Obviously AI also allows you to make predictions on individual entities using Personas.

Let us consider the case study of Predicting the Customer Churn. You have a historic dataset on customers and you have used Obviously AI to create a prediction.

Let's say you have a new customer Bob.

Based on the various attributes we can find whether Bob is likely to Churn or not.

Navigate to the Personas tab on the Prediction Report and click on "Personas".



Specify the name in the 'persona name' field and choose and avatar which best suits the entity.



Once you choose the name and pick an avatar the next step would be to choose the values of different attributes/columns. Let's say Bob is Male, makes the payments though Mailed check and has monthly charges of $*150**.To set the value for gender navigate to the gender tab and click on bottom center position indicated by the green marker.



Finally Choose the option Male from the dropdown.



To navigate to the next set of attributes/columns click on the right arrow.



Using similar steps mentioned above update the value of Payment method to Mailed Check. Next, you need to change the value of Monthly Charges column. Monthly Charges is a numeric variable and can take a broad range of values. To change the values select the current value of the field using the cursor. Once you select the existing value, it will be highlighted in blue.



Bob has monthly charges of $*150**. So enter $150 once the current value is highlighted. The value of MonthlyCharges will then be updated to $150.



Finally, click Save This Persona.



You will then see a new Persona named Bob in the '*Saved Personas**' section.



Click on the saved Persona '*Bob**' to get your Prediction probability. The prediction tab shows that Bob has 89% probability of Churning.



What's Next ?

Check out how can you export your predictions using Google Drive?
How can you make predictions on a future batch that is not a part of your present dataset?
Was this article helpful?
Cancel
Thank you!