Skip to Content

Salesforce Certified Tableau CRM and Einstein Discovery Consultant: How to Handle Null Values in Mostly North American Dataset for Einstein Discovery?

Learn the best approach for dealing with a small percentage of null values in a Region column when most of the data is from North America, in this sample question from the Salesforce Certified Tableau CRM and Einstein Discovery Consultant exam.

Table of Contents

Question

Universal Containers (UC) sells mostly to the North American market. A consultant is studying the spending habits of UC’s customers where the consultant collects the data with a Region column that is 94% North America, 3% Rest of the World, 3% null.

What is the appropriate action to take?

A. Leave the data as-is, let Discovery deal with nulls.
B. Replace the nulls with “Rest of the World” since Discovery rejects nulls.
C. Replace the nulls with North America since they are likely “North America”.
D. Drop the column since there is no valuable information from the column.

Answer

C. Replace the nulls with North America since they are likely “North America”.

Explanation

The reason for this is that Universal Containers sells primarily to the North American market, with 94% of the data in the Region column already being “North America”. With only 3% of the data being “Rest of the World” and another 3% being null, it is reasonable to infer that the null values are highly likely to also represent North American customers.

While Einstein Discovery is able to handle null values in most cases, it is still best practice to clean and prepare your data as much as possible before analysis. Replacing these few null values with the dominant category of “North America” allows the Region column to be fully utilized without dropping potentially useful information.

The other answer choices are less ideal:

  • Leaving the nulls as-is could work but doesn’t take advantage of the knowledge that most customers are North American
  • Replacing nulls with “Rest of World” incorrectly assumes those customers are not North American, contrary to what the data suggests
  • Dropping the column entirely discards the Region information, which could still have analytical value even with 94% of customers being from one region

So in summary, inferring that the small number of null Region values belong to the dominant “North America” category is the best way to handle this situation and allow Einstein Discovery to take full advantage of the Region data in its analysis and insights.

Salesforce Certified Tableau CRM and Einstein Discovery Consultant certification exam assessment practice question and answer (Q&A) dump including multiple choice questions (MCQ) and objective type questions, with detail explanation and reference available free, helpful to pass the Salesforce Certified Tableau CRM and Einstein Discovery Consultant exam and earn Salesforce Certified Tableau CRM and Einstein Discovery Consultant certification.