What it is & the business problem it solves
The Item Enricher is a module of Inspire (CM's AI personalization product). It adds recent popularity figures to every product in your Product Feed (the product catalogue that CM's channels read from):
|
Product Feed attribute |
Meaning |
|---|---|
|
interactionCount7Days |
Product detail views in the last 7 days |
|
interactionCount30Days |
Product detail views in the last 30 days |
|
transactionCount7Days |
Purchases in the last 7 days |
|
transactionCount30Days |
Purchases in the last 30 days |
The webshop, Web Personalization and recommendation modules can then sort and filter on what is trending or selling right now. You don't have to compute these numbers yourself or upload them separately.
How it works
-
On each run, Inspire counts for every product how often it was viewed and bought in the last 7 and 30 days. The counts come from the standard events in your customer data platform (CDP or CXP). These four counts and periods are fixed.
-
Optional parent-level counts. Many shops list one product (the "parent") in several sizes or colours (product variants). With Aggregate Counts On Parent Level switched on, the counts of all product variants are added up onto the parent. A popular T-shirt then shows its total popularity, not the popularity of a single size.
-
The four counts (interaction and transaction counts over 7 and 30 days) are written to the Product Feed as product attributes. They can then be used by every channel that reads the feed.
Example use cases
|
Scenario |
Channel |
Required products |
|---|---|---|
|
A fashion retailer shows a "Trending this week" carousel on the homepage, sorted by views in the last 7 days |
Web Personalization (sorting on a Product Feed attribute) |
CDP (with Retail & Ecommerce adapter) or CXP (with Templates), Webtracking, Product Feed, Web Personalization, Inspire Item Enricher |
|
A pet-food webshop shows a "Bestsellers this month" block sorted by purchases in the last 30 days |
Web Personalization / email product blocks reading the feed |
CDP (with Retail & Ecommerce adapter) or CXP (with Templates), Webtracking, Product Feed, Web Personalization or Email Campaigns, Inspire Item Enricher |
|
A DIY retailer's merchandisers filter out products with no purchases in 30 days from a promotional selection |
Merchandising filters on the Product Feed |
CDP (with Retail & Ecommerce adapter) or CXP (with Templates), Product Feed, Inspire Item Enricher |
|
A retailer boosts recently popular products inside its personal recommendations |
Inspire recommendation modules (boost settings that use Product Feed attributes) |
Item Enricher plus a recommendation module, Product Feed, Recommendations API or Email Campaigns |
What you can configure
You can configure the settings below in the cockpit. Settings marked Expert need expert rights. CM staff and customer admins can hold expert rights; request them via your CM contact person.
Item Enricher runs as a single module instance (one configured copy of the module) per workspace. Most settings live on the module's variant (the published set of tuning settings). You change them in a preview (draft) and then publish it. Schedules are cron expressions (time expressions) in Amsterdam time.
|
Setting (cockpit name) |
What it does |
Default |
When to change |
Who can change |
|---|---|---|---|---|
|
Schedule |
When the module runs |
Not scheduled |
Must be set (e.g. daily), otherwise counts are never refreshed |
Expert |
|
Aggregate Counts On Parent Level |
Adds up the counts of all product variants onto their parent product |
Off |
Your feed has separate rows per size/colour |
Standard |
|
Products Receiving Aggregated Counts |
Expression for which products get the summed counts (e.g. "is parent = true") |
Empty |
When aggregation is on |
Standard |
|
Variant Link Column |
Product Feed column holding the parent's item ID (e.g. a parent ID column) |
Empty |
When aggregation is on |
Standard |
|
Delivery Method |
Write the counts to the Product Feed (batch) or not at all (none) |
Batch |
"None" for a trial |
Standard |
|
Advanced/expert settings (CPU Request/Limit, Memory Request/Limit) |
System resources |
500m CPU, 4Gi memory |
Very large catalogues |
Expert |
The four counts and their periods (7 and 30 days) cannot be changed.
Switching parts on and off: Aggregate Counts On Parent Level switches the parent-level counts section on or off; it is off by default, and the settings in that section only become editable once it is on. If you do not see a switch for it in your cockpit, ask your CM contact to enable it.
Prerequisites & required CM products
|
Need |
Why |
|---|---|
|
CDP or CXP |
Source of the view and purchase events |
|
Webtracking |
Product detail views (interaction counts) |
|
Retail & Ecommerce adapter (CDP) |
Sets up CDP's standard event types (Purchase, Add to cart, ...), including the purchase (order line) events. It is included in onboarding, so there is nothing extra to buy or configure |
|
Templates (CXP) |
On CXP the Retail & Ecommerce adapter does not apply. CXP uses Templates to set up the standard events instead |
|
Product Feed |
Destination of the counts. Product IDs in events must equal the feed item IDs. |
|
Web Personalization / Email Campaigns / Recommendations API |
Channels that use the counts |
|
Cockpit |
Configuration, event viewer and statistics |
|
Data module |
Inspire's data module prepares the data for all modules (see the Data page) |
Packaging tier
Included from: Pro. Item Enricher is available in the MMC Pro and Enterprise packages. The Basic and Advanced packages do not include Inspire.
Data requirements
-
Events: product detail views (counted as interactions) and purchases (counted as transactions), using CDP's standard event types. Custom event types are ignored.
-
Product Feed: item IDs matching the event product IDs. A parent-ID column is needed for parent-level aggregation.
-
Technical minimum requirements: the module defines no minimum number of products, events or days of history. Products without views or purchases in the period simply get a count of 0.
-
Cold start: new products start at 0 and gain counts as soon as events arrive and the module runs.
How to check it works
-
Product Feed: open a few products and check that the attributes interactionCount7Days, interactionCount30Days, transactionCount7Days and transactionCount30Days are filled and updated after each run.
-
Event viewer "Enriched Items": shows a sample of enriched products with their counts.
-
Cockpit statistics: statistics are opt-in. In the cockpit you insert a statistics template and save. The module offers these distributions, each showing how your products are spread over tiers:
-
"Interaction (last 7 days)" and "Interaction (last 30 days)": products per view tier
-
"Transaction (last 7 days)" and "Transaction (last 30 days)": products per purchase tier
The tiers are fixed: none, 1-5, 6-10, 11-15, 16-20, 21-25, 26-30 and 30+ events in the period. They show how concentrated popularity is across your catalogue.
-
-
Sanity check: 30-day counts should be at least as high as 7-day counts. If all counts are 0, check that event product IDs match the feed item IDs.
Limitations & common pitfalls
-
Item-level only: the module enriches products, not customers. It does not produce per-customer lists such as "products this customer should reorder". For that, see Recurring Purchase.
-
Fixed metrics: only views and purchases over 7 and 30 days. No other periods, no add-to-cart counts, no popularity rank.
-
Not scheduled by default: without a schedule the counts are never refreshed. Counts are as fresh as the last run.
-
ID mismatch: if event product IDs differ from feed item IDs, counts stay 0.
-
Product variants vs parents: if the shop shows parent products but events are on product variants (or the other way round), switch on parent-level aggregation. Otherwise sorting on counts gives misleading results.
-
One instance per workspace.