🏎️PLG: Entity-level scoring

By the end of this tutorial, you will understand how to set up entity-level scoring in Cargo, use AI to enhance your scores, and effectively create segments based on those scores

Introduction to Entity-Level Scoring:

Typically, scoring mostly happens for the workflow level where records are processed and scored one-by-one.

However, you will come across use cases where it’s not efficient to leave the scoring to the workflow execution level and instead you want to create entire segments as a function of their score. There will be times when a simple numeric score won’t cut it and instead you’d use AI to use categorisation to create the scoring.

Picture this: You’re running a PLG motion want to identify high-potential users by scoring if the person is a key decision maker and whether they’ve sufficiently used the product automated score for each entity. You want to build a segment of high-potential users and engage them in a nurturing sequence

This is where entity-level scoring comes in. Let's delve deeper with an example of how we’re doing this for ourselves at Cargo

Setting Up Entity-Level Scoring with Cargo:

Begin with the Basics:

In your Cargo workspace, navigate to the entities tab and select the contacts entity (for instance)
Click the three-dots on the top-right of your screen and select the 'columns tab'

Add a computed column and click add to insert new fields

Now, you can start adding your scoring criteria
- Choose the data type for the new column. For this tutorial, we’ll choose #Number
- For the expression input, click on the small Library icon as in the image below to choose from one of the available recipes
- Now you can start adding your scoring criteria

Build a Score Using Defined Parameters:

Choose the attributes (columns) and assign weights to them based on their importance to your score
For example, here we’re taking the column app_run_count that tracks how many cycles of runs a user has performed on Cargo. Every time a user has executed more than 15 runs, we add a score of +20 to their score

Similarly, here we can attribute a positive score of 20 whenever the user signed up with a corporate email rather than a personal email

Other attributes we could have leveraged
- Deployed workspaces in the last month
- Total members on workspace made
- Engagement on marketing emails
- Duration since the last purchase

Enhancing Your Score with AI:

So far we were using columns or attributes that were easily classifiable using simple logic. What if we wan’t to use a score on something not so straightforward. We had a use case where we wanted to classify whether a user was a decision maker inside a team, and not just an individual contributor.
Setting one up is pretty simple. Here we chose an AI recipe from the library
In the configuration, pass the name of the column that contains the the job title of the user