Key Takeaways

1. Purpose of a Data Layer:

   • Delivers consistent and quality data across an entire digital estate
   • Enables joining data together for improved marketing, customer journeys, and bug fixing
   • Essential for companies with complex digital ecosystems (e.g., logged-in areas, products, checkouts across multiple platforms)

2. Data Layer Definition:

   • More than just a JavaScript object
   • Includes the schema definition and core principles underpinning the architecture
   • Designed to meet specific needs and built with consideration of who will maintain it

3. Data Layer Structure:

   • Reflects basic human interaction: Input (user action) and Output (view)
   • Everything else is context
   • Standardizes the "conversation" between customer and business

4. Architectural Considerations:

   • Data layer is a resourcing solution
   • Designed to allocate work efficiently between implementation teams and development teams
   • Must be easy for developers to build and maintain consistently

5. Design Principles:

   • Remove bespoke elements and standardize for automation
   • Keep it simple and event-driven
   • Make it centralized and delivered from the server
   • Create a modular structure with reusable components
   • Make the data layer granular for flexibility and reduced tech debt
   • Automate validation using tools like JSON schema

Action Items

1. Assess Need for a Data Layer:

   • Evaluate your digital ecosystem's complexity
   • Determine if you need consistent data across multiple platforms or technologies

2. Plan Data Layer Project:

   • Focus on delivering data consistency and quality, not just technical implementation
   • Align the project with revenue generation and cost-saving goals

3. Define Responsibilities:

   • Assign implementation team to define and maintain the data layer schema
   • Task development teams with building and maintaining the data layer as part of platform maintenance

4. Implement Design Principles:

   • Standardize and simplify data layer structure
   • Create a centralized, event-driven architecture
   • Develop modular, reusable data components
   • Implement granular data structure for flexibility

5. Establish Rigorous Processes:

   • Create strict acceptance criteria for new additions to the data layer
   • Implement automated validation using JSON schema or similar tools

6. Foster Cross-Team Collaboration:

   • Ensure implementation team understands development processes
   • Create clear communication channels between implementation and development teams

7. Automate and Validate:

   • Integrate data layer components into platform components for automatic consistency
   • Set up automated validation processes to catch errors early in development

8. Continuous Improvement:

   • Regularly review and refine the data layer schema
   • Stay updated on best practices in data architecture and implementation

Remember: A well-designed data layer is crucial for delivering value from first-party data. It should be approached as a strategic initiative that enables better decision-making and improved customer experiences across all digital touchpoints.

For more information:

• Twitter: @mattdoesdata
• LinkedIn: @Matts_At_Work
• Presentation Deck: Speakerdeck.com/mattsatwork24