The best way to measure NPS

NB The article is originally produced in Danish

How to measure?

The classic NPS measurement is conducted via e-mail and, as mentioned in the previous article, often only once a year and perhaps only as a supplement to, or as part of, a traditional questionnaire survey.

It has to be clear that the fundamental point of departure is that an NPS measurement has to stand alone to have the best possibilities of succeeding. If it has to be effective and have an operation character too, the surveys have to be conducted drip by drip so that the follow up can be timed appropriately. Furthermore, it would be preferable if the surveys are automated in a close-knitted process.

Systemize the campaign

The most important success criterion for an effective operational loyalty program is that the program is strategic and that the measurements are put into a system. First and foremost, the measurements must have an owner or sponsor who reaps the benefits of customer feedback. It should be a leader of a business unit or a team where NPS is included as a central KPI in the department’s results and development. Next, the manager must have the entire department’s support for the value and prioritization of the program, so that measurements are translated into concrete actions. When and how often can it be measured? NPS is often divided into two types of measurements:

• Relation survey (loyalty)
• Transaction survey (satisfaction)

Relation survey

As mentioned, the relationship measurement is the whole, ie the customer’s overall perception of the relationship. It can have arisen from many elements, including of course the product and the specific experiences the customer has in connection with finding, buying, receiving and using the product. However, many other factors can also influence customer satisfaction and loyalty such as. political conditions, events, etc., which are not directly related to the product or specific customer experiences.

In order to create a credible NPS loyalty measurement in the relationship, the measurement should therefore take place on an ongoing basis and not at a specific time. Measured e.g. among all customers right after announcing a sharp fall in prices, this can in all probability significantly affect short-term satisfaction. Therefore, customer loyalty should always be measured on an ongoing basis so that the NPS score becomes an expression of customer loyalty over time and not just in the present. This also makes it possible to see a trend in NPS development more quickly rather than having to wait for more measurements.

Continuous measurement of loyalty can be established in many ways. The right frequency varies depending on the product and relationship type, but a good rule of thumb is twice a year. As I said, it is important to have the measurements done in drips, rather than a total annual measurement. In this way, it becomes an ongoing program that both creates a credible NPS score and at the same time makes the relationship measurement and results always current, so operational follow-up can be distributed over time and becomes a natural part of everyday life and the company’s customer-related work processes.

Transaction and touchpoint survey

Transaction / touchpoint surveys can be used across the entire customer experience. The crucial thing is that there is a clear purpose to the surveys and that the results are used both operationally and for general improvements in the customer experience. The measurements must first and foremost be programmed for a touchpoint in the customer journey. The model on the right shows examples such as before or after customer meetings, just after a large offer has been made, by delivery, before re-signing an agreement and the like.

The touchpoint campaigns can then again have more transactions. For example. a Project NPS campaign where measurements could be made after start-up, midway through the project, and before and after completion.

The correct frequency for measurements depends on many parameters. There are e.g. large difference in the customer’s commitment in the relationship to an electricity supplier versus a supplier of a newly built house. An NPS.Today customer who builds houses successfully measures a full 12 times during the customer relationship and receives real feedback on improvements to the experience in all 12 touchpoints and a basis for individual follow-up and dialogue.

Of course, processes should be established, and preferably automation, to ensure that measurements in transactions do not take place with an annoying frequency for the recipient. If the company has a central system that collects customer data, rules and triggers for distribution can with great advantage take place here, as the distribution can be based on many parameters about customers and transactions. Of course, it requires the system to have functionality to support such rules and triggers. If this is not possible, it is crucial to choose an NPS system that offers function to avoid survey spam (also called Throttling).

Which channels should be measured?

The choice of channel for the measurement should be made on the customer’s terms. It must be the channel that is most relevant in relation to the transaction and that makes it as easy as possible for the recipient to submit his answer. Does the marketing department e.g. input to the company’s newsletter, it is obvious to merge the question here. If customer service has had an e-mail correspondence with a customer, the question can be included in the closing e-mail or in a separate e-mail when the case is closed. If you want feedback on the experience on a website or an app, it will be natural to ask the question here, either actively (eg pop-up) or passively. The latter I get more into in the section below.

Examples of channels are:

• E-mail
• SMS
• APP
• On a website or in a web shop
• On the product itself (for example a QR-code)
• iPad (for example a stand in a shop, in a cafeteria or at an on-site interview)

How to measure?

As a starting point the question can be asked in two ways:

• Active (prompt)
• Passive (visible)

The active question is usually sent to the recipient (eg via e-mail or SMS) but can also be asked in a pop-up or by interview such as e.g. Tivoli, which interviews guests about the experience at the end.

The passive question can appear in many different ways e.g. on a product, on a website, on a sign, in a document, etc.
The passive question is an exciting opportunity that many overlook. In particular, the opportunity to get feedback from unknown customers is interesting here. A shoe manufacturer could e.g. put a QR code on the box and request praise from the end customer.

Having an NPS system that can ask metadata questions (eg asking for name and email for follow-up) also provides new interesting opportunities.
You can read more about how measurements can be put into a system or programmed, in the next article “How do you put power into operational customer loyalty?”.

Content and wording of an NPS campaign
Before measurements are initiated, it is always a good idea to communicate to the recipients that measurement will take place, why and how it will be followed up. It can e.g. happen in the newsletter, on the website, in a prior email, etc.

The ideal NPS campaign consists of:

• Intro text: Eg. “We value our collaboration and want to create the optimal customer experience. Therefore, we hope you will spend a minute sharing your experiences ”
• The NPS question that is ideally formulated as: “How likely is it that you will recommend xxxxx?” “How likely are you to recommend xxxxx?”
• The follow up question, that should be reflecting according to the score to motivate comments: the score to motivate comments:

Detractor 0-6: What do you think we should do better?

Passive: 7-8: What should we do in order to get 9 or 10?

Promoter 9-10: What do you think works well?

• Completion text, that again can vary depending on the score, for example:

Detractor 0-6: We regret that your experience has not been optimal and will contact you as soon as possible to clarify solutions

Passive: 7-8: We are pleased that you are satisfied and we will continue to work to make it even better

Promoter 9-10: We are very pleased that you are satisfied and will recommend us to others. We will do what we can to continue like this (some are daring and then refer Promoters to eg Trustpilot in here in the completion text)

Useful information:

If there is no email permission on a contact, the NPS campaign must not contain marketing or sales-related elements or undertones. However, NPS campaigns with email permission should also not contain these messages as it removes focus from the real intent and seems disruptive to the recipient. Therefore, it is a good idea to keep both wording and appearance as neutral as possible.

When measuring in touchpoints and transactions, the NPS question can be reformulated so that the respondent understands that the measurement takes place specifically in relation to a specific customer experience and not in the general relationship (the overall experience). Very specific questions can also be asked, such as “To what extent did the service technician live up to your expectations?”. Then it is no longer a “pure” NPS measurement. nps.today supports both specific issues and the more official NPS methodology. Here it is just important to handle reporting on general loyalty (relationsNPS) and touchpoint or transactionNPS correctly.