Identifies high-frequency support topics missing KB coverage by extracting, aggregating, and comparing ticket data against existing articles.

How it works 14 steps

Inspect the data fetches, transforms, gates, and output this recipe runs.

Data fetch

▼ tickets ticket-list-reader-crm

Fetch list reader from crm

▼ kb_articles kb-article-list-reader-cms

Fetch article list reader from cms

▼

Process (12 steps)

⚠ guard guard

Stop if tickets is empty

■ hashed_tickets hash_replace

Anonymize sensitive fields for safe AI processing

✦ extracted ai-extract

AI extracts structured fields from data

■ topic_stats group_by

Group by question_topic

■ kb_titles map

Extract from each item

✦ matches ai-compare

AI compares two pieces of data

✦ parsed_matches ai-extract

AI extracts structured fields from data

■ frequent filter

Filter by condition

■ gaps filter

Filter by condition

■ sorted_gaps sort

Sort by

■ restored_data hash_restore

Restore anonymized fields back to real values

■ gap_table to_table

Format results as a data table

▼

Output

table gap_table

Settings 2 configurable

Configurable at install. Defaults shown — change them anytime in Recued.

lookback days number = 90

min occurrences number = 3