Audit intermediate Works with: Any AI

Knowledge Base and Self-Service Effectiveness Audit

Table output Has example

Check whether customers can solve problems without contacting you.

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

Audit self-service. Top twenty help articles with views: [TOP_ARTICLES]. Top search terms and searches with no result: [SEARCH_DATA]. Ticket topics and volumes: [TICKET_TOPICS].

Compare ticket topics with articles to find topics with high contact volume and weak or missing articles. Check article titles use customer words, steps are current, and search finds the right result. Estimate the contacts that better articles could remove, with the arithmetic and assumptions, for example 40 percent of 250 contacts is 100.

Return a table of gaps, a prioritised list of ten articles to write or fix, and a measure for deflection that does not rely on assumptions alone.

Add a worked gap analysis with invented volumes. Ticket topics: password reset 300 a month, article views 120 and the article is outdated; invoice copy 180, no article; delivery tracking 250, article exists but search for track order returns nothing; refund status 90, no article. If better articles and fixed search remove 40 percent of password and tracking contacts, savings = 0.4 x (300 + 250) = 220 contacts a month; at 4 per contact that is 880. Check the assumptions with a before-and-after comparison and allow for seasonal changes. Search terms: track order, invoice, cancel; zero-result searches 15 percent; add synonyms and redirect. Titles: replace Order Information with Track your order, using the customer's words. Build a plan for the top ten articles with owner, due date and review date, a template with the symptom, quick answer, steps, screenshots with alt text and last-checked date. Measure: helpfulness votes, share of article viewers who then contact support within 24 hours, and zero-result search rate. Add a monthly review of search terms and tickets, and a rule that every product change triggers a check of affected articles. Note that deflection figures are estimates until measured.

Add a closing note: review the top searches and zero-result queries monthly.

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.

Invented monthly ticket topics: password reset 300; delivery tracking 250; invoice copy 180; refund status 90.
Gaps: password article outdated; tracking article exists but search for "track order" returns nothing; no article for invoice copy or refund status.
If fixes remove 40 percent of password and tracking contacts: 0.4 x (300 + 250) = 220 contacts; at 4 per contact = 880 a month.
Zero-result searches: 15 percent. Retitle "Order Information" as "Track your order".
Measure: contacts within 24 hours after viewing an article.

• audit