Plan the Compliance Training
When To Use
Compliance training is a box to tick and the quiz record is kept as evidence against people. Everyone clicks through the yearly module, and nothing about the choices they make at work is any different afterwards. Run this to plan a mandatory programme or the refresh of one. It answers: on the job, which few decisions does each rule turn on, how do people practise them, and what is the record for?
When Not To Use
If you need questions for one course, use Write the Knowledge Check; to schedule retrieval for any other course, use Space the Practice. What a rule requires, what must be recorded and for how long are not training questions: check with a qualified adviser before planning anything.
Inputs
- The policy or rule text, as its owner gave it to you.
- Where people got it wrong or asked for help, with names removed: incident themes, helpline themes, audit findings.
- The current module and check, and how completion is recorded today, as you tell me.
If you have none of this, I start from the policy text and mark the output as a first draft, with every decision "to confirm with the policy owner".
Approach
Two methods carry this: practice on real problems (Merrill, 2002), so each rule becomes decisions practised in situations, and retrieval practice (Roediger and Karpicke, 2006), so the refresh asks people to decide again instead of rewatching. The retrieval research was done with students and study texts, so read it as a direction, not a promise. The judgement is to cut each rule down to the moments where people hesitate. The failure it prevents: the policy is pasted in clause by clause, the one situation that really trips people up gets a single line near the end, and the only thing the organisation can show afterwards is who clicked.
Workflow
- Ask at most three questions: which rules are in scope and who owns each, where have people got it wrong or asked for help, and what is the completion record used for today?
- Write the record-use statement before any content: the record shows that training was offered and completed. It is never used to grade, rank or build a case against a person, and answers are reported by item and in aggregate, with groups below a size you set suppressed. If the request is to design quiz records as disciplinary evidence, I refuse that part and say why.
- Per rule, list the few on-the-job decisions where people get it wrong, in the words of the work, not of the clause. The policy owner confirms each one; you take the list to them. The rest of the rule goes to a look-up reference.
- Write one scenario check per decision: a realistic situation, options built from the real mistakes, and feedback for every option that says why. Retakes are allowed; the check is practice, not a gate on the person.
- Plan the refresh as short decide-again prompts at intervals you set, in place of a rewatch of the whole module. Where a rule sets a frequency: check with a qualified adviser.
- List every legal, regulatory, retention and privacy point in one place for the policy owner, each marked "check with a qualified adviser". None is stated as fact in the plan.
Output Format
# Compliance Training Plan
Programme: [name] | Rules in scope: [list] | Policy owner: [role] | L&D owner: [role]
## Record use
The record shows: [training was offered and completed, on which date] | Item results are used for: [fixing the course; by item and in aggregate, groups below [size you set] suppressed]
Never used for: grading, ranking or building a case against a person.
## Decisions each rule turns on
| Rule | On-the-job decision | Where people go wrong | Where that came from | Confirmed by |
|---|---|---|---|---|
| [rule] | [decision, in the words of the work] | [common mistake] | [incident theme, helpline theme, audit finding] | [policy owner, date] |
Moved to the look-up reference: [parts of the rule]
## Scenario checks
| Decision | Situation | Options | Feedback for each option |
|---|---|---|---|
| [decision] | [realistic situation] | [options from real mistakes] | [why each is right or wrong] |
## Refresh
| Interval | Prompt format | Decisions covered |
|---|---|---|
| [interval you set] | [decide-again prompt] | [decisions] |
## For a qualified adviser
- [legal, regulatory, retention or privacy point] (check with a qualified adviser)
## Decision
[Policy owner] confirms the decisions by [date]; [L&D lead] and [legal or compliance adviser] sign the record-use statement by [date], before the build starts.
Done When
- The record-use statement is written and says what the record is never used for.
- Every rule has at least one on-the-job decision confirmed by its owner, and every decision has one scenario check with feedback for every option.
- Every legal or regulatory point sits in the adviser list, not stated as fact.
Quality Bar
- Decisions come from real mistakes, never from the order of the clauses in the policy.
- No legal requirement, frequency, pass mark or retention period is supplied by Claude; each one says "check with a qualified adviser" or is set by you.
- Results are read by item and in aggregate, no individual's answers go to a manager, and scenarios are fictional, never an incident a colleague could recognise.
- The record shows that training happened; it is never used to grade, rank or build a case against a person, and the policy owner confirms every decision.
Next
Run ldg-set-the-team-ai-rules (Set the Team AI Rules) to agree the last standing rule the team needs.
About the makers
This pack is made by 90n (90n.ai): AI-native ways of working and tools for Product, Marketing and HR teams, built by ex-McKinsey founders. We map where time and money leak, redesign the work around AI and stay until your people run it themselves, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).
1---2name: ldg-plan-the-compliance-training3description: Plans a mandatory training programme around the choices people make at work, producing the few on-the-job decisions each rule turns on, a scenario check per decision, a refresh by spaced retrieval, a record-use statement and the legal points listed for a qualified adviser. Use for "run ldg-plan-the-compliance-training", "compliance training", "mandatory training", "annual compliance training", "make compliance training engaging", "compliance refresher", "compliance quiz records", part of The Complete Claude Guide for L&D Pack by 90n.4metadata:5 source: https://skillmd.com/skills/polar-bear-org/ldg-plan-the-compliance-training6---78# Plan the Compliance Training910## When To Use11Compliance training is a box to tick and the quiz record is kept as evidence against people. Everyone clicks through the yearly module, and nothing about the choices they make at work is any different afterwards. Run this to plan a mandatory programme or the refresh of one. It answers: on the job, which few decisions does each rule turn on, how do people practise them, and what is the record for?1213## When Not To Use14If you need questions for one course, use Write the Knowledge Check; to schedule retrieval for any other course, use Space the Practice. What a rule requires, what must be recorded and for how long are not training questions: check with a qualified adviser before planning anything.1516## Inputs17- The policy or rule text, as its owner gave it to you.18- Where people got it wrong or asked for help, with names removed: incident themes, helpline themes, audit findings.19- The current module and check, and how completion is recorded today, as you tell me.20If you have none of this, I start from the policy text and mark the output as a first draft, with every decision "to confirm with the policy owner".2122## Approach23Two methods carry this: practice on real problems (Merrill, 2002), so each rule becomes decisions practised in situations, and retrieval practice (Roediger and Karpicke, 2006), so the refresh asks people to decide again instead of rewatching. The retrieval research was done with students and study texts, so read it as a direction, not a promise. The judgement is to cut each rule down to the moments where people hesitate. The failure it prevents: the policy is pasted in clause by clause, the one situation that really trips people up gets a single line near the end, and the only thing the organisation can show afterwards is who clicked.2425## Workflow261. Ask at most three questions: which rules are in scope and who owns each, where have people got it wrong or asked for help, and what is the completion record used for today?272. Write the record-use statement before any content: the record shows that training was offered and completed. It is never used to grade, rank or build a case against a person, and answers are reported by item and in aggregate, with groups below a size you set suppressed. If the request is to design quiz records as disciplinary evidence, I refuse that part and say why.283. Per rule, list the few on-the-job decisions where people get it wrong, in the words of the work, not of the clause. The policy owner confirms each one; you take the list to them. The rest of the rule goes to a look-up reference.294. Write one scenario check per decision: a realistic situation, options built from the real mistakes, and feedback for every option that says why. Retakes are allowed; the check is practice, not a gate on the person.305. Plan the refresh as short decide-again prompts at intervals you set, in place of a rewatch of the whole module. Where a rule sets a frequency: check with a qualified adviser.316. List every legal, regulatory, retention and privacy point in one place for the policy owner, each marked "check with a qualified adviser". None is stated as fact in the plan.3233## Output Format34```markdown35# Compliance Training Plan36Programme: [name] | Rules in scope: [list] | Policy owner: [role] | L&D owner: [role]37## Record use38The record shows: [training was offered and completed, on which date] | Item results are used for: [fixing the course; by item and in aggregate, groups below [size you set] suppressed]39Never used for: grading, ranking or building a case against a person.40## Decisions each rule turns on41| Rule | On-the-job decision | Where people go wrong | Where that came from | Confirmed by |42|---|---|---|---|---|43| [rule] | [decision, in the words of the work] | [common mistake] | [incident theme, helpline theme, audit finding] | [policy owner, date] |4445Moved to the look-up reference: [parts of the rule]46## Scenario checks47| Decision | Situation | Options | Feedback for each option |48|---|---|---|---|49| [decision] | [realistic situation] | [options from real mistakes] | [why each is right or wrong] |50## Refresh51| Interval | Prompt format | Decisions covered |52|---|---|---|53| [interval you set] | [decide-again prompt] | [decisions] |54## For a qualified adviser55- [legal, regulatory, retention or privacy point] (check with a qualified adviser)56## Decision57[Policy owner] confirms the decisions by [date]; [L&D lead] and [legal or compliance adviser] sign the record-use statement by [date], before the build starts.58```5960## Done When61- The record-use statement is written and says what the record is never used for.62- Every rule has at least one on-the-job decision confirmed by its owner, and every decision has one scenario check with feedback for every option.63- Every legal or regulatory point sits in the adviser list, not stated as fact.6465## Quality Bar66- Decisions come from real mistakes, never from the order of the clauses in the policy.67- No legal requirement, frequency, pass mark or retention period is supplied by Claude; each one says "check with a qualified adviser" or is set by you.68- Results are read by item and in aggregate, no individual's answers go to a manager, and scenarios are fictional, never an incident a colleague could recognise.69- The record shows that training happened; it is never used to grade, rank or build a case against a person, and the policy owner confirms every decision.7071## Next72Run ldg-set-the-team-ai-rules (Set the Team AI Rules) to agree the last standing rule the team needs.7374## About the makers7576This pack is made by 90n ([90n.ai](https://90n.ai)): AI-native ways of working and tools for Product, Marketing and HR teams, built by ex-McKinsey founders. We map where time and money leak, redesign the work around AI and stay until your people run it themselves, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).