LogoLogo
HomeDocumentation
  • 🐳Using Whaly Guides
  • Core concepts
    • 📚Getting started
      • Data stack architecture
      • Consumers vs Builders
      • Data layers in Whaly
      • License Mapping
    • 🪄Data modeling
      • Understanding data models
      • Designing data models
      • Common modeling patterns
        • Event schema
      • Maintaining data models
      • Data models best practices
    • 🖌️Explorations
      • Understanding Explorations
      • Designing Explorations
      • Maintaining Explorations
      • Mistakes to avoid
  • Training
    • 👁️For viewers
    • 👩‍💻For editors
    • 🧙For builders
      • Setting up the training material
      • Creating a chart
      • Using and editing explorations
      • Filtering a dashboard
      • Creating explorations and models
  • Inspiration
    • 🗒️Use cases
      • Billing / Invoicing
      • Customer success
      • Fundraising
      • Marketing
      • Partnerships
      • Product
      • Sales
      • Strategy
    • 💬Communication
    • 💡Tips
  • Recipes
    • 🤝Customer care
      • How to build a 360° customer dashboard
    • 🏦Finance
      • Modeling your recurring revenue
        • SQL for simplified MRR calculation
        • SQL for advanced MRR calculation
    • 📣Marketing
      • Track your entire Marketing Funnel
      • Calculate your Customer Acquisition Cost
      • Create a partner dashboard
    • 💼Sales
      • Analyze the impact of your Sales velocity on your closing rate
      • Create a sales performance dashboard
      • Build a target oriented sales dashboard
  • Misc
    • 🧐SQL Fanout
    • 📦Backup your data using BigQuery
    • ☁️Embedding reports in Salesforce
    • 👨‍💻Useful SQL operations
      • Flattening categories
Powered by GitBook
On this page
  • For Data Consumers:
  • For Data Providers/Builders:
  1. Core concepts
  2. Explorations

Understanding Explorations

PreviousExplorationsNextDesigning Explorations

Last updated 2 years ago

An Exploration is a query interface that will empower your data consumers to ask questions by themselves and get their answers. Properly designing Explorations is a key pillar of offering a "self service" BI experience to your company.

Explorations contains a collection of dimensions and metrics, coming from multiple tables that were created in the "modelization layer" of your Data Stack. Such collections should be fined tuned to contains the proper fields and metric definitions that data consumer need in order to get answers.

A proper setup of well designed Exploration will have the following benefits:

For Data Consumers:

  • No insight lag: Being able to get answers fast without having to write complicated SQL queries on the fly

  • Work on cleaned/consolidated data: Getting access to a consolidated repository of cleaned dimensions and metrics that are safe to use

  • Consistency: Getting the same results as the rest of the team and not having to troubleshoot difference in KPIs with co-workers

  • Empowerment: Being able to go beyond "dashboards" to answer complicated questions without having to open a ticket to the data team

For Data Providers/Builders:

  • Governance: By defining in a single place the definition of metrics, you don't allow every Data Consumer to invent a new way of calculating important KPIs, lowering data inconsistencies and improving organisation trust.

  • Low maintenance: Having a single place to build / update metrics. A single metric definition change will update all the dashboards/integrations that are using it.

  • Less low value work: Data Consumers can run their own extract / simple enquiries by themselves, freeing time for the more value added analysis.

  • Abstraction layer: Exploration are living above data tables but are not exposing them directly, you can change everything in your Data Warehouse without impacting any Data Consumer as long as you update your Exploration to use the newly created tables.

  • Avoid invalid SQL queries being used: Exploration are generating a safe SQL that doesn't contains errors that SQL novice can make easily which results in invalid numbers, like

As you see, properly designing Explorations is a high stake topic.

🖌️
SQL Fanout