WebEngage Collections – Introducing Realtime Context Aware Product Recommendations

We are excited to bring you our next upgrade in the Catalog based new Product Recommendation module – say hello to Collections!

Let me walk you through a scenario to explain – you’re running a feastive campaign (Diwali in this case). It’s a standard campaign adopted by e-commerce companies which is designed to give you the best recommendations. Of course, you want those recommedations to be personalized. These are two different user personas, for example sake..

Rina – Budget-conscious, looking for gifts under INR 1,500
Arjun – Abandoned a cart with premium formal shirts last night

Both get your “Diwali Offers” email. Despite your best intent and recommendation alogrithms at play, Rina sees a premium products worth INR 8,000 and Arjun sees standard discounted items. Both don’t engage further 🙁

Most Recommendations are either too generic (Top Items) or too opaque (Personalised). And more often than not, marketers find themselves looking at their recommendation widgets thinking: “This doesn’t feel relevant. And I can’t even explain why it showed up.”

While both, Top Items and Personalized Recommendations are useful in their own way; they miss a core theme – they aren’t context-aware. Personalized recommendations excel at behavioral patterns, but they’d deliver the same results whether it’s Arjun in Mumbai or Simone in London. Collections makes your Recommedation location aware in this case. With Collections, you’d be able to recommend “popular items” in e-commerce with range filters in a certain geography or “popular restaurants” in a certain area or “coupon codes you can redeem at shops near you”. 

Hello, Collections

Collections is a real-time curation layer that works alongside your existing recommendations. While Top Items and Personalized Recommendations continue driving discovery and behavioral patterns, Collections adds campaign-aware, context-driven control to your recommendations.

Think of it as moving from “What users generally prefer” (Top Items) or “What similar users liked” (Personalized) to “What would be absolutely relevant for THIS user, in THIS campaign, at THIS moment based on their recent behaviour” (Collections)

How Collections Work?

Let’s go back to Rina and Arjun. Start on your Recommendations screen..

Choose Collections and set up dynamic rules that filter your catalog in real-time

We have essentially exposed all the columns in your Catalog as “filterable criteria” to match with user demography and behaviour – seasons, tags, locations, stock levels, prices etc. Collections makes all this data usable in real-time, without any engineering overhead.

Summary of Recommendation features

Feature	Best For	What You Could Do?
Top Items	Discovery, trending products	Campaign-specific filtering + business rules
Personalized Recommendations	Behavioral patterns	Marketer-controlled context + real-time rules
Collections	Campaign-driven curation	Transparent rules that adapt per user, per session

When to use what?

1. Top Items: Homepage discovery, “What’s trending now”
2. Personalized Recommendations: “You might also like”, behavioral nudges
3. Collections: Campaign-specific offers, festive sales, clearance events, geo-targeted promotions and customised rules when you want control.

They work better together. Use Collections when you need precise control over what shows up in a specific campaign or context.

Benefits of Collections

1. Design campaigns around business goals (festive sales, clearance, geo-specific offers)
2. Apply rules by season, location, behavior, or any custom user attribute
3. Complete transparency – no mysterious algorithms
4. Reuse Collections across all channels – emails, push, in-app, web etc
5. Multi-sort and randomization to keep experiences fresh
6. Mix automated rules with manual overrides when needed

Ready for a demo? Ask your Account Manager to schedule one for you – our experts will have this module set up and activated for your account.