NPS Key Driver Analysis

NPS Key Driver Analysis is a NPS follow up question and an out-of-the-box dashboard tile.

The Key Driver question is a structured customer experience follow up question. Key Driver has a similar format to the Rating Grid question, but the prompt is more like a CSAT question. The Key Driver Question asks “How would you rate us in these areas?” You can customize the question so your survey recipients can rate the key areas of your business to measure the impact to your customer experience.

The Key Driver dashboard tile identifies those same areas of your business that impact your score so you can proactively make changes that improve customer loyalty.

Best Practices for Key Driver Analysis

Set up your question with the low score on the left side of the survey builder, and the high score on the right side. Like 0-10, Poor - Excellent, Very Unsatisfied - Satisfied. Switching the direction of the response may cause negative correlation and warning on your dashboard tile.
Phrase your key drivers positively. "Breadth of inventory" not "Inventory shortfalls"
Key Driver questions set up using shapes (hearts, stars, or circles) must have at least 5 shapes. This ensures we get a range of responses to track trends.
More responses collected on Key Driver questions make for better data analysis. GetFeedback recommends collecting at least 30+ responses, so we can see data trends.
Map your Key Driver question to a Salesforce number, text, or text area field. Map each item in the key driver grid to it's own field in Salesforce.

Key Driver Question

Building a new NPS survey program

Start building your GetFeedback survey. Choose +Add Question and select Key Driver from the menu.

Adjusting an existing NPS program

We recommend adding a Key Driver question to your existing survey. Use survey branching logic to skip over the original follow up question to preserve previously collected responses.

Copy your existing NPS survey and change the original follow up question (likely multiple choice) to the Key Driver question type. Compare your old and new NPS surveys side by side in your GetFeedback dashboard or Salesforce reporting.

We don’t recommend changing your NPS follow up question (likely multiple choice) to the Key Driver question type within your original survey without making a backup. When change a question type, previously collected responses to that question will be deleted.

Salesforce Mappings

Set up a new Salesforce custom mapping for your Key Driver responses to a custom Key Driver field in Salesforce. Supported field types in Salesforce are : Number, Text, or Text Area. You'll want to map each Key Driver item to it's own field in Salesforce.

Key Driver Dashboard Tile

Measure the correlation between your NPS score and key drivers to understand which areas of your business are impacting your customer loyalty.

Imagine you have a low NPS score with a high correlation to a key driver. The key driver dashboard tile highlights the particular area(s) that is problematic for your business so you can focus on what part of the customer experience needs improvement.

If you have a low NPS score, with a low correlation to your key driver, that variable may not be a factor in your NPS. This could be a positive area of your business that informs other opportunities for growth.

Reading the tile

Potential driver- An area of your business you hypothesize is important to customer’s overall perception and loyalty.
Correlation with NPS- The degree to which a given potential driver’s scores vary together with NPS scores. When you observe a high driver score you also observe a high NPS score, a low NPS score matches a low driver score.
Average driver score- An indicator of overall customer satisfaction with a potential driver.
Driver score distribution- See the spread of your response scores.

Interpreting Correlation and your Key Driver Score

GetFeedback uses Spearman's rank correlation on the NPS and Key Driver question to calculate your correlation score.

Hover over your dashboard tile to highlight the statistical significance of the data. This is also called the P Score. This helps us evaluate how likely it is that the score is not due to chance.

Strong correlation and high average driver score = A key value driver
Strong correlation and low average driver score = A key pain point
Weak correlation and a high average driver score = possible area to remove investment
Weak correlation and a low average driver score = no action recommended.