NPS calculation formula
Six Sigma – iSixSigma › Forums › Old Forums › General › NPS calculation formula
- This topic has 12 replies, 8 voices, and was last updated 12 years, 6 months ago by
Marty.
-
AuthorPosts
-
August 6, 2008 at 2:45 pm #50682
I know NPS is calculated as % promoters minus % detractors. My question is: are the percentages based on total # of all respondents or total # promoters plus total # detractors (i.e., # of “passive” responders are omitted from denominator)?
0August 11, 2008 at 5:17 am #174722Hi Don,
The %’s are based on total # of all respondents.
Cheers,
Daniel.0August 15, 2008 at 7:49 pm #174925Our operational definition has NPS caluclated based on the total number of respondents.
0January 9, 2009 at 3:23 pm #179544How do you calculate the minimum number of scores required to have a statistically accurate result? For example, if I have 250 sales people in the field, how do i know how many NPS scores I need for each in order to accurately compare the results (ie maybe have bonus structure that includes NPS).Any help is most appreciated.Thanks.
0January 12, 2009 at 7:03 pm #179623
Marty YuzwaParticipant@Marty-YuzwaInclude @Marty-Yuzwa in your post and this person will
be notified via email.Stuart,
This is the Excel formula that I use to calculate the +/- error percentage for my NPS calculations.
IF(“NPS %”<0.5,(NORMSINV(1-(1-0.95)/2)*SQRT("NPS %"*(1-"NPS %")/"No. of Surveys")),(NORMSINV(1-(1-0.95)/2)*SQRT("NPS %"*(1-"NPS %")/"No. of Surveys")))
The sample size will then depend on the how tight you need the error to be.
To give you an idea, if the NPS Score was at 50%, then a sample of 100 surveys would give +/- 10%. A sample of 200 surveys would be +/- 7%. It would take about 2000 surveys to get to +/- 2%.I find that in order to get all the samples I need I have to use a rolling 6 month or rolling 12 month NPS Score.
0February 10, 2010 at 9:10 pm #189262
Brian CarsonParticipant@Brian-CarsonInclude @Brian-Carson in your post and this person will
be notified via email.“if the NPS Score was at 50%, then a sample of 100 surveys would give +/- 10%.” Does this mean that the actual NPS is between 40%-60%? or between 45% – 55%?
0February 11, 2010 at 1:10 am #189267
Ken FeldmanParticipant@DarthInclude @Darth in your post and this person will
be notified via email.What do you think? Is 50% plus 10% equal to 55%?
0February 11, 2010 at 2:01 pm #189295
Brian CarsonParticipant@Brian-CarsonInclude @Brian-Carson in your post and this person will
be notified via email.My question is to clarify if the 10% is 10 points of NPS (50 + 10 = 60) or is it 10% of the current NPS score (50 x 10% = 5…..50 + 5 = 55)?
0February 11, 2010 at 2:01 pm #189296
Brian CarsonParticipant@Brian-CarsonInclude @Brian-Carson in your post and this person will
be notified via email.My question is to clarify if the 10% is 10 points of NPS (50 + 10 = 60) or is it 10% of the current NPS score (50 x 10% = 5…..50 + 5 = 55)?
0February 11, 2010 at 2:11 pm #189298
Ken FeldmanParticipant@DarthInclude @Darth in your post and this person will
be notified via email.This is what you posted:
if the NPS Score was at 50%, then a sample of 100 surveys would give +/- 10%All units are in % so why would you think otherwise?0February 11, 2010 at 2:22 pm #189300
Brian CarsonParticipant@Brian-CarsonInclude @Brian-Carson in your post and this person will
be notified via email.“if the NPS Score was at 50%, then a sample of 100 surveys would give +/- 10%.” comes from the post by Marty Yuzwa. I am asking for clarification.
0February 11, 2010 at 3:04 pm #189304
Ken FeldmanParticipant@DarthInclude @Darth in your post and this person will
be notified via email.Hopefully it is clarified or you can wait around for Marty to respond.
0February 12, 2010 at 1:05 pm #189337:)
0 -
AuthorPosts
The forum ‘General’ is closed to new topics and replies.