Self-Service Data


Overview

This page contains forward-looking content and may not accurately reflect current-state or planned feature sets or capabilities.

Data Democratization is a common goal for Data Teams and can be difficult to achieve given the variety, volume, velocity, and veracity of data to manage. Ultimately, all effective data democratization solutions must focus on the Data Customer and provide solutions that make data easy to find, easy to understand, and actionable:

  • easy to find: Data Customers must be able to easily find data relevant to them
  • easy to understand: Data must be labeled and organized in a consistent way using business-friendly terms and concepts
  • actionable: Data must provide or help provide insights that support decisions or action, towards delivering results

GitLab Self-Service Data Program

The GitLab Self-Service Data program helps GitLab move faster with trusted data and helps the Data Team scale by providing four distinct, but complimentary capabilities:

  • Data Catalog - the Data Catalog helps Data Customers find data definitions, dashboards, and models based on a subject area
  • Dashboard Viewer - all GitLab team members have access to pre-build Dashboards through an always-on portal accessible from Okta
  • Dashboard Developer - for GitLab team members who want to build their own data visualization charts or dashboards
  • SQL Developer - for GitLab team members who are familiar with SQL and want to conduct their own SQL-based analysis
graph LR
 E9 --> W1
 DV1 --> W3

    subgraph Extraction
        E1[Stitch] --> E9[DW Endpoint]
  E2[Fivetran] --> E9
  E3[Postgres] --> E9
  E4[Sheetload] --> E9
  E5[Custom] --> E9
    end

    subgraph Snowflake Data Warehouse
        W1[RAW Data] --> W2{dbt}
  W2 --> W3[Dimensional Model]
    end

 subgraph Data Visualization
        DV1[Sisense Dashboards]
    end

 subgraph Self-Service Data
  SS3[SQL Analyst] --> W3
  SS1[Dashboard Viewer] --> DV1
  SS2[Dashboard Developer] --> DV1
    end

Summary of Self-Service Capabilities

Dashboard Viewer Dashboard Developer SQL Developer
Access Data Catalog Yes Yes Yes
Access Pre-Built Sisense Dashboards Yes Yes Yes
Build New Sisense Dashboards No Yes No
Write SQL to Explore Data No No Yes
Requires Certification No Yes Yes

Self-Service-Data

Data Catalog

The Data Catalog is published in the internal handbook and contains an index of data dashboards, models, and definitions organized by function. As part of the handbook, the Data Catalog helps drive alignment to common standards and single-source-of-truth data. As part of the Data Team handbook section, the Data Catalog is regularly updated with relevant content.

Data Catalog Structure

graph TD
 A[Data Catalog] --> B[Finance Data Catalog]
 A --> C[Marketing Data Catalog]
  A --> D[Sales Data Catalog]
  A --> E[People Data Catalog]
  A --> F[Engineering Data Catalog]
  A --> G[Product Data Catalog]

Data Catalog Handbook Page Contents

graph LR
 A[Data Catalog Page] --> B[Definitions]
 A --> C[Dashboards]
 A --> D[Data Models]

Self-Service Dashboard Viewer

Self-Service Dashboard Viewer enables GitLab team members access to any pre-built dashboard published in Sisense.

Prerequisites:

  • None

Access Requests:

  • None Required - all GitLab team members are provided access during onboarding

Self-Service Dashboard Developer

Self-Service Dashboard Developer provides Certified GitLab team members the ability to build their own dashboards using Sisense Data Discovery based on well-defined and validated Kimball-standard data models powered by the Enterprise Dimensional Model.

Prerequisites:

Access Requests:

  • Open a New Access Request and use the Single Person Access Request
  • Include a link to the completed Knowledge Assessment
  • Request Access to Snowflake

Self-Service SQL Developer

Self-Service SQL Developer provides Certified GitLab team members a SQL prompt into the Snowflake Data Warehouse. Because not all of the data is currently modeled in a standard form, additional time may be required to perform research, read “RAW” data and unmodeled data, and learn from teams with more experience. As the Data Team builds the Enterprise Dimensional Model and delivers additional Level 2 Solutions with SQL examples and data models, SQL Developer productivity will increase and SQL analysis will become easier and faster.

Prerequisites:

Access Requests:

  • Open a New Access Request and use the Single Person Access Request
  • Include a link to the completed Knowledge Assessment
  • Request Access to Snowflake