Operations & ReportingProposed workflow

Draft an inbox handoff queue

Suggest staff, attorney, waiting, and reference queues from approved routing rules.

Tools for the documented steps

These products support separate steps. Their presence does not imply a tested connection or endorsement.

Practice relevance

Cross-practice idea · A general operational method. Review the stated tool and jurisdiction limits before use.

The problem

Suggest staff, attorney, waiting, and reference queues from approved routing rules.

Why it matters: A crowded inbox can obscure work that staff could handle and messages that need an attorney. A reviewed routing queue makes those decisions explicit without automatically hiding, forwarding, or completing messages.

Sources & evidence

Proposed / not tested. The public source describes the problem. Documentation supports only the stated method components; no result from this recipe has been verified.

What’s been verified

Source methods and reported problems have been reviewed. We have not run the adapted workflow. The sample below describes expected behavior on synthetic data, not a customer result.

What to measure: In a synthetic pilot, count correctly found exceptions, false alarms, and unresolved inputs. No time or cost saving is assumed.

  • Pain evidence: Emails are increasing my blood pressure ↗Emails are increasing my blood pressure, 2025-09-16, older. A solo litigator describes unsustainable email volume and explicitly wishes Outlook could send suitable items to staff. Several responses describe batching, delegating, waiting and task categories. Direct open failed; indexed original text was available.
  • Method: Microsoft Outlook Quick Steps ↗Microsoft Outlook Quick Steps documents categories, flags and repeatable message actions. These native actions are available building blocks; automatic legal-urgency classification is not an established feature.

Source checked: October 15, 2026. Pain accounts are self-reported. Documentation supports components; this adapted recipe and prompt have not been tested. We wrote these prompts. The adapted workflow hasn’t been tested.

Suggested approach

Start when: An approved inbox-review window

  1. classify selected deidentified message summaries against firm rules
  2. Keep source references and report unknowns instead of filling gaps.
  3. The responsible reviewer checks the report before any separately approved action.

Limits and review

The assistant drafts or reports only. Check source records and review the output before a human performs any separately approved send or change. No automatic legal decision, money movement, account connection or publication is included.

Who reviews it: Attorney validates routing before it affects the inbox; all uncertainty remains visible. Test a familiar sender discussing a new matter, an innocuous subject containing a deadline and a misdirected message.

What you need

Your firm-approved AI assistant · prepared files

  • Existing firm-approved Microsoft Copilot work access in Outlook with permission for the chosen threads; account/tenant/feature access must be checked.
  • Bounded manually selected thread list, lawyer-approved routing policy and alias role/owner matrix. This is not a claim of exhaustive inbox discovery.
  • Preserve original evidence unchanged in restricted storage. Use synthetic material first. Prepare a separate, explicitly approved minimized AI input; aliases alone do not anonymize narrative or attachments. Do not upload raw originals or an alias key merely because another tab/file is sanitized.
  • Use synthetic inputs first. Keep real data in firm-approved accounts and processing, with appropriate access, confidentiality and retention controls. Aliases alone do not anonymize records. Never paste credentials, API keys or unnecessary client details.

How to use

Documentation-verified component path; proposed adaptation, not tested

Before you start
  • Summary by Copilot is native thread summarization; it does not establish mailbox completeness or safe delegation. The exact classification prompt below is an editorial adaptation.
  • No output changes folders, rules, importance, completion status or ownership automatically. Missing source references or ambiguous/consequential requests require attorney review.
Check access and setup
  • Actual account access, source versions/completeness, privacy approval and reviewer sign-off were not observed.
  1. Read the selected thread with native AI

    Outlook Copilot · Open selected conversation > Summary by Copilot (or Summarize)

    Use only a thread approved for the firm’s existing Copilot processing. Review the generated summary and open available numbered citations to the source messages. Repeat only for the bounded thread list and record capture time. If Copilot is unavailable, read the messages manually; do not claim an AI route ran.

    Expected output
    Thread summary grounded in the selected conversation, subject to human completeness checking.
    Next step
    Use the verified thread identity and approved routing policy in the Outlook Copilot chat step.
    Check before continuing
    Review all messages in the thread for omitted requests or context; native summaries are not evidence of absence.
  2. Ask for a source-linked routing proposal

    Outlook Copilot Chat · Copilot icon in Outlook navigation header > chat prompt field

    Paste the exact thread reference and the approved client-free routing policy/role matrix into the prompt. Keep the request scoped to this chosen thread. If the chat cannot resolve the exact thread or returns uncheckable source references, stop that row and perform manual review.

    Expected output
    Proposed routing/action/owner queue for the selected thread.
    Next step
    Copy the answer to a new local Word review document; preserve actual source-message references in restricted review and use aliases in any broader summary.
    Check before continuing
    Verify every proposed action against the original message; do not adopt apparent urgency or a legal deadline from an AI inference.
  3. Human review and final handoff

    Final human review

    Responsible staff reviewer and lawyer/process owner · Local reviewed report beside the frozen source set

    Attorney validates every proposed handoff and privacy scope. Staff records only approved action/owner/source reference in the local handoff document; if a later task or message is authorized, verify its completion separately.

    Expected output
    Reviewed inbox handoff queue for the explicitly selected threads.
    Next step
    No folder/rule changes or communications occur in this recipe.
    Check before continuing
    All unresolved or unreadable evidence stays open. No inferred approval or external action follows from AI output.

What you’ll get

inbox-review.csv: message_alias, proposed_queue, explicit_action, proposed_owner, rule_id, evidence, uncertainty

Inputs and a sample test Synthetic example · not a measured result

Input files

  • messages.csv: message_alias, deidentified_summary, sender_role, received_at, explicit_request, source_ref
  • routing-policy.csv: rule_id, condition, allowed_queue, proposed_owner, attorney_review_trigger

Synthetic example

M1: routine request with explicit staff-routing rule
M2: familiar sender but new matter
M3: deadline-related message

Expected output

M1: proposed staff queue with rule
M2: ambiguity for attorney
M3: attorney review, no deadline calculation

Test a failure

Attorney validates routing before it affects the inbox; all uncertainty remains visible. Test a familiar sender discussing a new matter, an innocuous subject containing a deadline and a misdirected message.

Try with the sample files

Follow the tool steps above and use the inputs named in each step. Where provided, native-* files are blank scaffolds and NATIVE-INPUT-MAPPING explains their fields. Original examples and the optional full review prompt use a separate prepared-input schema. Keep them separate unless the mapping explicitly connects them. Downloading a kit does not connect accounts or prove the workflow works.

Download inputs and checklist 8 files

Instructions, blank templates and separate synthetic examples; these files do not install an automation. Blank templates still need verified source data.

Optional full review prompt for prepared inputs

Prepared-input review prompt with Your firm-approved AI assistant · prepared files

Proposed / not tested · editorial prompt. This prompt uses the original prepared-input schema shown in the sample test. Step prompts may use different tables. Follow NATIVE-INPUT-MAPPING where supplied; do not mix the original examples with native-step inputs without mapping and checking every field.

Read original method Read pain evidence

Published by Law Firm Agents. These are source-linked editorial adaptations; no end-to-end execution is claimed. Read the plain-text version.