When maintenance requests happen outside the CMMS

A conversation can start a request. It cannot be its only record.

An operator mentions an issue over the radio. A supervisor sends a message. Someone asks a passing technician to take a look. The computerized maintenance management system, or CMMS, shows none of them. Before concluding that people refuse to use it, follow one request from the moment it was raised to the decision about what happens next.

The problem may be capture, access, review, or feedback. A requester who cannot find the asset needs a different response from one who submitted the issue twice and heard nothing. A supervisor dispatching work from a private list creates a different problem again: the team is committing capacity outside the visible queue.

For routine work, agree on one accountable intake process even when observations arrive through several approved channels. The goal is a traceable request and decision, not a ban on talking to each other.

Diagnose the route before changing the form

Ask an operator and the person who reviews new requests to walk through the same recent example. Compare what each believes happened.

• Access: can the requester reach the approved intake route during the shift, with their own authorized access? What happens when the device, network, or asset lookup is unavailable?
• Capture: who records a verbal request, and how does the requester know it happened? “Someone will enter it later” leaves two people assuming the other owns it.
• Information: does the form ask for observations the operator can provide, or planning and diagnosis they cannot yet know?
• Review: who checks new items and covers absence? Is an unreviewed request visibly different from accepted work waiting for planning?
• Response: can the originator find a decision, an explanation, and a next review point without contacting a particular supervisor?
• Dispatch: do supervisors use the recorded disposition when assigning work, or does the private message still outrank the queue?

Record the point at which the trail disappears. More reminder posters will not fix a login problem, and a shorter form will not fix a queue nobody reviews.

Define what a request needs and what it does not promise

Capture the asset or identifiable location, what was observed, when it happened, and the reported operational impact. Preserve the originator’s description. Add a way to ask follow-up questions and a supporting reference when available. A report of an unfamiliar noise is an observation; it should not require the requester to name a failed component.

The reviewer adds the routing decision under the site’s priority rules. Record whether the item needs clarification, links to existing work, enters planning, requires escalation, or is declined with a reason. A request number confirms receipt. It does not promise a completion date or authorize execution.

Agree a review cadence and absence cover that the team can deliver. Show the next checkpoint for items awaiting information. If the site uses a paper or offline fallback, name who transfers it into the CMMS, preserves the original observation time, and returns the reference to the requester. Reconcile those fallback records so they do not become another permanent backlog.

Keep urgent communication available through the site’s existing emergency and escalation procedures. Do not delay an urgent report while someone completes a form. Assign responsibility for the associated record within that process.

Check whether the route earns trust

Trial the intake agreement in one area. Compare a defined selection of shift-log entries and reported requests with the CMMS record. Count unmatched items, explain the sample, and examine why they were missed. Do not describe that sample as the share of all plant work.

Follow declined and duplicate requests as carefully as approved ones. Can the requester see the reason and, where relevant, the surviving work-order reference? Check items awaiting clarification for missing owners. A low number of new requests may mean fewer problems, but it may also mean people stopped reporting.

At the review, choose one barrier to remove and one response commitment to check again. Keep any work discovered during the trial subject to normal review. Recording a request does not reserve crew capacity or grant it priority over existing commitments.

Bring a simple intake record to the next discussion

Use the free Notification Template as a starting point for observation capture. Adapt it to approved access, reporting, and compliance requirements at your site. The 24-Hour Triage Checklist shows the book’s proposed review routine; its timing is an operating model to assess locally, not a universal standard or an emergency response rule.

The Reject-Pile Review Agenda helps operations and maintenance examine decisions that reporters did not understand. When accepted work later disrupts commitments, use the weekly schedule guide. When the record exists but responsibility disappears at a shift change, use the handoff guide.

For the wider rollout problem behind persistent paper workarounds, read The Floor Report #18: Adoption Debt. Its additional field materials are for paid Substack subscribers. The companion tools linked above remain free and require no email.