Skip to main content

Hey everyone,

We are currently in the process of adding the NDR to Gainsight. Currently we calculate them on the level: Market Segment → Region → Subregion. Since it is a percentage, we cannot then calculate it correctly for the whole company (or per market segment or region), as the average over percentages unfortunately leads to a distorted result. 
 

Have any of you ever had a similar problem and can help me?

 

Thanks in advance

Adrian 

Good question, @AdrianGlos .

For context, I’m making an assumption that NDR = Net Dollar Retention, where 

(Revenue at the start + upgrades – downgrades – churn) / Revenue at the start

I’ve found anytime percentages of a segment are in play, the best practice is to aggregate your “Revenue at the start” and your “Revenue at the start + upgrades - downgrades - churn” up to the level needed, then calculate the NDR quotient.

Put another way, you’ll likely need to work up the data stack by obtaining the Ending Revenue (the numerator) and Starting Revenue (the denominator) values for each Subregion. With that, you can configure Gainsight to aggregate via SUM, then calculate the NDR at the appropriate grain. Having only the percentages by Subregion likely won’t be enough, because to your point, the Subregions may not have equal weighting or influence on NDR as you rollup through various layers of the organization.


Thanks for the answer @matthew_lind

But in the Report Builder, it is not possible to sum up fields first and then calculate a calculated field based on these summed fields, is it?


Hey again @matthew_lind,

Is there a way to build the report so that you can filter the following via the dashboard filter? 

Region in EMEA or APAC

AND

Subregion not in DACH

 

Thanks in advance

Adrian


@AdrianGlos , glad to help. 

But in the Report Builder, it is not possible to sum up fields first and then calculate a calculated field based on these summed fields, is it?

You are correct. A calculation on top of a calculation isn’t supported inside Report Builder. The Rules Engine has some solutions here, though you need a Rule, some data destination in the MDA to write the output, and then finally a Report for analysis. Data Designer could also be a potential solution here, where you CAN do more “multi-stage” aggregation, math and processing, and then build a Report from your outputs.

 

Is there a way to build the report so that you can filter the following via the dashboard filter? 

Yes, and this is one of the most powerful use cases for Reports and Dashboards: Global Filters. I build them frequently, and they are fairly straightforward to build, providing all kinds of flexibility when presented to your end users. More on exactly how that’s done at this link.

 


Reply