AFEST: workplace-based training and Qualiopi compliance explained

An employee who picks up a professional skill alongside an experienced colleague, with no written record of the process, isn’t doing AFEST: they’re doing informal mentoring — not fundable, and not acceptable evidence during a Qualiopi audit. Workplace-based training is a format in its own right, legally recognised since 2018, but its framework demands a rigour of instructional design that providers new to it often underestimate.

What the law says

France’s “Avenir Professionnel” law of 5 September 2018 wrote AFEST into article L6313-2 of the Labour Code as a recognised training format, on equal footing with classroom or distance training. It breaks with the historical separation between the workplace and the training venue: learning happens directly within the professional activity, but framed by a formalised instructional design.

This legal recognition has a direct effect on certification: an AFEST designed or supported by a training organisation falls fully within the scope of a Qualiopi audit, just like classroom or distance training — see our article on distance training (FOAD) obligations for the equivalent logic on the remote-learning side.

AFEST instructional design in four stages

1. Analysis of real work activity. Before designing anything, you need to observe and break down the work situations that carry learning potential, with the target skills explicitly identified.
2. Adapting or adjusting the work situation for teaching purposes: slowing the pace down, deliberately exposing the learner to varied cases, temporarily relaxing certain production constraints.
3. The work-situation exercise itself, overseen by a trainer-facilitator who observes and steps in if needed.
4. Reflective sequences, distinct from the exercise itself: dedicated time where the learner and the facilitator revisit what was just done, to make the learning explicit and identify what still needs consolidating.

It’s this fourth stage that most clearly sets AFEST apart from plain on-the-job training: without an organised, tracked reflective sequence, there is no AFEST in the sense the framework requires.

A format studied by education research

AFEST has been the subject of French research, particularly drawing on “didactique professionnelle” (professional didactics). A study by Anne-Lise Ulmann and Béatrice Delay, published in 2024 in the journal TransFormations (“L’action de formation en situation de travail (Afest): révélateur troublant d’une nouvelle épistémologie de la formation?”), examines how AFEST challenges the classic reference points of instructional design by treating the analysis of activity — rather than pre-defined content — as the raw material of learning (read the article on TransFormations). This research confirms that the quality of an AFEST depends less on the teaching materials than on the facilitator’s ability to organise genuine reflection on the lived activity.

Evidence expected during a Qualiopi audit

Stage	Evidence expected
Activity analysis	Analysis report on the work situations selected and the skills targeted
Instructional design	Document formalising the exercise sequences and the reflective sequences
Delivery	Dated attendance records and reflective-sequence summaries
Assessment	Skills assessment grid, consistent with the operational objectives set beforehand

Indicator 28 of the National Quality Reference Framework, covering “workplace-based training and apprenticeship,” is the direct entry point for an auditor assessing an AFEST. It’s often combined with indicator 21 on the skills of those delivering training: an AFEST facilitator must demonstrate a specific competence in activity analysis, distinct from mere subject-matter expertise.

Funding and administrative conditions

An AFEST is funded like any other training action — skills development plan, OPCO, FNE-Formation depending on the scheme in force — provided it is properly formalised in a compliant training programme and agreement. Missing formalisation is the leading cause of funding refusal: a funder cannot approve an action it cannot distinguish from ordinary production time.

The most common mistakes

• Confusing AFEST with informal mentoring: without organised, documented reflective sequences, the action cannot be recognised as training.
• Skipping the prior analysis of the activity, which alone identifies genuine teaching potential in the chosen work situation.
• No identified or trained facilitator, even though this role is distinct from that of a plain subject-matter expert.
• No written evidence of the reflective sequences, still the element most often missing during audits.

Take action

The Complete Qualiopi Kit includes design templates and evidence for indicator 28, alongside the other 31 indicators of the framework. To build your training offer from the moment you launch your organisation, the ebook “Create Your Training Organisation in 30 Days” walks through the approach step by step, and the complete pack brings both resources together.