Create recommendations with demo data

This article provides guidance on how to leverage omnichannel product recommendations in Tier-1 single box environments using prepopulated, customizable demo data.

Omnichannel product recommendations provide a set of editorially curated or programmatically generated list of products. These lists can be used in several scenarios, depending on the business need. For more information about product recommendation lists, see Product recommendations overview.

For Tier-2 and higher Dynamics 365 environments, product recommendations are automatically computed based on customer data. Using product recommendations demo data doesn't disable any product recommendations solution already provisioned in the environment and any costs associated with its usage.

For Tier-1 environments, product recommendations are based only off the static demo data stored in a .csv file.

Enabling product recommendations demo data in an environment

To enable product recommendations demo date, you need to deploy the Dynamics 365 Commerce Preview Demo Extension to the respective environment. Doing so automatically enables product recommendations demo data.

Default demo data

Each OneBox type environment comes with a preloaded set of product recommendations demo data stored in the coma separated 'reco_demo_data.csv' file, located on the Commerce Scale Unit.

The data is structured along the following columns.

Column name	Mandatory	Description	Possible values
RecoList	✔️	The specific product recommendation list type that the demo data point is to generate.	RecoBestSelling RecoNew RecoTrending RecoCart RecoPeopleAlsoBuy RecoPicks RecoSimilarVisual RecoSimilarTextual
OperatingUnitNumber	✔️	The specific operating unit number where product recommendations are expected to be surfaced.
Category		The category the specific list should be returned for. If no category is specified, the list is for top of navigation hierarchy only.
SeedItemId		For lists that require seed (RecoPeopleAlsoBuy and RecoCart), the product those lists should show additional products for.
CustomerId		For lists that require a customer identifier (RecoPicks). The default value '0' applies to all customers.
ItemIds	✔️	One or more products to be returned as the result, separated by ';'.

Customize demo data

You can edit the default demo data with any product and category information configured in HQ. After you update the .csv, the product recommendations that are returned to customers will immediately reflect the changes.

The extension contains a datafile called 'RecoMockDataset.csv', which allows you to control the dataset used to power the mock recommendations results. The file name can be controlled through extension configuration using the ext.Recommendations.DemoFilePath setting. This enables you to have multiple datasets available that can be switched between easily through configuration.

<settings>
    <add name="ext.Recommendations.DemoFilePath" value="RecoMockDataset.csv" />
</settings>

Additional resources

Product recommendations overview

Enable Azure Data Lake Storage in a Dynamics 365 Commerce environment

Enable product recommendations

Enable personalized recommendations

Opt out of personalized recommendations

Enable "shop similar looks" recommendations

Add product recommendations on POS

Add recommendations to the transaction screen

Adjust AI-ML recommendations results

Manually create curated recommendations

Product recommendations FAQ