Zero values pushed in via custom metrics could be counted either as "null"/missed values or actual zero value. You can decide which is the right one for your use cases, and a Planhat Admin can help you edit.

What the setting does:

  • "Yes": days with "0" value will now count as a "0"

  • "No": days with "0" values will count as null

Practical examples:

Let's assume we have 3 different custom metrics being pushed in to Planhat, found in the table below.

Day 1

Day 2

Day 3

Day 4

Day 5

Log-ins

7

5

9

0

4

Temp (C°)

22

15

15

0

17

NPS

5

3

7

0

5

The question, which the setting is for, is how to look at Day 4's data.

  • Log-ins: most likely 0 log-ins on Day 4 > setting should be "Yes"

  • Temperature: most likely missing data on Day 4 > setting should be "No"

  • NPS: most likely missing data on Day 4 > setting should be "No"

In other words: when you push in data where 0 means 0, the setting should be "on". The most common use cases for this is when pushing in usage activities, since it's a "counter"-type metric where 0 is the starting point. In the Temperature and NPS case, there is no "starting point" at 0, which means 0 likely means "no value":

Note that days with no values will always be counted as null.

The implications:

Whether the setting is Yes/No mostly impacts subsequent calculated metrics, e.g., when using the "LAST" operator or do averages over time. Using the "Log-ins" example above:

  • If I use the "LAST" operator on Day 4, then:

    • If "Yes", then Day 4 = 0 and so the last value is 0

    • If "No", then Day 4 = null and so the last value is Day 3 = 9 > Day 4 = 9

  • If I use the "Average" operator on Day 1-5, then:

    • If "Yes", then Day 4 has a value and is included in the average calculation, which produces an average of 5 (25/5)

    • If "No", then Day 5 does not have a value and is not included in the average calculation, which produces an average of 6.25 (25/4)

Note: this is a global setting for all custom metrics. If you have multiple metrics that have different logics (like the three above), then you can:

  • Transform the data you push in (e.g., turning the Day 4 value for log-ins from 0 to null before pushing it into Planhat)

  • Prioritize value of accuracy between the metrics, and set logic accordingly

Did this answer your question?