What is Catalog useful for?
Catalog is the main place for the products and services your AI Employee needs to know. Instead of spreading prices, details, images, and search information across different places, keep each item together so Mr. AI can recommend the right item and answer customer questions accurately.
- If you are building a Catalog from scratch, start by giving AI the item data and business context before creating the items in Catalog.
- Let AI propose the Folders, Semantic Tree / Semantic Classification, Semantic Tags, AI Search Attributes, and Search Names from the full set of items.
- Review that structure, then use Build & Manage with AI to create and place the items. This is the path we currently recommend because it is easier to keep the Catalog consistent and review the work in batches.
Recommended setup order for a new Catalog
- 1
Prepare the source that contains your products or services: a file, list, or other clear item data with useful business context.
- 2
Use Build & Manage with AI and ask it to analyze the whole set and propose Folders for internal organization.
- 3
Ask it to propose the Semantic Tree / Semantic Classification and useful Semantic Tags from the patterns and meaning across the items.
- 4
Ask it to propose AI Search Attributes, values, Aliases, and Search Names for precise customer questions, using facts supported by the source.
- 5
Review the proposals and correct anything that does not match how your business actually works.
- 6
Once the structure is approved, let AI create the items and attach each one to the right Folder, classification, Tags, and Search Attributes.
- 7
After indexing, test natural discovery questions and precise lookups in Test Assistant.
How does Mr. AI search?
- If a customer says, “I need a course that helps me learn sales,” that is a discovery request and semantic search helps find relevant items.
- If the customer says, “show every course by Hassan Faleeh” or “how many courses are in the Cairo branch?”, AI Search Attributes support the more exact lookup.
- When an exact business fact is maintained in Catalog, such as price, instructor, or branch, the system can rely on that structured fact instead of guessing from similarity.
A quick example
- 1
Give AI the course list and useful context about the business.
- 2
Ask it to propose Folders, a Semantic Tree, and Tags for the whole set.
- 3
Ask it to identify useful Attributes such as Instructor and Level, plus Aliases and Search Name where needed.
- 4
Review the structure, then ask AI to create and place the courses instead of adding every item manually.
- 5
Review the result, then test one natural discovery question and one precise lookup in Test Assistant.
Choose the guide for the step you are on
- Start with “Manage a large Catalog with AI” when you have a source or item set and want AI to propose the structure before creating the items.
- Use “Organize Catalog: Folders, Semantic Classification, and Tags” to understand and review the structure AI proposes.
- Use “Precise search: AI Search Attributes and Search Name” to understand and review exact search facts.
- Use “Add and edit a product or service correctly” as a field-by-field reference for one item or for manual edits.