# Planning a Cartographer Implementation
The following are a set of recommended activities to be reviewed internally with cross-functional teams before you begin building a Cartographer recommender. The teams should include those who will be involved in setting up, developing, and maintaining the Cartographer recommender experiences, which typically include Site Operations/Merchandising, Product Managers, Development/IT teams, PIM teams, etc.
# Plan Information Architecture
Build low fidelity wireframes to determine the desired recommendations flow
Identify questions and decide on page grouping
Define question types and answers
Establish product attributes mapping for each question, and map attribute values to answers
Determine product attribute model extensions
Decide on attribute extensions mechanism: at the source, in Salesforce Commerce Cloud; or using Cartographer’s extension data.
Define product recommendations type
Define results segmentation
Define supported locales
# Plan Data Import
Determine data import method, OCAPI or Excel spreadsheet data source
If using OCAPI method, configure OCAPI on staging and production environment
Consider creating Cartographer specific OCAPI client id and configuration
If using Cartographer Data Augmentation, create related Excel spreadsheet data sources
Determine source product catalog categories
Determine if image URL transformation is required, and define transformation rules
Define product import schedule
Configure data sources
# Plan Project Authoring
Configure Cartographer recommender project as defined by information architecture section. See the Cartographer User Guide (opens new window) for more details on how to set-up the recommender in the Project Editor.
Capture draft and production API keys
Publish project
# Plan Front-end Implementation
Familiarize with developer documentation: (this document; http://drivecommerce.com/cartographer/developer/)
Design recommender experience and build mobile, tablet, and desktop wireframes and high fidelity comps
Define interactive elements, including page transitions
Decide on supported browser versions and select appropriate integration script loading method
Decide on implementation strategy: content asset vs pipelines
Implement front-end recommender experience
If necessary, deploy Salesforce Commerce Cloud code updates
Consider using site preference or another environment specific configuration to store project API key
# Plan Performance and Conversion Tracking
- Determine analytics integration method: Salesforce Commerce Cloud back-end code or third-party tag manager
If using Salesforce Commerce Cloud back-end code method:
Implement Cartographer tracking tag for add-to-cart action
Implement Cartographer tracking tag for checkout confirmation page
If using tag manager:
Determine whether tag manager can be triggered for add-to-cart action and checkout confirmation page
Determine whether required product data is available to tag manager events
If necessary, update Salesforce Commerce Cloud implementation of tag manager to allow tracking of required events and data
Implement tracking tag for add-to-cart action
Implement tracking tag for checkout confirmation page
Determine IP address ranges to be excluded from analytics tracking
Define third-party analytics (Google Analytics, Omniture, etc) tracking requirements
Implement third-party analytics integrations
Review captured performance and conversion metrics, including conversion time interval
# Plan QA and UAT
Validate project API keys on staging and development environments
Validate recommender implementation
# Ongoing and Maintenance
- Review analytics reports either in a separate analytics system (e.g., Google Analytics, Omniture, etc.) or within the Cartographer Analytics dashboard (more details on available metrics can be found in the Cartographer User Guide (opens new window)).