The correct use of Smart Labels in NPS Analysis

November 21st at 11:01am by Susan Levermann

Those who hear about NPS often consider just the score, and totally forget that NPS Platforms like zenloop offer a not just an NPS measurement tool but also a complete NPS system. The advantage: an automatic analysis of feedback. How our smart labels feature helps with NPS analysis is explained in this article.

What are smart labels?

In our NPS Platform zenloop, smart labels are our tool for the administration and understanding of feedback in the broadest of terms. They form the starting point for semantic and uncontrolled text analysis. That means that the incoming answers are automatically labeled and can be categorized on arrival.

Why should you work with smart labels?

Thanks to the use of smart labels, the clustering and analysis of this valuable feedback in the zenloop NPS system is fully automated. Smart labels therefore offer an innovative solution for NPS analysis. This kind of fully automated segmentation is to date unique and allows you to cluster and theme feedback and way you wish.

How do smart labels work in an NPS Analysis?

Smart labels are a part of the AI that we at zenloop developed. Our algorithms mimics hereby human intelligence in it’s behavior.

When our software receives feedback from your customers, every single answer runs through the same process. We call this “Smart Label Flow”. During this flow customer comments automatically receive the appropriate smart label.

The smart label in detail

Language detection:

The first stage of the Smart Label Flow is language detection. The AI recognizes the feedback language and gives this information, along with the feedback to the next process

Parser:

The second step is the parser. This checks the feedback for grammar and analyses their meaning. You could say that the parser digests the feedback and extracts the message.

Removing of stopwords:

n the next stage the AI removes stopwords. These can also be described as “filler” words, which give structure but no further information to the message. This makes them superfluous to the labels and these words are then removed. The AI does this to just the right degree to keep the meaning of the feedback.

Stemmer:

The steamer has the job of further simplifying the feedback whilst retaining all the necessary information. For example, conjunctions, declinations or typos are removed and replaced with the base form of the word. For example, from “cars” the steamer would make “car”.

Keyword matching:

In the penultimate step of the Smart Label Flow keywords in the extracted information from the previous steps is matched to smart labels. If a message doesn’t fit to any particular keyword then no label is attached to the comment. This also means that some messages can get numerous labels, if these match up with the right keywords.

Labelling:

As soon as the matched keywords are identified, the individual comments are labeled accordingly. This is the basis for feedback analysis and the process of making customers into fans of your company.

How do I work with smart labels?

In order to gain deeper insights into customer feedback, labels can be combined with attributes and/or scores. This makes it easier for you to recognize the strengths and weaknesses from your customer feedback.

With a direct message, every customer comment can be immediately responded to. If feedback arrives which absolutely must be shared with or answered by a specific part of your organization, zenloop users have the possibility to forward this feedback to the appropriate person or department. Automated e-mails or other alerts, for example through Slack, Zendesk or similar can also be easily set up.

As previously mentioned, smart labels allow for feedback to be clustered by topic. This makes it possible to filter feedback according to label. This allows even complex associations to made between several areas. For example, you could choose to view all the answers which are labeled with quality and pricing.

Through the addition of new keywords to existing smart labels, the definition of those labels can be continually refined. This makes filtering and analysis ever preciser.

As well as all this, zenloop users have the possibility to create new smart labels which explicitly relate to their content and feedback. During the creation of these smart labels, careful attention must be paid to which keywords are used to define the labels.

The advantages of smart labels

The use of smart labels brings with it many advantages. The times when all the answers - sometimes from differing touchpoints - had to be sorted in order to gain an overview are long gone.

With zenloop it is now possible to filter all the appropriate answers and the associated NPS score at the click of a button, and then use our analysis tool to view the development of sour NPS.

This makes it much easier to generate a valid prognosis or to develop positive processes which affect the whole of your company. Smart labels are therefore very much essential for your future prospects and analysis of sour company.

Check out zenloop


Susan Levermann

Content Marketing Manager