Zendesk Macro Suggestion from Knowledge Base

Overview

This flow gives Zendesk support agents an instant, governed "what to send" suggestion on every new ticket. When a ticket is created, it runs a semantic search over the knowledge base in Azure AI Search and a keyword search of the Zendesk Help Center, then posts the best-matching articles/macros as a private internal note on the ticket, logs the suggestion to Dataverse for adoption tracking, and notifies the support team in Microsoft Teams.

Why it matters: Agents repeatedly resolve the same known issues. Surfacing relevant KB content per ticket cuts handle time, improves answer consistency, and (via the Dataverse log) lets you measure which suggestions agents actually use.

As built: This is a complete, connector-first reference implementation. It ships Off — going live requires only authorizing the connections and setting the environment-variable values (index name, Teams IDs, Zendesk auth header).

Use Case

A support team with a maintained Help Center / KB wants relevant self-service content auto-suggested on each new ticket, without auto-replying to the customer. Agents review the suggestion in an internal note and choose whether to use it. Suggestion adoption is tracked so KB gaps and high-value articles can be reported on.

Flow Architecture

Environment Variables

Connectors & Connections

Connector	API name	Actions used
Zendesk	shared_zendesk	GetOnNewItemsV2 SearchArticles
Azure AI Search	shared_azureaisearch	IntegratedVectorSearch
Microsoft Teams	shared_teams	PostMessageToConversation
Microsoft Dataverse	shared_commondataserviceforapps	CreateRecord

Customization Guide

Almost every realistic variant of this flow can be implemented by changing environment variable values. A few cases require small edits inside the flow definition — those are called out explicitly below.

Tune the search

Adjust flowlibs_AzureSearchVectorField / flowlibs_AzureSearchSelectFields to match your index schema, and raise/lower the Top Article Count variable.

Macro mapping

Extend the note (or add a Switch on ticket topic) to map a matched article to a specific Zendesk macro id.

Deflection

For simple intents, branch to a *public* reply with the top KB link instead of a private note.

Adoption reporting

Build a Power BI report over flowlibs_kbmacrosuggestion on flowlibs_usagestatus to show suggestion-usage rates and KB gaps (tickets where no match was found).

Source switch

The flow uses both AI Search and Help Center; disable either loop if you only maintain one KB.

Key Expressions

The flow is intentionally light on Power Fx / WDL gymnastics — the heaviest expressions are the branch-name concatenation and the approval outcome check. They are listed below in the order they appear in the flow.

@{trim(concat(variables('varTicketSubject'),' ',coalesce(triggerOutputs()?['body/description'],'')))}

@add(length(coalesce(outputs('Search_Help_Center_Articles')?['body/results'],createArray())),length(coalesce(outputs('Search_Knowledge_Base')?['body'],createArray())))

@concat('* ', coalesce(items('For_Each_Help_Center_Article')?['title'],'(untitled)'), ' - ', coalesce(items('For_Each_Help_Center_Article')?['html_url'],''), decodeUriComponent('%0A'))

@{concat('https://', parameters('EnvVar_ZendeskSubdomain'), '.zendesk.com/api/v2/tickets/', variables('varTicketId'), '.json')}

Customize & download