How to use NPS: Can I get a hell yes?

Since the very beginning of onefinestay, we’ve always used NPS as our primary measure of guest satisfaction. NPS has become industry standard over the past 10 years - 2/3 of Fortune 100 companies measure NPS. 

NPS measures the likelihood of a future referral - hence, the primary NPS question of ‘how likely are you to recommend x product or service to a friend’. 

At onefinestay, we tend to be pretty hard on ourselves if our NPS dips below +60 - +70. As we should be - we’re a hospitality company, and still relatively young, so should always be striving for the highest scores possible. Any small business should view NPS both the cheapest sales and marketing channel, as well as evidence of product market fit. 

But to put things in perspective: any NPS above +50 is already excellent by industry standards. This is deceiving, because we tend to anchor to a ‘exam’ scoring system of 0-100, whereas NPS is on a -100 to +100 scale:

(per wikipedia) 

NPS can be as low as −100 (everybody is a detractor) or as high as +100 (everybody is a promoter). An NPS that is positive (i.e., higher than zero) is felt to be good, and an NPS of +50 is excellent.

Netflix's NPS score is +68 (as of early 2016). 

Apple’s NPS score is +72.

These businesses are generally perceived to have industry-leading products. 

NPS scores are deflationary due to its calculation methodology: 

(Promoters - detractors) / Total responses

where: 

9-10 = promoter

0-6 = detractor

7-8 = neutral (e.g. these scores are counted in the denominator only, which drags down overall scores)

NPS is a phenomenal way to benchmark customer satisfaction over time, and is a decent proxy for virality. But to tease out of NPS a true proxy for escape velocity, I like to go one level deeper.

Managing the customer, not the system

It’s important to remember its not the specific NPS score that matters (at least outside of a VC pitch deck), but how that score is a proxy for what it is intended to measure - likelihood of recommendations, and the virality that results from these recommendations. 

Like any system, its easy to fall into the trap of managing the system - rather than managing the customer. 

It’s easy to feel good with the score of ‘9’ because its a promoter - but there’s a reason the 9 isn’t a 10. It’s easy to be annoyed by an ‘8', but to a layperson unfamiliar with the NPS system, 8/10 is pretty good. Likewise, it feels bad to get a ‘5’, but its not a horrible result outside of the NPS framework. It doesn’t imply that a customer will never give the product another chance, or that a conversation couldn’t turn the situation around. 

Separately, but for similar reasons, its important to pay close attention to the comments. Every piece of customer feedback provides a valuable opportunity to improve. The scoring system does not replace qualitative feedback. 

Can I get a hell yes?

I’m a believer in Kevin Kelly’s 1,000 true fans concept. The foundation to a great business is a small group of people who absolutely love the product, and spread the word. 

On the flip side, if enough people have a horrible experience it can kill any business. 

I don’t think conventional NPS scores factor in strong emotional reactions to new products or services, often where companies need to grow to keep pace with expectations from investors & other stakeholders.

As such, I like to look at our ‘hell yes’ and ‘hell no’ scores - the extreme end of surveys to see who is saying ‘hell yes’ to the product (10/10), and who is saying ‘hell no’ (something went badly wrong, and the customer gave us a 0/10). And we make sure that the percentage of 'hell yes' is always rising. 

The 'hell yes' customers also create a rich cohort that will often take the time to provide feedback on new product ideas or offerings, elements of the service that really matter to them, etc. 

All new businesses need to create evangelists in order to survive. So for me, the ‘hell yes’ score is the best way to know you’re on the right track.