Social media training strategy framework: six decisions from work problem to review

A practical social media training strategy rests on six decisions: the work problem, the eligible learners, the assessed task, the delivery route, the controls and the continuation rule. Settle them in this order. Each one limits the next, and none can be skipped.

What to take away

1. Pin down the work problem

Name the workplace, the role and the recent task that failed or became risky. Keep the artefact, the date and your analysis of the cause. A vague worry about "skills" cannot be assessed later, so it cannot justify a purchase either.

Write the no-course alternative next to the proposed course. A corrected approval workflow, a clearer line of authority or safer account access may fix the problem without any teaching. If it does, stop here and save the budget.

2. Define who is eligible

State each learner's job responsibility, prior knowledge, location in England, access needs and exclusions. Do not swap a broad labour-market category for a named learner list. A list of job titles tells you nothing about who actually performs the failed task.

Record invited, ineligible, declined, withdrawn, attended and assessed people separately. The denominator for completion is eligible starters. The denominator for competence is every eligible assessed attempt, with missing attempts reported rather than silently removed.

3. Lock the assessment first

Describe one output that resembles the real work without using a production account. The learner never touches a live brand profile during practice.

A learner might create a synthetic post, attach evidence for an objective claim, add an advertising label where one is required, write accessible text alternatives and follow an approval route. Define the permitted references, the assessor, the rubric, the pass rule and the reassessment path before anyone enrols.

4. Match delivery to the constraint

Compare in-person, live remote, asynchronous and coached-workplace routes against the same task. Record learner time, trainer and assessor time, travel, technology, adjustments, support, authentication and recovery. Our delivery routes article sets out the four options side by side.

Choose the least burdensome route that can still produce the required evidence. A blended route needs a stated reason for every component. If a component adds cost but no evidence, drop it.

5. Refuse gate trading

Create separate approvals for learner-data handling, electronic marketing, account security, advertising claims, accessibility, education status and procurement. Passing one review does not clear the others. A strong security result cannot make up for a missing privacy decision.

For the data gate under UK GDPR, you must have a valid lawful basis to handle personal information (see the lawful basis guide), while PECR and the Equality Act 2010 section 20 sit behind the marketing and accessibility gates. Our risk scenarios show how each gate can fail in practice.

Two other gates have their own legal homes. Electronic marketing gets its own sign-off because the rules on electronic marketing apply alongside data protection law. The accessibility gate asks whether reasonable steps were taken to avoid substantial disadvantage to disabled people.

Keep synthetic identities, data and brands in every rehearsal. A gate owner records approve, reject or approve with conditions, plus the evidence, the expiry and the retest trigger.

6. Decide with bounded evidence

Set the review date before delivery. Report activity, assessed competence and workplace observations in separate columns. Merging them hides which layer failed. When you plan the review, consider what matters when designing an evaluation, which is what the Magenta Book covers.

The decision owner may expand, revise, keep the present method or stop. Our measurement and reporting guide shows how to lay out the columns. A ninety-day plan turns the decision into a capped pilot.

Illustrative example

The numbers here are invented and show the arithmetic only. Suppose 20 eligible learners start, 16 complete the activity and 12 of 15 assessed attempts meet every mandatory rubric criterion.

Completion is 16 out of 20, or 80 per cent. Competence is 12 out of 15, or 80 per cent. One completer has no assessed attempt, and that gap is reported, not removed.

Before you act