence gaps, relevant context and agreed next action. Cite the supporting notes. Where the goal changed, say so. A source-linked summary can acknowledge uncertainty and still help a coordinator prepare.

For the fictional example, the summary could say: “Preparation was observed on 16 March. On 6 April the participant reported leading an agenda item and requested practice handling questions. A further application check is planned.” It should not claim a measured confidence increase or sustained leadership improvement without supporting evidence.

How can Sopact support continuity?

A manually maintained timeline works for a small review. Sopact’s approach connects notes, check-ins and other authorized evidence to continuing records so analysis can use the relevant history. Shared definitions help keep the meaning of fields visible across cycles.

Test whether an AI-assisted summary distinguishes carried-forward context from new evidence and proposed actions from completed ones. Keep source dates, reviewer corrections and access controls available. People interpret the journey and agree the next step; a generated trajectory is not a validated assessment of a person.

Practice: review a delayed opportunity

Use the three-session example. Write a five-line summary and identify the next evidence needed. Then imagine the participant changes the goal in May. Add that change without deleting the earlier goal or describing the change itself as failure.

Frequently asked questions

Is attendance an outcome?

Attendance describes participation. It may matter to delivery, but it does not alone establish learning or growth. Choose evidence related to the agreed goal and retain attendance as context where it helps interpret the journey.

How many sessions should trigger review?

There is no universal number for every goal. Use the action’s due date, the opportunity to act and the organization’s process. Review missing follow-through without assuming that a repeated entry proves the participant is failing.

What if progress cannot be measured numerically?

Use a clear qualitative account with dated evidence and agreed review criteria. Explain what was reported or observed and what remains uncertain. A defensible narrative can be more useful than a score with no consistent meaning.

Can AI summarize the whole history?

It can assist with authorized records, but the summary needs source checks. Test missing dates, contradictory accounts and changed goals. Preserve reviewed outputs and do not assume repeated generation will produce identical wording or judgments.