NPS doesn’t say anything particularly useful • Hacker News Board

Re: NPS doesn’t say anything particularly useful

#81

Post by itsdrewmiller » Sat, Jan 01, 2022, 5:06 PM UTC

rkk3 wrote: post #76

itsdrewmiller wrote: post #70

Earlier quoted context omitted.

Yeah, I agree it doesn't have the universal agreement on implementation that NPS does. Here's an alternative that I believe does and is referenced in some of the NPS skeptical research: https://en.wikipedia.org/wiki/American_Customer_Satisfaction...

That "measures the satisfaction of consumers across the U.S. economy", very different from Brand's Customer Experience Surveys. Having worked as a software vendor in the CX field, I saw a lot of different implementations of the "same" survey metrics. NPS was only standardized one (I think because its from b school in the 80's).

Nevertheless it seems to be roughly as predictive of growth - https://web.archive.org/web/20200716065914/https://pdfs.sema...

Re: NPS doesn’t say anything particularly useful

#82

Post by dcposch » Sat, Jan 01, 2022, 8:15 PM UTC

sokoloff wrote: post #68

dcposch wrote: post #59

Earlier quoted context omitted.

NPS has the advantage of being concrete and simple, producing evenly sampled feedback. This is hard to get any other way. A proper NPS asks only one question and requires only a single click to answer -- eg "Scale of 1-10 how likely are you to recommend to a friend?" > click [8] > done. There's a free-text "Tell us more" field that is totally optional. This means users actually answer it. NPS feedback is the closest…

In an almost Goodhart’s Law case, now that I know how NPS is used, I answer these questions differently, being far more prone to use 6, 7, or 8 to express my opinion.

Eh this is a niche issue. I bet less than 1% of people answering NPS have ever heard of "NPS".

The only way this would be a meaningful effect is for a dev tools or similar product where your audience is the HN audience.

Re: NPS doesn’t say anything particularly useful

#83

Post by Dave3of5 » Thu, Jan 13, 2022, 4:03 PM UTC

austenallred wrote: post #25

Yes, NPS is an imperfect measure, as all are. What I’ve never seen is a great argument for what one should use instead.

Company I was working with done an internal study on this I believe CSAT or a simple average were better if your goal is to improve revenue.