How to turn a policy document into SCORM training

Most organisations already own the content their people need to be trained on. It is sitting in a policy document — a data-protection policy, a safeguarding procedure, an anti-money-laundering handbook — usually as a PDF that gets emailed round once a year under a subject line asking everyone to "please read and confirm". The trouble is that confirming you have read something is not the same as being able to do it, and a reply in someone's inbox is not a training record. This guide walks through how I turn a policy document into interactive, scored SCORM training in Intle: what to upload, how to write the brief, how to check the proposed structure, and how to export something you could actually put in front of an auditor.

Why a policy PDF doesn't train anyone

A policy PDF is a reference document wearing a training badge. It has three gaps that matter. First, there is no interaction — the learner scrolls, their eyes glaze, and nothing forces a decision. Second, there is no scoring, so you have no evidence anyone understood the part that counts. Third, there is no completion record your LMS can report on. For a compliance topic that last gap is the expensive one: when someone asks you to demonstrate that your workforce was trained on, say, your obligations under the UK GDPR and the Data Protection Act 2018, an email chain of "read and confirmed" replies is not evidence of competence. You need to show who completed the training, when, and that they met a threshold.

What "good" looks like when a policy becomes training

• Scenarios built from real situations the policy governs — not "what does clause 4.2 say", but "a parent emails you a photo of another child; what do you do?" The decision, not the definition, is the learning.
• Scored checks with a pass threshold the LMS can record and report against, so completion means something.
• Per-question feedback that cites the policy — every answer explanation points back to the specific rule or regulation it rests on, so learners see the reasoning, not just a red cross.
• A completion status and score written back to the LMS through the SCORM runtime, so your evidence trail is automatic rather than manual.
• Facilitator context kept in authoring but stripped from the learner-facing package, so you can annotate the source of truth without leaking notes into the export.

The honest test is whether a learner leaves able to make the right call under pressure, not whether they can recall a clause number. That is why scenario- and assessment-shaped output beats a slide deck for anything with a real-world consequence — see the scenario-based learning examples for how that plays out in practice.

Step 1 — Upload the policy as a source

You do not need to paste the policy in by hand. On the AI compliance training generator, attach the source file and Intle extracts the text for you. The document parsers cover PDF, DOCX, PPTX and plain text, up to 10 MB per file. If your policy lives on an intranet page or a briefing video, you can add a URL instead — a webpage, or a YouTube or Vimeo link — and you can attach reference images too. Whatever you give it becomes grounding for the generation, so the questions and scenarios draw on your actual wording rather than a generic template of the topic.

Step 2 — Write a brief that names the audience and the behaviour

The brief is the single biggest lever you have, and you get up to 5,000 characters for it. The mistake is to describe the topic: "make training on our data protection policy" gives the model nothing to aim at. Instead, name the audience and what they must be able to *do*. Who is this for — front-line staff, line managers, a specific department? And what decision should they get right afterwards? Be concrete about the behaviours you are worried about, because those are the moments the scenarios should recreate.

Step 3 — Review the proposed structure before you generate

Intle classifies your request first — its detection step decides whether the policy is best served as compliance training, a scenario or an assessment — and that classification is authoritative rather than a guess you have to accept. Before any content is written, you get a proposed structure: the sections, where the scored checks sit, which situations become scenarios. This is the cheapest place to intervene. If the plan front-loads too much reading before the first decision, or misses a situation you know trips people up, adjust it here. You can also override the content type if you want an assessment rather than a walk-through, or lean into a scenario for a judgement-heavy policy.

Step 4 — Generate, then edit what it got almost right

Generation runs in the background; processing time varies with source length, output depth, and service load. Under the hood, quality gates check the output before it reaches you — interactivity ratio, visual density, the longest run of non-interactive screens, and variety in how questions are answered. What arrives is a draft, not gospel. Open it in the editor and fix the things a model can only get *almost* right: a scenario that needs your organisation's real reporting line, a feedback note that should quote your exact policy wording, or a threshold that should match your internal standard. This is where your subject-matter expertise earns its place.

Step 5 — Export SCORM or host a session

When it is right, export it. Intle packages target SCORM 2004 4th Edition and SCORM 1.2. The public matrix identifies the exact build behind dated package-level checks and keeps named platforms pending until tested. Check /lms-compatibility, import the package into your exact LMS configuration, and verify launch, score, completion and resume before rollout. If you would rather skip the LMS, host it as a session link instead.

A worked example: a data-protection policy

Take a school's data-protection and acceptable-use policy. Uploaded as the source, with a brief aimed at office and teaching staff, it becomes something like the school phishing and data security example: a learner meets an email that impersonates the head teacher, has to decide whether to release pupil data, and gets feedback that points back to the policy and the underlying duty under the UK GDPR. If your starting point is a blank page rather than an existing document, the compliance templates give you a structured brief to adapt, and the wider compliance training examples show the range — from AML for retail banking to safeguarding assessments framed around Keeping Children Safe in Education.

The pattern holds for almost any policy you already have: the document is your grounding, the brief is your aim, the structure review is your cheap intervention point, and the SCORM export is the evidence you were previously faking with read-receipts. The policy was never the problem. It was the format.

A policy tells people what the rules are. Training proves they can follow them. If your compliance record can't tell the two apart, you don't have a training record — you have a distribution list.— Rebecca Hartley, Compliance Content Lead