---
title: Decision Component
slug: xlsmart/decision-component
docTags: 
createdAt: 2026-04-06T23:03:08.801Z
---

In this section, we will browse the different component of Adobe Decision engine to understand how we select the best offer.



## Collections

The offers you want to propose may depend on the context. For example if you want to urge your customer to renew his mobile plan, you may want to propose a list of eligible plans. 

For other communication such Loyalty, you may want to propose a list of offer he can redeem with point. 

To define the list of offer eligible for a context, we use **Collection**. 

Go to Decisioning > Catalog > Collection and open the **Telco - All offers **collection. 



![](https://api.archbee.com/api/optimize/1Jp0iolWv35Z_mkI7oUiT/mUPYBIrKhj8qKFcrGxsJI_image.png)

Click on preview collection, and ensure your offer is present. If not, maybe you forgot to add the Telco Tag as this is the condition we used to include offer in this collection.



![](https://api.archbee.com/api/optimize/1Jp0iolWv35Z_mkI7oUiT/iMRkJWMlM_YJb_9RQRbmA_image.png)

If your offer is here, that means it may be proposed in campaign who use this collection. 



## Ranking

Now that we have a list of offer, we have to define how to select the best. We have a few options : 

- using business priority, defined at the offer level
- Define Business logic, such as boosting offers for specific customers
- use AI to determine the best chance of engagement or purchase
- A combinaison of those strategy.  



We already seen how to set up priority at the offer level. Let's check business logic. 

Go to Decisioning > Strategy set up, and click on **Ranking formula**. 



![](https://api.archbee.com/api/optimize/1Jp0iolWv35Z_mkI7oUiT/jf3W-r_vaq89NEtzd9bl2_image.png)

Click on **Telco - Ranking by ARPU**. You will see the logic here is to boost offers that have more data than your cusotmer avg consumption.  

This ranking formula will be applied for each offers each time we call the engine for a customer. Each offer will end up with a priority, or score, then we will sort those offer to keep the best(s) of eligible.  



![](https://api.archbee.com/api/optimize/1Jp0iolWv35Z_mkI7oUiT/v3S9e5UksIeAXTCgGTF8q_image.png)

Next, you can explore the AI Model, you can click on create AI model to see the option. 

We propose 2 mode: Auto Optimisation, which basically will select the best offer overall, or the personalized optimisation. This one will select the best offer at the individual level, based on the audience is qualified for. 

We will not create any AI model today as it take some time to train. just close the screen and move to next exercice.  

![](https://api.archbee.com/api/optimize/1Jp0iolWv35Z_mkI7oUiT/XIX84tB1VFoGN2uHi7pIk_image.png)

## Selection Strategy

Lastly, we need to define the selection strategy. The Strategy just combine the different element we've seen previously : Which offer (collection do I want to use, and how (ranking Strategy) do I want to select the best. 

Go to Decisioning > Strategy Setup > Selection Strategy to browse existing strategy. 

You can eventually click on Create if you want to see the option proposed, but for this HOL we will reuse the existing one.  

![](https://api.archbee.com/api/optimize/1Jp0iolWv35Z_mkI7oUiT/WTU359D23rF5_sO7CdXKj_image.png)

Now that we have everything setup, let's see how to display those offers.