Analytics Fulfillment Using Zendesk

Introduction

The Analytics team is now utilizing Zendesk to log and fulfill all stakeholder requests.  We believe that Zendesk offers benefits for analysts and stakeholders alike.  For Analytics, centralizing our backlog inside Zendesk will elevate the service we provide to stakeholders.  For Stakeholders, Zendesk offers several convenient ways to submit work requests.  This document is intended to guide stakeholders through the new fulfillment process.

Creating a Ticket

Zendesk offers Stakeholders two convenient methods to submit an analytics work order: email & Slack.

  1. Request Analytics Support Using Email (Recommended)
    • New requests can be emailed to analytics@deseretdigital.com
    • Each new subject line will automatically create a new ticket in Zendesk
    • Once sent, you will receive an automated email response confirming the receipt of your request
  2. Request Support via Slack
    • Open a direct message with an analytics team member
      • Note: Zendesk cannot currently join channels, so a direct message is required
    • In the chat box, type @Zendesk and press enter
      • If prompted, add Zendesk to the conversation
    • Next, type /Zendesk create_ticket and press enter
    • This form will appear:

 Snap.png

  • Fill out the form:
    • Select yourself as the Requester
    • Select Analytics as the Assignee
  • Once submitted, Zendesk will post a reply in the conversation confirming receipt. In addition, you (the designated Requestee) will receive an email confirmation

Ticket Description Field

The lack of a complete Description is the leading cause of delays during the fulfillment process.  Expedite the turnaround of your request by making sure you have provided these six pieces of information:

  1. Due date (if you have a hard deadline): When do you need this information returned to you and why?
    1. Meeting your deadlines is extremely important to us.  Just let us know. 
    2. Examples:
      • I'd like to have the information by the end of the week, because I am presenting on Tuesday.
      • I'm considering some product changes that would go in the next roadmap, so having this by end of month would be great.
      • My boss needs the information as soon as possible, so today or tomorrow please. 
    3. If you do not provide a due date, the ticket will be processed on a first in, first out basis
  2. Date Range: What is the time period you want to examine?
    1. Seasonality influences all products and biases almost all analyses, so Analytics will always confirm the reporting period before starting an analysis.
    2. Examples: 
      • Year to date
      • Last 6 months
      • February 2020 compared to February 2019
  3. Context: What is the specific behavior or phenomenon you're interested in understanding?
    1. Providing this helpful information to the analyst will result in him/her giving you a more considered and thorough answer to your question.
    2.  Examples:
      • I’m trying to figure out drop out in the posting process.
      • I’m trying to see if [some behavior] varies by time of year & device.
      • I’m trying to get the total number of Sessions from Social Media over time.
  4. Dimensions: Where is the data produced? 
    • This is surprisingly difficult to get right.  Miscommunication here often results in the analysis having to be restarted, so your analyst will want you to be very explicit.
      • Vertical, sections, and pages: Detail Pages in the Goats Subcategory
      • Platform: App Only, Browser Only, Both?
      • Devices: Phone, Tablet, Phone & Tablet, Desktop only?
      • Geo-restrictions:  Site traffic is delineated by City, State, & DMA; using these values, analytics can a generalize to County and Zip Code if needed.  Please be sure to differentiate whether you want to examine traffic that originates from a particular region (demand), or whether you want to examine listings where the good is located in a particular region (supply).  
  5. Metrics: What is the number you want back?
    • Users: Unique browsers that matched said Context+Dimensions at least once
    • Sessions: Every visit that contained said Context+Dimensions, no matter if it all came from a single User
    • Instances: Every time said Context+Dimensions were matched, no matter if it all came from a single User during a single Session
    • Something else: Average featured ad revenue per self-serve poster
  6. Breakouts: the “per” questions.
    • How do you want the number expressed?  
    • Examples
      • Total Users & Sessions per month
      • Posting conversion rate per device type

What Happens After You Create a Ticket

Analysts will review all new tickets, and respond with a personalized note within 1 business day.  If you need same day support, submit the ticket and then immediately speak with someone in the analytics department.

If the description looks complete, the analyst's response will restate your question (confirming understanding) and provide you with an expected delivery date.  The expected deliver date in this email is a estimate, not a guarantee.  Fulfillment might take longer than expected if there is an influx of other urgent tickets, or if the analyst gets sick, or if for some reason another ticket requires more time than was expected.  If you have a hard deadline, please provide a due date in the ticket description.

If the analyst needs more information, his/her response will contain a list of questions of clarification.    

Communicating on an Existing Ticket

After creating a ticket with Analytics (whether you create it in Slack or email), all subsequent communication must be done via email

Communication in Slack, other than creating the ticket, does not get recorded in Zendesk.  Remember when you create a ticket, you will always receive a confirmation email. The confirmation email serves as a great jumping off point to initiate a thread that will be recorded in Zendesk.

If you send a follow up email with a new subject line, you will create a new ticket in Zendesk.  To keep the queue manageable, Analytics will probably delete the duplicate ticket, and respond to you with a request to reply on an existing email thread.

Re-opening Solved Tickets

Once analytics has delivered a response to you, and hopefully delivered a solution to your request, the analyst will leave the ticket open.  You have 14 days to review the deliverable & reply with questions.  If after 14 days, no further communication is exchanged, the ticket will be permanently closed.

If you have a follow up question that is only tangentially related to the existing ticket, or requires a new data pull, you may be asked to enter a new ticket.

How Analytics Prioritizes Requests

Analytics will work the backlog in the following order of precedence:

  1. New requests that require expedited turnaround
    • The need for expedited turnaround is indicated by the inclusion of a due date in the ticket description
  2. Re-opened tickets in which Analytics failed to answer your question as stated
    • In other words, the request wasn't resolved due to an oversight by the Analyst
  3. First in, First Out
    • Re-opened tickets with follow up questions & new requests are worked in the order received
    • The clock starts once all requisite information is fully detailed by the stakeholder

 

 

 

 

 

Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request

Comments

0 comments

Please sign in to leave a comment.