A person's engagement is a metric used to measure people's satisfaction. Engagement is similar to eNPS but allows us to have more granularity thanks to the metrics and submetrics it is composed of.
In Nailted, a person's engagement is achieved through the completion of a battery of 113 questions distributed in the following categories: Direction, Feedback, Career, Recognition, Wellness, Company image, Happiness, Fellowship, Managers and Satisfaction. One question from each category is included in each of the surveys sent out every 2 or 3 weeks or on a monthly basis.
How is the value of a metric calculated?
To calculate the value of a metric we will group all the responses sent by person per month, average them and apply this logic to all the people and finally apply the average of the sum of the people to calculate the total value of the metric. See below an example of how the value of metric D is calculated over a period of time.
Person 1

Answers question A: 4

Answers question B: 6

Answers question C: 5
Person 2

Answers question A: 8

Answers question B: 6

Answers question C: 4
The calculation of the value of metric D will be found by applying the following formula:
Person 1 = (4+6+5)/3 = 5
Person 2 = (8+6+4)/3 = 6
Metric value D = (5+6)/2 = 5,5
This formula is applied to find the value of engagement, metrics and submetrics.
Why is engagement not calculated with the average of the metrics?
At Nailted we always try to get the most accurate data we can calculate and in the case of the engagement metric or any other metric this is not achieved by calculating the average of the average of the metrics. Let's look at an example that demonstrates this:
The answers obtained to the question asked for metrics A and B were as follows:

Metric A answers: {2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2}

Metric B answers : {4}
If we perform the averages of metric A and metric B we obtain the following results:

Average of metric A: 2 (the number 2 is repeated 13 times).

Average of metric B: 4 (the number 4 only appears once)
If we average the averages, we obtain the following result:
((4+2)/2) = 3
On the other hand, if we take the average of all the answers obtained in both metrics, this is the result obtained:
(2+2+2+2+2+2+2+2+2+2+2+2+2+4/14) ≃ 2.14
There is a variation of more than 0.8 points between the two calculations, with the average of all the answers providing a more representative figure.
Furthermore, this way of making the calculation helps us to give the correct weight to each answer within the metric, since if one question has 100 answers and another has 20, its weight will be in accordance with the number of answers received, which is not the case if the average of the averages is taken.
In the example of the image, it would not be correct to calculate the Wellness metric by averaging the submetrics (healthy habits, stress level and quality of work) because the number of answers in each of them is different and we would not be giving the correct weight to each of the answers.