Facilitator guide: Reduce the Precision

Suggested session: 15–20 minutes: 3-minute framing, 6-minute individual rewrite, 5-minute comparison, and 5-minute debrief.

Preparation

Give each learner or pair a fictional rewrite card. Do not ask learners to improve real customer, family, student, employee, account, or health information. A chat tool is optional; this is a paper-first classification and rewriting exercise.

Facilitation priorities

• Ask both what the task needs and whether the learner is authorized to use another person’s information for this purpose.
• Ask learners to preserve useful context rather than producing blank, generic requests.
• Separate precision in the drafting request from precision in the finished artifact; model placeholders that an authorized person fills in afterward.
• Name the difference between reducing precision and deciding that a task belongs somewhere else.
• Make a group-level identification risk visible: several broad details can still point to one person or organization in a small setting.
• Treat uncertainty honestly. Do not replace an unknown fact with a plausible invented one.

Common difficulties

• Learners delete everything: ask what a helpful first draft would still need to know.
• Learners keep an identifier “just in case”: ask how that detail would materially change the requested output and whether it is theirs to use for this purpose.
• The finished artifact needs an exact value: put a placeholder in the drafting request, then have an authorized person fill or verify the value afterward.
• A category is still too identifying: ask whether it identifies someone in this class, workplace, neighborhood, or family.
• The exercise becomes product policy advice: return to the durable principle and record product/account approval as a question for the responsible organization.

Evidence of learning

Look for a request that retains the actual goal, uses a defensible approximation, names at least one tradeoff, and identifies a boundary that a rewrite alone cannot solve.

Accessibility and inclusion

Offer cards as selectable text and read them aloud on request. Let learners work independently, silently, or with a partner. Use labels and written reasoning rather than color alone. Do not require anyone to explain a real private situation or disclose why a particular category feels sensitive.

Provisional timing for facilitator planning

These inherited estimates are starting points for facilitation, not completion targets for learners. Refine them after observing setup, access needs, discussion, and actual participation.

• Lab: 15–20 minutes (short: 10; extended: 30)

---

Source: labs/reduce-the-precision/facilitator-guide.md