Open-ended responses can be tricky data to work with for market researchers.
There is undoubtedly value in asking these questions of respondents in online surveys, but the insights are often less obvious.
By comparison, listed answer options in a survey can be easily analyzed by just about any data management tool.
It is no wonder text analytics is increasingly growing in popularity.
What are the actual advantages of using text analytics?
Below are five tangible benefits you can expect from the top platforms available to market researchers.
Text Analytics vs. Manual Coding
Traditional approaches for open-ended responses include manually coding the data into distinct buckets or listing out all the responses to be read.
However, coding takes time and listing answers doesn’t offer quick takeaways from the data.
To achieve both a quick timeline and useful insights, text analytics may be the best option for many surveys.
For a more detailed rundown of what you can do with open-ended responses, read about your options here.
#1: Identify critical topics in the data
Perhaps the closest replacement for manual coding, an automated topic assignment is the primary benefit to text analytics.
Tools often use artificial intelligence (AI) to find themes in the open-ended text responses by grouping similar terms in respondents’ answers.
Curious about other uses of artificial intelligence in market research? Read more about AI in text analytics and other research applications here.
The tool performs the tedious job of pulling to the top 10 to 15 topics based on the open-ended responses.
You can expect some level of customization in the number of total topics as well as merging similar topics together.
This all helps the researcher to easily quantify the responses and see what overall themes resulted.
#2: Eliminate noise in the data
Just as text analytics identifies key themes, it may also weed out less important words and phrases to the analysis.
This may come in the form of an option before running the analysis or a customized filter after the analysis is complete.
You can typically exclude common words such as “the,” “and, ” or “is.”
Terms also worth leaving out are obvious words that were used in the original question, which may simply be restated by respondents.
The AI aspect of text analytics should be able to stem similar words under the same topic such as “buy” and “bought.”
#3: Pinpoint drivers of positive sentiment
Once the topics are established, many text analytics tools allow users to run a sentiment analysis on the open-ended responses.
Positive sentiment, for example, can be measured by calculating a net value from predetermined values assigned to most English words via natural language processing (NSP).
Researchers can see what percentage of responses for each topic have a net positive sentiment.
This indicates what words and phrases may be most associated with happy or optimistic feedback.
In terms of ad testing, topics with the highest positive sentiment scores can help you feature terms and imagery that resonate better with our target audience.
#4: Understand sources of negative sentiment
Text analytics can just as easily provide insights into the topics with the highest negative sentiment.
Some tools even specify different types of negative sentiment such as hate, fear, and skepticism.
Identifying these topics is equally important but for different reasons than positive sentiment.
Knowing which topics have the most negative associations gives researchers clues as to where something is falling short.
For example, text analytics may reveal that certain traits of a product are often mentioned in the open-ends in a negative light.
The next step for a researcher and client might be to dig deeper into whether the negative sentiment is the result of a misunderstanding or general discontent for the trait.
Future messaging for the product may benefit from explaining the trait in an alternative way or removing it altogether.
Recommended Reading: Sentiment Analysis and Social Media
#5: Cross topics with other key metrics
Lastly, the assigned topics have even greater value when applied to other data in the survey.
While the underlying themes from an open-ended question are useful, you may unveil nuggets of information by crossing them with key metrics measured in other questions.
These metrics may be rating scale questions from the survey such as Net Promoter Score (NPS), willingness to buy, or likelihood to renew a contract.
This feature offered by some text analytics tools gives a researcher the ability to tie the most important metrics back to the open-ended responses.
Learn more about Net Promoter Score in market research.
Disadvantages of Text Analytics
Like anything else, there are always some trade-offs. At this time, text analytics tools are not standard in data analysis tools.
This often means there is an additional out-of-pocket expense for researchers and end clients.
There may also be a question of the tool’s accuracy because automated processes aren’t always better than a human.
In many cases, though, the time and work saved through text analytics help justify the investment.
Utilize Text Analytics with Drive Research
Drive Research is a national market research company located in New York.
Our market research firm leverages the power of text analytics for many clients to uncover insights beneath the surface of open-ended responses.
Interested in learning more about our market research services? Contact us today!
- Message us on our website
- Email us at [email protected]
- Call us at 888-725-DATA
- Text us at 315-303-2040
Tim Gell
As a Research Analyst, Tim is involved in every stage of a market research project for our clients. He first developed an interest in market research while studying at Binghamton University based on its marriage of business, statistics, and psychology.
Learn more about Tim, here.
