Rounding in the dashboard user interface

Your main score is 9.0. The top 5% benchmark percentile is also 9.0, and above.

However, the dashboard indicates that your score sits in the top 10% benchmark percentile.

Your unrounded engagement score is below the necessary threshold dictated by the unrounded top 5% benchmark percentile.

Example: your main score is 8.96123864, while the top 5% range is 8.97363652. Both numbers round up to 9.0 in the dashboard, however your score is actually 0.01239788 below the threshold for the top 5% percentile.

You're using the 'Difference to benchmark' viewing option in the segment heat map.

You've expanded the Engagement driver to display the benchmark difference per individual driver/subdriver score, as well as the overall difference for the Engagement driver.

All individual Engagement driver & subdriver scores display a difference of 0.0, but the overall engagement driver displays a difference of 0.1.

Although the heat map displays a difference of 0.0 per driver question, the actual difference can range from -0.05 to 0.0499.

The overall Engagement driver difference of 0.1 also uses rounding, and could range anywhere from 0.05 to 0.1499, based on the individual driver/subdriver scores.

You calculated the raw percentage of employees in each NPS category.

The total percentage displayed on WPEV adds up to 100%, however, if using standard rounding rules, it should be 101%.

Does different weighing apply to each NPS category rounding up or down?

WPEV rounds percentages to the nearest whole number. The rounded percentages from the 3 NPS categories must add up to 100%.

If the first decimal point on a percentage is 5 or above, it must round upward. Example: 48.6% becomes 49%. However, if this is the case for all 3 NPS categories, the total becomes 101%.

To mitigate this edge case, WPEV identifies the NPS category where the first decimal point is the lowest, and rounds it down, rather than up.

You've 6 active questions in your survey schedule, and the approximate time to complete the survey displays as 2 minutes.

Before launching the survey, you activate 2 additional questions, so the total is now 8.

However, the approximate time to complete the survey still displays as 2 minutes.

WPEV customer data indicates that the average time to answer each question is about 0.25 minutes.

We multiply the number of questions by 0.25 minutes to calculate the approximate survey completion time. For the survey estimation, we always round up to the nearest whole number.

Using the above calculation, 6 questions taking 1.5 minutes to answer and 8 questions take 2 minutes to answer.

Both versions display as approximately 2 minutes in the schedule and survey emails.