Preserve Knowledge — Tacit Capture Playbook

Welcome — Why tacit capture matters

Tacit knowledge—what people do, notice, and decide without writing it down—keeps work moving but disappears when people leave, switch roles, or get too busy to teach. This playbook helps teams make capture practical, ethical, and repeatable so experience becomes reusable rather than fragile.

What this playbook delivers

Concrete, low-friction practices you can use with existing teams: who should run capture, when to schedule it, scripts and prompts for interviews and shadowing, recording and privacy checklists, a simple metadata/tagging scheme for findability, and a short-transfer ritual to accelerate onboarding and handoffs.

Core roles

• Knowledge Owner: The person with the expertise (subject matter expert).
• Capture Lead (Librarian/Cultivator): Facilitates capture, prepares questions, records artifacts, adds metadata and stores materials where teams find them.
• Shadowing Partner / Trainee: Observes and practices; helps surface contextual cues and questions.
• Keeper / Curator: Maintains indexed artifacts, ensures access, and schedules refreshes.

When to capture: timing and triggers

Prefer proactive, short captures versus crisis-driven documentation. Key triggers:

• Planned role transitions, promotions, or exits (start capture at least 4–8 weeks before handoff when possible).
• New initiatives, process changes, or production line launches.
• Recurring incidents or near-misses—capture lessons immediately while memories are fresh.
• Periodic refresh cycles (quarterly or semiannually) for critical roles and practices.

Prioritize what to capture

If you can’t capture everything, prioritize by impact and rarity:

1. High business risk or cost if lost (safety checks, regulatory steps).
2. High frequency but variable execution (customer onboarding steps with many exception paths).
3. Unique individual expertise (tacit heuristics, troubleshooting shortcuts).
4. Hard-to-recreate contextual knowledge (local suppliers, site-specific quirks).

Step-by-step capture workflow

1. Plan — Define scope (what outcome or activity), select roles, schedule short sessions, and name a Capture Lead. Timebox activities to 30–90 minutes per session.
2. Consent & privacy — Run a short consent script (sample below). Agree on what can be recorded and how artifacts will be used.
3. Prepare — Create a one-page capture brief: objectives, prompts, expected artifacts, suggested tags, and delivery deadline.
4. Capture — Use structured interviews, paired observation (shadowing), and short video or annotated screenshots. Prefer multiple short sessions over one long interview.
5. Synthesize — Immediately summarize (10–20 mins) what was learned into a short capture note with key steps, exceptions, and links to raw recordings.
6. Index & store — Add metadata and tags, store in a consistent place, and link from role pages, onboarding flows, or playbook collections.
7. Transfer ritual — Execute a short-transfer ritual with the incoming person: 60–90 minute coached walkthrough, quick quiz or checklist, and a 7‑day shadowing schedule with checkpoints.
8. Review & refresh — Schedule follow-up in 30–90 days to update artifacts after practice and real work exposure.

Consent & privacy — a simple script

Use plain language and respect boundaries. Example:

"Thanks for doing this — our goal is to make your knowledge reusable so the team keeps running smoothly. We'll record audio/video and summarize the highlights. Is it okay to record? Are there any topics, documents, or customer details that must not be recorded or shared? You can review and request edits to any summary before we publish it."

Interview script — prompts that surface tacit knowledge

Start broad, then move to concrete steps, exceptions, and examples.

• What outcome are you responsible for and how do you know it's successful?
• Walk me through a typical day when everything is going well. What do you check, and in what order?
• Show me a real example or evidence of the work (screens, forms, physical checks).
• What are the common surprises or exceptions, and how do you decide what to do?
• Which informal tricks, heuristics, or shortcuts do you use and when not to use them?
• What mistakes have you learned from? Describe one that used to happen often and how you prevent it now.
• Who do you contact for problems you cannot solve? What do you say to get help quickly?

Shadowing checklist — what to observe

• Sequence of actions and timing (note delays, checks, and dependencies).
• Decision points—where the person chooses one path over another and why.
• Artifacts used: documents, tools, dashboards, physical gauges.
• Implicit rules or cues (how they know something is 'off').
• Communication habits (phrases used with coworkers or customers).
• Workarounds and why they're used.

Recording & capture checklist (audio/video/notes)

• Confirm consent on record.
• Good lighting and stable camera (for video); clear audio (external microphone if possible).
• Short segments: prefer five to ten minute clips annotated with timestamps and brief summaries.
• Capture artifacts separately (screenshots, forms, sample emails) and link them to the clip.
• Immediately add a one-paragraph summary and 3–5 bullet ‘must-know’ points while memory is fresh.

Simple metadata and tags for findability

Use a consistent, lightweight metadata model so material is discoverable.

• Title: Role — Activity — Short descriptor (e.g., "Line Tech — Shift Handoff — Critical Checks")
• Role: job title or functional role
• Activity: short controlled vocabulary (onboarding, troubleshooting, maintenance, shift-handoff, supplier-issue)
• Risk/Impact: high / medium / low
• Version/Date: capture date and last-verified date
• Tags: site, machine ID, system, customer, regulatory, quick-win

Short-transfer ritual (30–90 minutes plus 7-day support)

1. Quick context (10 minutes): owner explains purpose, key outcomes, and critical constraints.
2. Walkthrough (20–40 minutes): live or recorded demonstration of the activity and one common exception.
3. Practice (20 minutes): trainee performs the activity with coaching.
4. Commit (5 minutes): trainee names two follow-up questions to check after first shift.
5. Schedule 7‑day checkpoints: brief standups at day 1, day 3, and day 7 to address questions and update documentation.

Storage, retrieval, and housekeeping

• Store raw recordings and a short, searchable summary side-by-side.
• Keep an index (spreadsheet or simple DB) mapping role → essential artifacts → last-verified date.
• Set review cadence for critical roles (every 6 months) and lower-risk areas annually or on trigger.
• Apply access controls for sensitive content and respect privacy requests recorded during consent.

Measures of success

• Ramp time for new hires or role swaps (target reduction: e.g., 25% in 90 days).
• Number of repeat incidents for previously-captured problems.
• Percentage of critical roles with current capture artifacts and a verified date.
• User feedback: usefulness rating after first 30 days on the job.

Common pitfalls and how to avoid them

• Waiting until exit day—start early and make short captures.
• Over-documenting everything—prioritize what matters and prefer concise summaries.
• Keeping raw recordings unreadable—always create a short searchable summary and tags.
• Not honoring consent—publish only what was agreed to and enable redaction or edit requests.

Templates & quick artifacts

Use these as starting points and adapt to your team’s language.

Capture Brief (one paragraph)

Objective: [What outcome]. Scope: [Steps, systems, exceptions]. Owner: [Name]. Capture Lead: [Name]. Deliverables: short summary + linked clips + tag list. Deadline: [Date].

Summary template (to store with recordings)

Title, Role, Date, One-line purpose, 3 must-know bullets, 2 common exceptions, Link to raw recording, Tags, Last-verified date.

Practical experiments to try

• Run a 90-minute capture sprint for one priority role and measure ramp time for a replacement after 60 days.
• Use the short-transfer ritual for the next three hires and collect their "usefulness" rating after two weeks.
• Publish a monthly "captured knowledge highlight" to encourage use and reduce 'documentation debt.'

Next steps for teams

1. Pick one high-impact role and run the full workflow this month.
2. Agree on a single storage location and a short metadata scheme your team will use.
3. Assign a Capture Lead and schedule the first 30‑minute planning session.

Appendix — Quick reference: interview starter phrases

• "Show me what you would do if X happens."
• "When you say ‘check the system’, what exactly are you looking for?"
• "Tell me about a time when this failed—what signs did you see first?"