{
  "schema": "vedokrok.public-item.v1",
  "release_id": "MHC-RPUB-20261011-9e4b62c5",
  "url": "/knowledge/practice-the-backup-before-you-need-the-backup",
  "id": "MHC-D-RESEARCH-9226",
  "version": "0.2.0",
  "title": "Practice the backup before you need the backup",
  "summary": "A backup on a map is not a backup until the handoff has survived a realistic attempt.",
  "kind": "protocol",
  "body": "Choose one recurring, consequential workflow with a single point of failure. Let the backup perform a realistic case while the primary observes only at agreed stop points. Capture missing decision cues, access, artifacts and escalation conditions, fix those gaps, then repeat with the primary unavailable. Cross-training studies support shared interaction models and coordination, but much of the evidence comes from simulations and does not imply that every role should be duplicated.",
  "limits": [
    "Do not rehearse on live high-risk systems without safeguards. Licensed, regulated or safety-critical work may require formal authorization and supervised training. Cross-training depth produced inconsistent findings in the experimental literature, and much of the evidence is simulation-based; demonstrated backup readiness does not make people interchangeable experts."
  ],
  "topics": [
    "union-transactive-memory-expertise-routing"
  ],
  "intents": [],
  "evidence": [
    {
      "claim": "Across two computer-simulation experiments with 45 and 49 teams, cross-training enhanced shared team-interaction mental models; coordination mediated the relationship between those shared models and team performance.",
      "role": "supports",
      "note": "The tasks were simulations, and the authors reported inconsistent findings about the depth of cross-training needed to improve shared mental models.",
      "locator": "Abstract",
      "source_id": "RS-85FBF9764660FB3E9E89"
    },
    {
      "claim": "A 2007 meta-analysis found a small-to-moderate tendency for team training to improve team performance, while its analyses suggested that coordination and adaptation training contributed more strongly than cross-training alone.",
      "role": "limits",
      "note": "The evidence base was small; the main analysis treated multiple tests from some studies as independent, although a supplemental seven-study analysis showed the same broad pattern. This does not identify one universally optimal training mix.",
      "locator": "Abstract and supplemental independence analysis",
      "source_id": "RS-F8EB6521EF7C07182E49"
    },
    {
      "claim": "A 2024 scoping review of 213 cross-disciplinary sources on team adaptation recommends realistic training with actual teams and highlights shared mental models, contingency planning, roles and responsibilities as relevant learning objectives.",
      "role": "contextualizes",
      "note": "The review synthesizes heterogeneous experimental, observational and conceptual work. Its training recommendations are not a causal test of the rehearsal protocol in this card.",
      "locator": "Training implications and discussion",
      "source_id": "RS-4B65D6D807889A08CCF5"
    }
  ],
  "use_when": [
    "Your expertise map names a backup, but nobody knows whether that person can actually carry a recurring critical task when the primary is unavailable."
  ],
  "avoid_when": [
    "Do not rehearse on live high-risk systems without safeguards. Licensed, regulated or safety-critical work may require formal authorization and supervised training. Cross-training depth produced inconsistent findings in the experimental literature, and much of the evidence is simulation-based; demonstrated backup readiness does not make people interchangeable experts."
  ],
  "example": "An integration team lists a second engineer as backup for payment-retry incidents. During rehearsal, the backup discovers a hidden dashboard permission and an undocumented rollback threshold. The team fixes the access and records the threshold; on the next drill, the backup can complete the case without the primary.",
  "check": "With the primary unavailable, the backup can either complete the defined case safely or stop at the right boundary and escalate using current artifacts and permissions.",
  "source_ids": [
    "RS-85FBF9764660FB3E9E89",
    "RS-F8EB6521EF7C07182E49",
    "RS-4B65D6D807889A08CCF5"
  ],
  "published_at": "2026-10-01T14:27:00Z",
  "steps": [
    "Pick one task where primary-person absence would materially delay or endanger the work. Define success, stop conditions and the escalation route.",
    "Give the backup the current artifact and a minimal walkthrough, then let them execute a realistic case rather than merely watch the primary.",
    "Record each failure as a concrete gap: missing context, permission, tool, decision cue, artifact or escalation rule. Repair the smallest useful gap.",
    "Retest after the repair or a meaningful system change with the primary unavailable. Keep the formal owner and escalation path explicit."
  ],
  "editorial_overlay": null,
  "editorial": null,
  "notices": [],
  "sources": [
    {
      "id": "RS-85FBF9764660FB3E9E89",
      "title": "The impact of cross-training on team effectiveness",
      "url": "https://doi.org/10.1037/0021-9010.87.1.3"
    },
    {
      "id": "RS-F8EB6521EF7C07182E49",
      "title": "Testing Three Team Training Strategies in Intact Teams: A Meta-Analysis",
      "url": "https://doi.org/10.1177/1046496407304332"
    },
    {
      "id": "RS-4B65D6D807889A08CCF5",
      "title": "Team adaptive capacity and adaptation in dynamic environments: A scoping review of the literature",
      "url": "https://doi.org/10.1016/j.hfh.2024.100089"
    }
  ],
  "relations": [],
  "collections": [
    {
      "id": "RC-943DB1954DE483B80004",
      "title": "Route team knowledge with a living expertise map",
      "url": "/collections/route-team-knowledge-with-a-living-expertise-map"
    }
  ],
  "topic_hubs": [],
  "solution_packs": []
}