Net promoter score

Customer NPS survey responses are broken into three categories. Though each customer has their own interpretation of the numeric scale, these categories allow for a clear overall NPS calculation.

Customers who score either 9 or 10 are considered loyal, enthusiastic and likely to recommend your business to someone else.

Customers who respond with a score of either 7 or 8 are satisfied but not necessarily likely to recommend your business to others. 

Customers who respond with a score of 6 or below are unlikely to return. They won’t recommend your business, They could potentially discourage others from buying from you. 

Broadly, there are two kinds of NPS surveys. Transactional surveys are sent after a particular interaction in the customer lifecycle. Meanwhile, relational surveys measure customer response to an ongoing relationship. 

Net Promoter Score score is calculated by subtracting the percentage of detractors from the percentage of promoters. Passives are not counted, they are considered neutral. 

Usually, a business does not only have a single NPS. NPS can measure customer satisfaction per product, outlet or loyalty program. NPS is a popular means of tracking the success of customer service/customer support teams. 

Using NPS to track satisfaction with each stage of the customer journey. For example, customers may be satisfied with the product itself but dissatisfied with the customer experience that goes along with it. Exploring this further, the NPS for customer support may be high but the lack of an omnichannel strategy means they don’t feel this level of care is being given to the rest of the business.

Many businesses believe a quality product will sell itself but revenues for businesses that prioritize customer service actually rise 4-8% above their market average. Comparing NPS is a quick way to determine which part of the business needs the most attention. 

Broadly, any score over 0 is considered somewhat positive as it means a company has more promoters than it does detractors. Between 30 and 70 is good and 70+ is excellent. However, what’s considered a good NPS varies widely by industry. An NPS in the minus figures may seem poor when taken in isolation. However, even this may be the case once compared to industry benchmarks. 

The popularity of NPS hinges on the simplicity of the survey and the results: one question that can be broken into just three categories. There will always be flaws in a system this simple. In the era of omnichannel retail, it makes sense to gain a better picture of your audience.

Here are some NPS flaws and what methods you can use to tackle them:

NPS ignores passive results. This is a measure bought in to make calculations even simpler but can paint a distorted picture of how a business or department is perceived. A business with a large number of passive results and detractors giving scores of 5 or 6 could have the same score similar to a business getting much more negative feedback.

Measurements based on concrete results like referral programs give a more realistic reflection of how customers are speaking about a business. 

As noted by the Wall Street Journal, it is easy to end up concentrating on improving an NPS by manipulation, rather than working on the core issues. This can be both intentional and unintentional. A business may decide that artificially improving a score, e.g. through selective surveying or psychological queues in UX design, are simple fixes. Conversely, lack of context might leave that as the only possible action to take. 

Again, optimizing customer experience by the engagement of your most dedicated customers with a loyalty program gives results based on action, not intent.

According to the Harvard Business Review, acquiring a new customer can be five to 25 times more expensive than retaining an existing customer. How a customer interprets a one-question NPS survey can be very different from how those results are interpreted by a business. A long-term customer loyalty strategy needs to be based on more nuanced information based on actual customer behavior.