This blog promotes knowledge sharing through experience and collaboration. For more product information, visit our WebSphere Commerce CSE page. For easier navigation, utilize the Categories to find posts that match your interest.
Watson Commerce Insights Data Flow
Watson Commerce Insights (WCI) consumes data from multiple sources such as store catalog, inventory, cost, plans, analytics etc, and presents in a way that allows merchandisers to identify problems and make better business decisions. In this post, let's go through the data flow between WCI and different systems it communicates with (only applicable when it's integrated with WebSphere Commerce) by using following flow diagram.
Note: Numbers mentioned in the above diagram are just for reference and it doesn't mean any sequence.
1. Catalog / Inventory
Catalog data is extracted from WebSphere Commerce database using data extract utility and pushed to WCI FTP server for WCI to consume it. Inventory data usually comes from Order Management System and the generated .csv files are pushed to WCI FTP server, if generated inventory files are accessible from WebSphere Commerce system, then FTPUpload schedule job can be used to push these files to WCI FTP server for WCI to consume it.
2. Analytics Data
WCI pulls product analytic data overnight based on the analytics system configured in WCI configuration page. Usually WCI pulls analytic data once a day and it's used to display product metric overlays and to generate intelligent sequencing files.
3. Sequencing update
Whenever you save a category product sequencing change in WCI, either through manual or intelligent sequencing, WCI make a REST call to Websphere commerce to persist the changes in database and to create/update the external index files (only if it's manual sequencing). You can refer Sequencing in Watson Commerce Insights blog to get more details and understand about this sequencing behavior.
4. Plan Data
Merchandiser can upload their sales plan data from WCI user interface, these sales plan data are used to compare your store / category's actual performance data against your plan and helps to validate how these categories are performing compared to your plan.
5. Cost / Event Data
Cost refers to average amount of money that your company pays for a product. You can upload cost data to WCI by pushing an input file (.txt) into WCI FTP server for further processing and it helps to calculate margin for a given product by comparing actual selling price from analytic data against your cost data. These calculated margin metric can also be used in intelligent sequencing condition to sequence products under a category. Event data covers events that have the potential to impacts sales, for example holidays. Event data input file can be pushed to WCI FTP server for WCI to consume it.s
6. Data to Watson Analytics
This functionality is only available for customers who are using IBM Digital Analytics as their analytic provider. Digital Analytics generates digital data feed (.ddf) file overnight and pushes it to WCI FTP server, WCI consumes the file, formats as per Watson Analytics specification and sends it to Watson Analytics.
7. Intelligent Sequence files
Whenever WCI pulls new analytics data from analytic provider or if customer uploads an inventory data, WCI generates new intelligent sequencing (.kpi) files based on newly uploaded metrics in WCI FTP server for WebSphere Commerce to consume it through TriggerMetricFileDownload schedule job and update the external search index file in WebSphere Commerce search.
8. Push to Live
WCI is only integrated with WebSphere Commerce authoring / staging environment, therefore all the WCI sequencing update goes to WCS authoring environment and customer need to run indexprop & stagingprop etc to push the changes to live environment in order to see the changes in live storefront.
For overall validation on different steps, you can refer Validating Watson Commerce Insights integration with WebSphere Commerce System blog post for further details.