> For the complete documentation index, see [llms.txt](https://docs.harmony.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.harmony.io/dashboard-and-analytics/using-harmony-insights-for-kb-and-service-catalog-recommendation.md). # Using Harmony Insights for KB and Service Catalog Recommendations Harmony Insights continuously analyzes your resolved support conversations to surface patterns that help you improve self-service. Previously, the engine focused exclusively on recommending new knowledge base (KB) articles. It now also recommends service catalog items, so you can address both knowledge gaps and recurring service requests from a single workflow. This article explains how the recommendation engine distinguishes between the two suggestion types, where to find them in the dashboard, and how to act on each one. *** ### How the Recommendation Engine Works When enough resolved conversations share a common theme, Insights groups them into a pattern and classifies the pattern into one of two categories: * **Knowledge gap** - Users are repeatedly asking troubleshooting or how-to questions that could be answered by a self-service article. Insights generates a KB article suggestion, including a recommended title, outline, and relevant source conversations. * **Recurring service request** - Users are repeatedly submitting requests for something to be done on their behalf (for example, software provisioning, access grants, hardware orders, or account changes). Insights generates a service catalog item suggestion, including recommended form fields, a ticket type, and a priority level. The classification happens automatically. You do not need to tag or categorize conversations manually. The engine uses the content and resolution patterns of closed tickets to decide which suggestion type fits each detected pattern. *** ### Viewing Recommendations in the Insights Dashboard To see your current recommendations: 1. Open the **Harmony** dashboard and navigate to **Insights** in the left sidebar. 2. Select the **Recommendations** tab. 3. Use the **Type** filter at the top of the list to toggle between: * **KB Articles** - suggestions derived from knowledge gap patterns. * **Service Catalog Items** - suggestions derived from recurring service request patterns. * **All** - both types displayed together. Each recommendation card shows: * The detected pattern and a summary of the conversations that triggered it. * The suggestion type (KB article or service catalog item). * For service catalog suggestions: the recommended form fields, ticket type, and priority. * For KB article suggestions: the recommended title and content outline. * The number of resolved conversations that contributed to the pattern. *** ### Acting on a KB Article Recommendation 1. Click a KB article recommendation card to open the detail view. 2. Review the suggested title, outline, and the source conversations linked in the panel. 3. Choose one of the following actions: * **Create Article** - opens the KB editor pre-populated with the suggested title and outline, ready for you to write and publish. * **Dismiss** - removes the suggestion from your active list if it is not relevant. * **Snooze** - hides the suggestion temporarily and resurfaces it after a period you specify. *** ### Acting on a Service Catalog Item Recommendation 1. Click a service catalog item recommendation card to open the detail view. 2. Review the following fields that Insights has pre-populated: * **Item name** - a suggested display name for the catalog entry. * **Form fields** - a list of input fields recommended based on what agents collected in the source conversations. * **Ticket type** - the category Insights believes best fits the request pattern. * **Priority** - the default priority level recommended for tickets created from this item. 3. Choose one of the following actions: * **Create Catalog Item** - opens the Service Catalog editor with all recommended fields pre-filled. You can adjust any field before publishing. * **Dismiss** - removes the suggestion from your active list. * **Snooze** - hides the suggestion temporarily and resurfaces it later. > **Tip:** You are not locked into the pre-filled values. The recommendations are a starting point. Review the linked source conversations before publishing to confirm the form fields cover everything your agents typically collect. *** ### Frequently Asked Questions **How many resolved conversations are needed before a suggestion appears?** The engine requires a minimum threshold of similar resolved conversations before surfacing a pattern as a recommendation. This threshold is designed to filter out one-off requests and focus your attention on genuinely recurring topics. **Can I convert a KB article suggestion into a service catalog item, or vice versa?** Not directly from the recommendation card. If you believe the engine has misclassified a pattern, dismiss the suggestion and create the item type you prefer manually. You can also use the source conversations linked on the card as a reference. **Will dismissed suggestions ever reappear?** A dismissed suggestion will not reappear unless new resolved conversations create a fresh pattern that meets the detection threshold independently. **Who can view and act on Insights recommendations?** Access to Insights recommendations follows your account's role-based permissions. Agents and admins with access to the Insights section of the dashboard can view suggestions. Creating a KB article or service catalog item from a suggestion requires the same permissions as creating those items directly. *** ### Related Articles * Setting Up and Navigating the Harmony Insights Dashboard * Creating and Managing Knowledge Base Articles * Building and Publishing Service Catalog Items * Understanding Harmony Insights Data and Conversation Analysis --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.harmony.io/dashboard-and-analytics/using-harmony-insights-for-kb-and-service-catalog-recommendation.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.