Explore the Four Levels
of Text Analytics
key takeaway
Best Practice
We suggest using text analytics software to screen customers’ answers to the open questions in your surveys, then letting humans make the final decisions about priorities.
Here are the four levels of text analytics that we recognize:
- Single word concepts are often shown as word clouds and are marginally useful. Yes, your customer talked about ‘service’ a lot, but was it good service or bad service? And what service are they referring to? ‘Service’, ‘Services’, and ‘Support’ will appear as separate items.
- The addition of basic language structure recognition provides multi-word concepts such as ‘Deluxe hamburger’ and ‘high-end burger.’ Note that these will be shown as different topics.
- The next level adds sentiment into the equation. The multi-word concept is qualified in some way, usually in terms of positive or negative sentiment We start to see themes, usually grouped in terms of what customers liked and did not like. ‘The air conditioning was too cold’, and ‘The whole family was freezing’ would appear on a long list of themes with negative sentiment but would not be grouped together.
- The addition of a taxonomy gets to the heart of Natural Language Processing. This allows the grouping of topics that mean the same thing but use completely different words. ‘The clown frightened the children’ and ‘The main act was unsuitable for young people’ would be grouped together, as the software will recognize that they probably mean the same thing.
It’s a sophisticated area. Many software providers will produce a free analysis for you. We suggest providing them with the largest file of open text responses you have and comparing what they can do with them.
go beyond basic nps analytics
Just as you can only act on what you know, you can only analyze the data you have. Even the most comprehensive, multi-part attitudinal data will only ever cover a fraction of your customer base. Visionary programs use analytics that incorporate machine learning to transform operational data into customer insights, with attitudinal data as a critical tool for calibration. To get there, download our guide, take our comprehensive training course, or read about the future of NPS.
REPORT
The Complete Guide to NPS Basics and Beyond
TRAINING
Leading an Outcome- Oriented
CX Program
REPORT
The New NPS
Manifesto
REPORT
The Complete Guide to NPS Basics and Beyond
TRAINING
Leading an Outcome- Oriented
CX Program
REPORT
The New NPS Manifesto
ABOUT OCX COGNITION
OCX Cognition delivers the future of NPS. We ensure customer experience success by combining technology and data science with programmatic consulting. In our Insights section, we present a comprehensive and evolving collection of resources based on our research and expertise, collected for CX leaders committed to delivering business outcomes.