Reference · built on requestOperator by FibricReporting & analysis

Contact Drivers

Classifies why customers contacted support each week, from conversations and order records, and proposes a drivers report and a fix task.

Request accessPricing ↗

About

A support team knows its volume. It rarely knows, in one table, why the volume came. Contact Drivers reads each week's tickets from Zendesk or Gorgias, each call and chat from Amazon Connect with its transcript and category, and matches each contact to the order it concerns in Shopify or Magento by order number or customer email.

Each contact gets a driver: what the customer wanted and, where an order is attached, what state that order was in when they asked. The operator proposes a drivers report ranked by count and by the order state behind it, and a fix task for the driver that a change upstream would remove. Both wait for a person to approve them.

This is a reference listing. It documents what Fibric would read from Contact Drivers and what it could propose, based on the vendor's published interfaces. Fibric builds it under a managed deployment when you request it; selecting it here installs nothing.

Inputs

  • Tickets with tags, custom_fields, via.channel, and requester from Zendesk, through the tickets list and incremental exports
  • Tickets with channel, tags, customer, and created_datetime from Gorgias, ordered by created_datetime
  • Contacts by channel, queue, and initiation method from Amazon Connect SearchContacts, with transcript text where conversational analytics is on
  • Orders by customer email, created_at, and status through the Shopify Admin GraphQL orders query
  • Orders searched with searchCriteria by status, created_at, and customer_email through the Magento REST /V1/orders endpoint
  • The driver taxonomy you maintain, and last week's drivers so movement is shown

Proposed actions

  • Target capability: propose the weekly drivers report in Slack, each driver with its count, its share of contacts, and the order state behind it
  • Target capability: propose a fix task in Asana for the top removable driver, with the tickets and orders that show it
  • Target capability: propose a Jira issue when the driver points to a defect in the store or the fulfillment flow
  • Target capability: propose applying the driver as a tag on each Zendesk ticket it classified, through Update Ticket

Proposed actions are target capabilities. Every action runs propose-first and needs a validated deployment and the appropriate permissions.

What you can build

  • Where is my order, sorted by why

    Contact Drivers splits the week's shipping questions from Gorgias by the Shopify order state at the time of contact: not yet shipped, in transit, delivered. The report shows which state drove the most contacts.

    With Gorgias, Shopify, Slack

  • Calls about one product page

    Transcripts from Amazon Connect show a cluster of calls about a sizing question on one Magento product. The operator proposes an Asana task for the content owner with the call excerpts and the order ids.

    With Amazon Connect, Magento, Asana

  • Tag the tickets so the driver can be watched

    For the driver the team decides to fix, the operator proposes tagging its Zendesk tickets through Update Ticket, so next week's count can be read straight from a view.

    With Zendesk

Requirements

  • A service desk the connector can read: Zendesk with ticket and user access, or Gorgias with ticket read
  • For calls and chats, an Amazon Connect instance the connector can search, with conversational analytics on for transcripts
  • A store the connector can read orders from: Shopify with read_orders, or Magento with an integration token
  • A driver taxonomy, even a short one, that the operator classifies against and a person can extend
Authentication
Contact Drivers keeps no keys. It reads through the Zendesk, Gorgias, Amazon Connect, Shopify, and Magento connectors you authorize and writes only through Slack, Asana, Jira, and Zendesk after approval.

Limits

  • A contact is matched to an order by order number or email. A customer writing from a different address is left unmatched and reported so.
  • Zendesk lists at most one hundred tickets per page; full weeks come through the incremental export, which trails real time.
  • Amazon Connect SearchContacts returns at most one hundred contacts per page and needs a time range on every request.
  • It classifies and proposes. It does not answer tickets, edit orders, or change routing.

Access and pricing

Reference listing. Fibric builds the operator under a managed deployment when you request it. Your quote covers the build, capabilities, usage, and support.

Request Contact Drivers ↗

Questions and answers

What does a person approve?
The report as it will post, the fix task with its title and evidence, and the tag proposal, each on its own. You can approve the report and hold the task, or reassign a driver before the report goes out.
What record does it leave?
Per week: each contact read, the driver assigned, the order it matched or the reason it did not, the report as approved, and each task or issue created. Re-classified contacts keep both the original and the corrected driver.
Does it read customer messages?
Yes, to classify them. It reads ticket text and transcripts through the connectors you grant, keeps the driver and the ids, and quotes short excerpts only in proposals a person sees. It never writes to a customer.
Ask about Contact Drivers

Ask about the capabilities and requirements in this listing.

For project-specific requirements, contact Fibric.

This operator is developed, published, and supported by Fibric. Third-party names and logos identify the systems an integration connects to; they are the property of their respective owners, who are not affiliated with Fibric and do not sponsor or endorse this listing. Trademark policy