Find gaps in recorded time

What you need

Your firm-approved AI assistant · prepared files

• Clio Manage with access to the selected matters; Manage AI entitlement only for the optional AI Actions view.
• Timekeeper review is mandatory. Actual durations come only from contemporaneous recorded evidence, never AI estimation.
• Firm-approved Excel; optional Copilot entitlement and tenant approval for the exact minimized data. Native Excel analysis can be done manually if Copilot is absent.
• Keep raw exports and identifying source pointers in a separate restricted location. The AI workbook contains only approved alias-based tables and visible policy/scope text, never raw client records on another tab.
• 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-checked proposed route; source collection and human approval required

1. 1

   Inspect native unlogged-activity suggestions

   Clio Manage AI, optional · Matters > chosen matter > Dashboard > AI Actions card

   Review suggestions for unlogged communications, notes or completed tasks. Open the cited source to check that the activity exists; do not accept or save a suggested entry. No prompt is required for this native proactive feature. Without AI, inspect the matter’s Communications, Notes and Tasks subtabs directly for the bounded period.

   Expected output

   Candidate source activities with actual timestamps and source references, not proposed billable time.

   Next step

   Create a reviewed activity ledger in the next step.

   Check before continuing

   AI suggestions are refreshed overnight and limited in scope. No suggestion and no visible time entry is not proof of missing work.

   Step documentation 1 ↗Step documentation 2 ↗
2. 2

   Check existing recorded time

   Clio Manage · Selected matter > Activities; main Activities table for date/user filtering

   Read matching time entries for the same matter, person and period. Use actual recorded time and work description; distinguish hourly duration from flat-rate quantity. Keep any source link between an activity and entry only when explicit. For details use the entry’s existing detail/Edit view without changing or saving it.

   Expected output

   Reviewed entries with recorded duration, actor and source reference. This route uses bounded manual capture; a general activity CSV export schema is not verified.

   Next step

   Keep source pointers restricted; map alias-based Activities and Entries tables.

   Check before continuing

   Different timestamps or wording do not prove distinct work; a single entry may legitimately cover multiple activities.

   Step documentation 1 ↗Step documentation 2 ↗
3. 3

   Map the observed records to the starter input

   Excel and authorized staff · Separate analysis workbook > Home > Format as Table

   Populate activities.csv with activity_id,matter_alias,actor_alias,activity_datetime,activity_type and actual_duration_minutes_if_recorded. Leave duration unknown unless actually recorded. Populate entries.csv with entry_id, an explicitly known activity_id_if_known, matter_alias,actor_alias,recorded_minutes and a safe narrative_category. Add reviewed period, timezone and coverage exclusions to Scope. Create named tables for Activities and Entries. Put the policy and review scope in visible worksheets or paste them into the next prompt.

   Expected output

   Two staff-created tables retaining actual source evidence, with no invented time.

   Next step

   Check the mapping against each original source before using AI.

   Check before continuing

   Aliases and missing policy fields are staff-created; do not relabel a native export as the kit schema. Unknown stays blank/unknown, not false or zero.

   Step documentation 1 ↗
4. 4

   Review the bounded evidence with Copilot

   Excel Copilot, optional · Approved analysis workbook > Copilot icon at lower right > Chat only

   Use Chat only and submit the prompt. Without Copilot: compare explicit activity IDs, then manually inspect ambiguous matters/persons/dates without adding time.

   Excel Copilot, optional promptCompare Activities and Entries within Scope. First list exact explicit activity_id matches. Then list unmatched activities or ambiguous possible matches as review candidates, citing source row IDs. Return activity_id, candidate_entry_ids, finding and missing_evidence. Never estimate duration, convert timestamps into worked minutes, decide billability or create a charge. A missing activity link is not necessarily missing billed time. Preserve all recorded minutes unchanged.Copy step 4 prompt

   Expected output

   Timekeeper-review candidates, not time entries to post.

   Next step

   Copy the draft exception list to the reviewer; retain row references.

   Check before continuing

   Check every flagged row and the scope. A fluent answer is not evidence of completeness; no automatic workbook or source-system edits.

   Step documentation 1 ↗Step documentation 2 ↗
5. 5

   Approve the review result

   Final human review

   Responsible staff reviewer · Reviewed local worksheet

   The actual timekeeper checks contemporaneous records and decides whether any entry is missing or wrong under the firm’s billing rules. Close a candidate as already captured, unresolved or requiring a separately approved correction.

   Expected output

   Reviewed gap checklist with no inferred billable minutes.

   Next step

   Any external message, source change, payment or deletion requires its own authorized process.

   Check before continuing

   Do not use AI-generated duration or a default minimum charge. The workflow stops before posting time.

   Step documentation 1 ↗

What you’ll get

Approved assistant workspace for matching; output time-review.csv with activity_id, candidate_entry_ids, match_status, duration_evidence, question.