ReviewDesk
← Back

ReviewDesk guide · By the Ingenious AI team · Published 14 November 2026 · Reviewed 14 November 2026

What should happen when a human-review request goes overdue?

Generate a free adverse-outcome letter — no sign-up Build a free pre-use notice See pricing

The short answer. The case stamps itself overdue, surfaces to the owner and leadership views, gains an escalation note stating what blocks it and what happens next, and — if the assignee is the block — moves to a new owner with its full history. The applicant sees the escalation through their status link rather than discovering silence. ReviewDesk computes escalation on read from the case's own deadline, so an overdue case cannot hide between reports.

What is review-request escalation?

Escalation is the managed response to a review case approaching or passing its deadline: visibility raised, ownership reviewed, blockers named, the applicant informed. In ReviewDesk's SLA workflow, escalation states are computed on read from the case's deadline with provenance, and escalation events are emitted once per state change, so the record shows exactly when the case escalated and why.

Why compute escalation on read?

Because batches lie by omission. A nightly job that flags overdue cases cannot flag the case that becomes overdue at noon and is complained about at two. On-read computation makes the case's state a function of now: every list, every open, every status check re-derives urgency from the deadline and the current time. There is no stale view in which an overdue case looks fine. ReviewDesk's workflow was built this way in its Phase-8 SLA work — escalation computed on read, emitted once — and the pattern deserves copying whatever platform a team uses.

Emitted once matters as much. An escalation that re-fires on every read becomes noise, and noise trains people to ignore the signal entirely. The state change is the event: the case entered escalation at this time, for this reason, seen by these roles. After that, the case wears its stamp until it is decided; the decision closes the story with the lateness on the record.

What should trigger escalation before the deadline?

The final-third rule is a sound default: when a case has consumed two-thirds of its window without a decision, it escalates to the owner's attention — not as an emergency, but as a question: is this case moving, and does its assignee need help? Cases waiting on external evidence escalate differently from cases waiting on a reviewer; the escalation note should say which, because the remedies differ. Evidence-wait escalation triggers chasing; capacity escalation triggers reassignment or reinforcement.

Some cases should escalate at birth: high-consequence decisions — a home, a job start date, care access — where the person's window is shorter than the statutory one. Configure those decision types with tighter internal clocks from the start, per the clocks guide, rather than relying on heroes to notice urgency in a queue of equals. Escalation designed is escalation that rarely has to shout.

What does the escalation note need to say?

Three things, in the reviewer's or owner's own words: what is blocking the case, what is being done about it, and when the applicant will next hear — a date, not a sentiment. 'Awaiting county record confirmation, chased 12 May, applicant updated, next update 19 May' is an escalation note. 'In progress' is a status wearing a note's clothing. The note joins the case history permanently; when the case is later read — by leadership, by the applicant's adviser, by your own future selves — the note is the difference between a delay that was managed and a delay that merely happened.

Write notes as if the applicant will read them, because through the status link's summary they effectively do. Professional, factual, free of internal shorthand and blame. The officer who writes 'vendor slow again' learns to write 'third-party record confirmation outstanding; chased twice' — the second version is better management information anyway, which is the usual reward for writing for the reader.

How should reassignment work when the assignee is the block?

Cleanly, completely and without stigma. The case moves with its file, clock, provenance and history to the new owner as a recorded event; the original assignee's queue relieves; the applicant's status view shows the case in review, undisturbed. Reassignment fails when it is treated as a verdict on the first reviewer — teams then hide stuck cases instead of surfacing them, and escalation dies as a practice. Frame it as load management, audit it as load management, and thank the people who escalate their own cases early; they are the system working.

The assignment system in the assignment guide supports the pattern: named owners, visible queues, moves recorded. What it cannot supply is the culture — a weekly owner review where escalated cases are discussed as operational facts, blockers removed in the room, and nobody's competence is inferred from a case that waited on a county records office for three weeks.

How do you talk to the applicant whose case is late?

Early, plainly and with a date. The worst communication strategy for lateness is silence until completion; the applicant experiences unmanaged time as disregard, and composes their own explanation — usually a worse one than the truth. A short update beats a long apology: your case is still in review, here is what we are waiting for, here is when you will next hear from us, here is your status link. Where the statutory window itself will be breached, say so before it happens, with the remedy you are applying; admitted lateness with a plan is survivable, discovered lateness is not.

The closing communication for an overdue case should acknowledge the lateness in its first lines, before the outcome — the pattern the letter guidance in the tone guide prescribes. Burying an apology after three paragraphs of outcome reads as concealment. Applicants judge the whole process by its worst moment handled; a late case handled with candour can end with more trust than an on-time case handled opaquely, and your outcome data will show it if you read it honestly.

What should leadership see, and how often?

A small, stable panel: open cases by age band, escalations this week with causes, breaches this quarter with remedies, median days to decision by type. ReviewDesk's real-case completion metrics count actual case records — the numbers leadership sees are the numbers the inbox holds, not a sample dressed as a census. Review monthly; the point is trend and cause distribution, not case-by-case interference, which teaches owners to hide cases from the panel.

Close the loop quarterly into the programme: recurring evidence-wait causes become supplier conversations; recurring capacity causes become staffing cases; recurring configuration causes become workflow fixes. Escalation data is the programme's self-knowledge arriving pre-analysed. Organisations that collect it and never read it have built a very accurate record of their own inattention — make sure yours is read, in a room, with actions minuted.

Your five actions for review-request escalation

  1. Compute escalation from the case's own deadline on every read; emit the state change once and keep the stamp permanently.
  2. Escalate in the final third of the window, and at birth for decision types whose human window is shorter than the statutory one.
  3. Require an escalation note with the blocker, the action and the next-update date — written as if the applicant will read it.
  4. Reassign blocked cases completely and without stigma; audit reassignment as load management, not performance.
  5. Update the applicant before any breach, acknowledge lateness first in the closing letter, and review cause patterns monthly.

Worked example: an escalated case that ended in trust

A fictional applicant for a care role — Jonas Weber — requests review of a screening decline based on a registry entry he says is a namesake's. The case escalates in its final third: the registry's confirmation is outstanding. The escalation note records two chases and the applicant is updated through his status link with the reason and the next date. On day twenty-six the registry confirms the entry belongs to another person; the correction is appended, the re-assessment clears, and the offer process restarts — inside the window, with the applicant informed at every stage.

Jonas's feedback, recorded on the case, is the kind money cannot buy: he describes the process as slow but straight. The case took twenty-six days — hardly brisk — yet every day of it was accounted for to him. The escalation system did not make the registry faster. It made the wait legible, and legibility is most of what applicants are actually asking for when they ask for a person.

Frequently asked questions

What does 'computed on read' mean for escalation?

The case's urgency is re-derived from its deadline and the current time every time the case is viewed or listed, rather than by a nightly report. An overdue case can never appear fine in a stale view, and escalation fires once at the state change.

When should a case escalate?

By default when it enters the final third of its window undecided, at once if its deadline passes, and from birth for decision types given tighter internal clocks. The escalation note must name the blocker, the action and the next update date.

Should overdue cases keep their stamp after completion?

Yes — the overdue fact stays on the record with its cause and remedy. Permanently visible lateness is what makes the statistics honest and the programme's improvement demonstrable.

How do you tell the applicant a case is late?

Before the breach where possible: what is awaited, what is being done, the next update date, and the status link. In the closing letter, acknowledge the lateness first, before the outcome.

Is this guide legal advice?

No. ReviewDesk is templates and workflow tooling, not legal advice. Escalation is described as ReviewDesk implements it; consult counsel for your obligations.

Keep reading: How does an applicant status link change the experience of waiting for a review? · How should human-review cases be assigned across a team? · What should a review case's decision history record? · How does the 3-year evidence vault export a complete decision file? · What should you ask a software vendor before you rely on their tool for consequential decisions? · All ReviewDesk guides · ReviewDesk home · Try the free desk · Free letter generator · Free notice builder · Pricing

ReviewDesk is built by Ingenious AI. Templates and workflow tools — not legal advice.