1. Knowledge Base
  2. The Faraday app
  3. Your predictions: Pipelines and Outcomes

How to create an outcome

How to build predictive models based on business outcomes you care about the most, in just a few clicks.


Table of Contents

  1. What is Outcomes?

  2. Accessing Outcomes

  3. Creating an outcome

  4. Analyzing your outcome

What is Outcomes?

With Outcomes, you're able to easily describe the business outcomes you’re trying to predict in just a few clicks. Let's say a major business outcome of yours is that you want to acquire more high-quality leads (who doesn't?)–you'll plug the outcome's cohorts in, and Faraday will create a likely-to-buy predictive model for that goal by applying dozens of strategies to build a range of candidate models, then selecting the one that most accurately predicts your outcome.

Accessing Outcomes

To access Outcomes, select the Consoles menu on the left-hand navigation bar, then find the Outcomes icon. If your screen size is large enough, the consoles menu many already be expanded.

Creating an outcome

Inside Outcomes, you'll find a list of available outcomes, as well as columns for:

  • The attainment cohort: people in the attainment cohort are those who achieved this outcome’s goal.

  • The attrition cohort: people in the attrition cohort are examples of people who failed to achieve the outcome's goal & enter the attainment cohort.

  • The eligibility cohort: people in the eligibility cohort are those that are eligible to achieve this outcome's goal and enter the attainment cohort.

  • The outcome's status: whether your outcome is Ready, or able to be used, or Error, or the outcome had an error when building.

  • Outcome actions: edit outcome and duplicate outcome.

To create a new outcome:

  1. Select + New outcome in the upper right of the Outcomes screen.

  2. Next, select your attainment cohort, which is required in order to create an outcome. In this example, we're going to create a simple lead conversion outcome, so the attainment cohort will be Customers.

  3. Optionally, select an attrition cohort for users that fail to attain this outcome, and an eligibility cohort to restrict which candidates are eligible to achieve this outcome. An important on eligibility cohorts, however, is that your eligibility cohort should not be a subset of your attainment cohort: if your attainment cohort is set to Customers, your eligibility cohort should not be something like Customers with a basement.

    🗒️ For example, say you want to create an outcome that scores leads based on the likelihood that they'll convert and become customers. In your outcome, you'll select the attainment cohort Customers, as the goal of your outcome is that leads will enter the Customers cohort and become customers. You'll leave the attrition cohort empty, as it's not relevant to this lead scoring model. Lastly, your eligibility cohort will be your Leads, as you're only interested in how likely it is that your leads will become customers. With these selections, Faraday will use your current customers as a baseline against which your leads will be scored on the likelihood that they'll become just like your customers.

  4. Once your cohorts are selected, give your outcome a unique name.

  5. With your desired fields are filled out, click Save changes to save the outcome, after which you'll receive a popup telling you that your outcome is building. You'll receive an email when the outcome is ready for use, and its status will display as Ready.

 

💡 The Outcome summary on the right will update as you add or remove cohorts in your outcome.

Analyzing your outcome

Once your outcome is complete and its status is Ready, click its name in the Outcomes list view to analyze its performance.

Your model will receive a score based on what results you can expect when using this outcome in your campaigns. The score can range from:

  1. Misconfigured: This can happen when your cohorts have too few people in them to make meaningful predictions.

  2. Weak: Expect your results to have minimal improvement when leveraging predictions from this outcome.

  3. Moderate: Expect your results to improve modestly when leveraging predictions from this outcome.

  4. Good: You can expect good results when employing this outcome in most supported use cases.

  5. Excellent: Your outcome is strongly predictable. You can expect great results in many use cases.

  6. Warning: Your outcome is predicted better than we would typically expect. You should check that only predictors known prior to the outcome are included. In other words, the model's performance is too good to be true. This can happen when the model calls on first-party data that's directly related to the outcome.

For a full report on everything involved in creating this outcome, click the technical report button.