When a mandate is flagged as at-risk, the question isn't "what should we do?" — it's "which actions have recovered this specific failure type before, and in what sequence?" That distinction is what separates a Recovery Playbook from a to-do list.
Majhi OS builds and maintains a library of Recovery Playbooks — action sequences matched to specific failure types, ordered by historical effectiveness, and updated as outcomes accumulate. The system doesn't prescribe generic best practices. It executes the specific sequence that has worked for this type of failure on comparable mandates.
The Failure Taxonomy
Before a playbook can be assigned, the failure type must be identified. Majhi OS classifies mandate failures into six primary categories, each requiring a distinct recovery approach.
Outreach Decay
Response rates declining faster than baseline. Cause: usually a combination of outreach fatigue (too many messages to the same population), timing misalignment, or messaging that doesn't match the candidate's current context. Recovery: message recalibration, ICP expansion, channel diversification.
Intake Misalignment
Shortlist approval below 38%. Cause: the candidate profile being sourced doesn't match what the hiring manager actually wants — often because the intake brief captured the job description, not the actual hire criteria. Recovery: structured intake session to reset criteria, historical shortlist review to identify the approval pattern.
Recruiter Overload
Recruiter managing too many active mandates. Cause: load accumulation without rebalancing. Recovery: mandate reassignment or load reduction through prioritization. Speed of rebalancing matters — overloaded recruiter performance degrades fastest in the first 10 days after threshold breach.
Engagement Collapse
Candidates going silent after first contact or first interview. Cause: usually process speed — if competitors are moving faster, candidates exit without explanation. Recovery: interview cadence compression, candidate re-engagement sequence, process bottleneck removal.
Sourcing Exhaustion
Primary sourcing channels saturated. Cause: the ICP definition is too narrow or the reachable population has been fully contacted. Recovery: ICP expansion, adjacent market sourcing, channel shift (e.g., from LinkedIn outreach to referral activation).
Decision Stall
Strong candidates in pipeline but no offer movement. Cause: usually hiring manager hesitation — budget uncertainty, role scope disagreement, or indecision on criteria. Recovery: hiring manager escalation, stakeholder alignment session, offer package modeling.
How Playbooks Are Built
Failure Type Classification
The system identifies the primary and secondary failure types driving the mandate's health score decline. Multiple failure types can be active simultaneously — the playbook is sequenced to address them in dependency order (you can't fix outreach decay if recruiter overload is still active).
Action Sequence Selection
From the Recovery Playbook library, the system selects the action sequence with the highest historical recovery rate for this failure type on comparable mandate profiles. Actions are ordered — some must happen before others to be effective.
Execution and Tracking
Each action in the sequence is assigned, tracked, and timestamped. The system monitors whether actions are completed and whether the relevant signals are improving after each action. If the expected signal improvement doesn't materialize, the playbook escalates to the next intervention.
Outcome Attribution
After the mandate closes — successfully or not — the recovery sequence is attributed to the outcome. Which actions correlated with score recovery? Which didn't move the needle? This feedback loop is what makes the library more accurate over time.
"The playbook library is the operational memory of the system. Every recovered mandate teaches the system something that applies to the next one. That compounding is the moat."
What Autonomous Execution Means Here
For recovery actions that don't require human judgment — outreach message recalibration, interview calendar optimization, candidate re-engagement sequences — Majhi OS executes them directly. Recruiters are only pulled in for actions that require human decision-making: intake resets, offer conversations, escalations. The system handles everything else automatically.