← Back to the support-ticket project

PulseDesk · Design rationale

Support ticket
learning design

The rationale behind Write a Useful Support Ticket and a proposed way to assess the work it prepares learners to do.

Sarah McLean / Self-directed PulseDesk concept / October 2026

The learning decision

The Rise sample asks employees to turn an observed software problem into a record a support teammate can investigate. The working hypothesis is that a model and short writing task can help people include useful facts without guessing at a cause.

That need has not been verified with a workplace team. Before recommending training, I would review approved, de-identified tickets and their follow-up questions with a support lead. I would also observe the intake process: a confusing form, missing access or unclear ownership may require a process change.

What is already built

Learners examine a SalesView report export that returns error E214, then draft a ticket about an inactive Save button in ProjectBoard. The second case changes both the app and the symptom.

The course provides a worked example, an evidence comparison and a reusable five-prompt reference. The writing task is self-checked; no learner results are claimed.

The work to assess

Given a new incident and local ticket rules, the learner writes a specific title and records the task and observed result, relevant context, available evidence, attempts already made and work impact. Unknowns are stated plainly. Unsupported causes and prohibited sensitive details are left out.

A fact from the practice case

At 2:15 p.m., Save stayed inactive after a comment was entered. Reloading once did not change it.

This supports a report of what happened. It does not establish why the button failed. That distinction is part of the assessment.

Language and access

Judge the accuracy and usefulness of the record. Equivalent wording, dictation and approved translation can be accepted where they fit the role; grammar and English fluency are not stand-ins for ticket quality.

Before a pilot, check keyboard access, readable prompts and enough time to compose a response. Confirm required fields, permitted evidence and the treatment of unknowns with the support lead.

Ticket review

A draft rubric for a new fictional incident. Agree the required facts and local rules before scoring. Genuinely unavailable information should be flagged, not invented; score against what the case actually provides.

Issue summary

0 — Missing
No usable task or symptom
1 — Incomplete
Task or result only; title absent or vague
2 — Usable
Specific title plus attempted task and observed result

Context and time

0 — Missing
Required available context absent
1 — Incomplete
Some context; follow-up needed
2 — Usable
Relevant app, environment and time

Error and evidence

0 — Missing
Available evidence absent or irrelevant
1 — Incomplete
Some detail; a key gap remains
2 — Usable
Exact message or observed steps; relevant evidence

Already tried

0 — Missing
No account of attempts
1 — Incomplete
Attempt without its outcome
2 — Usable
Attempt and result, or explicitly none

Work impact

0 — Missing
Absent or unsupported
1 — Incomplete
Affected work named
2 — Usable
Specific task, scope or deadline, without exaggeration

Factuality and privacy

An invented fact, an unverified cause presented as fact, or prohibited sensitive information makes a ticket unsuitable for submission regardless of score. Correct it before calling the ticket ready.

Proposed pilot

With a support lead, prepare two comparable cases. Two reviewers independently score three example responses and resolve differences against the rubric. Agree a provisional readiness rule before collecting pilot responses.

Invite 6–10 volunteers to write a baseline ticket, use the course and reference, then write about the second case without its model. Vary case order where practical. Double-score at least 20% of responses. With approval, review the next 1–3 eligible workplace tickets per participant after 1–2 weeks.

The evidence to report

Compare baseline and transfer responses descriptively. Report ready tickets and factuality/privacy passes with their denominators, criterion scores, non-submissions, exclusions and reviewer agreement. Record difficulties using the reference. A small volunteer pilot can expose usability problems; it cannot establish that training caused improvement. Share aggregated findings only.

PulseDesk is fictional. This is a proposed evaluation; no workplace research, pilot or measured impact is claimed.