Project Team
Digital Loyalty | |
|---|---|
Operations | |
Data Engineering |
🌎 Project Overview
Tim Hortons (“TH”) collects guest satisfaction data on an ongoing basis utilizing Qualtrics as a facilitator for collecting & processing the data. This data provides direct voice-of-customer insights into: positives, pain-point and risk & opportunities, with respect to guest satisfaction. In parallel, TH’s loyalty programme collects a variety of data on registered guests, allowing for detailed analyses of guest behavior over time within the Databricks environment. In isolation, both datasets provide valuable insights to the business.
The aim of the project is to ingest the guest satisfaction data into the Databricks environment on a daily basis. This will allow users to join known guests' loyalty and satisfaction data together, with the aim of providing richer and more actionable insights based on the insights from combining these two datasets together.
📚 User Story
Title | Description | Priority | Notes | |
|---|---|---|---|---|
| 1 | Analyze guest satisfaction data in Databricks. | A user wants to use Databricks to query and analyze guest satisfaction data (loyalty and non-loyalty) on a daily & ongoing basis to, in part, review feedback from guests to gather insights on positives, risks and opportunities for improvement. | MUST HAVE | |
| 2 | Join guest satisfaction and loyalty data in Databricks. | A user wants use Databricks to write a query to combine guest satisfaction data (loyalty) with existing loyalty data tables on a daily & ongoing basis to, in part, develop insights based on a consolidated loyalty-and-feedback view from guests. | MUST HAVE |
|
✍️ Requirements
The guest satisfaction data should be ingested into the Databricks environment on a frequency no less than daily, and available for end-users across TH to query and analyze. The dataset as a minimum should include all non-PII columns, most notably: the loyalty customer identifier(s) and response-specific information such as date and time of response, questions & answers, restaurant information, and offer-related information. A sample of columns provided in the guest satisfaction dataset is available here: . Some sample questions users would aim to answer with the consolidated data in Databricks are:
What guest feedback are leading & lagging indicators of loyalty spend behavior?
What guest feedback are leading & lagging indicators of restaurant churn?
By building comprehensive data models of guest loyalty and feedback data, users are better positioned to obtain insights to the above example questions.
🖥️ Technical Specifications
Data Engineering will
Translate requirements into a plan with engineering activities to meet due date.
Engage with TH Operations and Qualtrics to agree on data processing requirements (e.g. establishing pipeline, curating dataset, etc.).
Complete the data governance requirements for the project (such as availing metadata such as a data dictionary for users).
Create a readable view of the guest satisfaction data to facilitate the user stories defined above.
Conduct the required quality assurance (“QA”) steps to ensure the data is accurate as per Qualtrics.
Provide support and ongoing maintenance of the guest satisfaction data in Databricks to ensure the data continues to meet the requirements defined herein.
Risks
Personally-identifiable information being included in the data model in the Databricks environment.
Project timeline overrun.
Changes to Questions & Answers on source data without prior notification to engineering and business teams.