What’s Your Text Analysis Plan?
Why ask for comments if you don't analyze what customers say?
Surveys, chats, and interviews... the greatest insight comes from customers’ verbatims. But only if you have research-driven consulting for text mining to extract the meaning in the comments. With our scientific Text Analysis consulting, you’ll know in detail how to grow customer loyalty.
Discover Your Options
See examples of Text Analysis options & strategies. Find what’s best for you.SEE EXAMPLES
Review The Plan
We design a custom coding framework for your open-ended survey comments. When you love it, we start.
Impress Your Boss
Monitor your text mining portal. Share insights and actions with your team!
Find out: The 5 Things You Need for Great Text Analysis ➔
Analyst Raven Susu-Mago explains Text Analysis
Length: Under 2 minutes
- The value of open-ended survey questions
- The problem with merely reading the comments
- Our Text Analysis advantages
- Client Example
- Interactive Live Dashboard
Find out: The 5 Things You Need for Great Text Analysis ➔
Interaction Metrics’ Text Analysis: Your Benefits
- Red alerts identify customers that need immediate action.
- Metrics show your progress in real time.
- AI software analytics become smarter and more specific.
- Instead of a staff member reading comments, you have an objective approach.
- Proven protocols build trust with stakeholders.
- Your future surveys will ask more relevant questions based on what’s most important in your customers’ minds.
- Cultural nuances are included in our analysis.
Ready to Unlock the Insights in Your Customers’ Data?
Are you sitting on your customers’ chat logs, emails, and other feedback, unsure how to analyze it meaningfully? Good news; you’ve found the right team!
Interaction Metrics’ Text Analysis is the solution for extracting objective insights from troves of customer comments, conversations, and other feedback.
Contact us to find out what our Text Analysis can do for you. In just 25 minutes we can start to establish a Research Plan that fits your data.
The best way to understand your customers is to take their words seriously. Companies routinely collect customers’ comments in surveys, chats, and support calls. Then they fail to analyze this trove of feelings and thoughts.
Could your customers be buying more—or buying more often? Perhaps it’s time upgrade your customer listening! There’s no better way to do this than with our Text Analysis.
3 Use Cases for Interaction Metrics Text Analysis
- Improve AI Text Analytics: If you are using AI software to decipher the meaning in your customers’ comments, it is critical to have Research Analysts code against your AI to check for accuracy and improve insights.
- Survey Insights: If you have a survey and want to understand why customers are ranking you the way they are, Text Analysis is an essential addition. Read more about how to improve your open-ended survey questions here.
- Other Verbatims: Emails, chats, customer reviews, and customer service conversations are all sources of text that’s worthy of Interaction Metrics Text Analysis.
Not sure if Interaction Metrics’ Text Analysis is the right solution for your data? Contact us to schedule a short session. In less than half an hour, we can start to establish a custom Research Plan based on your company’s needs and objectives.
The Beauty of Unstructured Data
Structured data is organized neatly into a searchable, ordered pattern, like a rating system where respondents give a numeric ranking, or “yes/no” answers.
In contrast, unstructured data doesn’t follow a predetermined schema; it can include videos, audio files, social media, and chat conversations. In fact, around 90 percent of data is unstructured, according to the MIT Sloan School of Management. In customer experience, a survey’s open-ended comments, along with phone calls, chat logs, and interviews are all sources of unstructured data.
Unstructured data can be intimidating to analyze because it doesn’t follow the ordered schema of structured data. How do you create a pivot table of 8,000 customer comments and then extract actionable conclusions to present to your CEO? Often, companies just ignore the unstructured data in their hands.
But unstructured data has a power and authenticity that structured data doesn’t, especially when it comes to understanding your customers. When your friend asks you whether you enjoyed the new restaurant you tried last night, how do you respond?
- “I give it a six out of 10?”
- “The ambience was great, but parking was hard, and our food came out lukewarm! I wanted to like it, but I don’t think we’ll be back.”
Rating questions give you scores, but customers’ responses to open-ended survey questions do the heavy lifting.
Comments get to the heart of the customer experience, often revealing issues you didn’t even know to ask about—and showing what to improve and for whom.
Can’t You Just Read the Comments?
Ok, unstructured data is clearly valuable. But do you have time to read through 8,000 customer comments? Of course not. Besides, it’s not enough to read customers’ comments, even if you had the time, patience, and mental fortitude to do so.
The problem with merely reading comments is that the brain’s working memory capacity starts cutting off at around seven items. This limit to our short-term memory is known as Miller’s law, after the Harvard psychologist George Miller.
Although researchers have since developed other understandings of short-term memory, Miller is correct that our cognition is quite limited. Even if you have time to read thousands of comments, there’s really no point to doing so, because you simply can’t synthesize all that information. You’ll overlook subtleties and topics you’re not prepared to recognize.
Even worse, you can’t quantify anything about your text data, and you won’t have a compelling report. Business audiences demand numbers, not meandering stories.
Text Analysis Coding
Text Analysis coding is the process of systematically categorizing and extracting information from customers’ unstructured data. It applies frameworks to identify patterns, themes, and specific features within a body of text.
Throughout the history of social science, anthropologists, psychologists, and sociologists have been coding narrative data making coding a proven technique for deriving meaning from qualitative data—and it turns out that it’s highly useful in customer experience science, too. For many companies, coding is the most effective way to extract meaning from their survey verbatims and other unstructured data, like interviews and reviews.
So How Does our Text Analysis work?
Several Analysts work together to build a coding framework. The framework is built iteratively, going through multiple cross-checks until it captures the true meaning of the text. Then, that framework is used to classify comments by their various elements, getting progressively narrower and narrower in specificity.
Once the coding is in place, we quantify the codes to reveal themes in rank order priority. Quantification is a crucial step because it allows us to correlate your verbatims’ themes with other outcome scores like the Net Promoter Score.
The key to our coding is our experienced researchers who know how to delineate subtleties and interpret comments accurately as they code your customers’ comments.
You Deserve More than Word Clouds
Some companies turn to word clouds because they are a default chart for survey platforms. Word clouds are the cheapest and easiest “insight” you can glean from your data: just input the entire spreadsheet of customer comments to see which words occur the most often.
But word clouds don’t find the story in the data, and they don’t show what actions to take. Perhaps the word “product” occurs most often in your word cloud. That doesn’t tell you whether customers are satisfied or displeased with your product.
You deserve more than word clouds—and that’s where our Text Analysis comes in.
Text Analysis: The Deep Dive
Customers often write their comments informally, and it takes some work to categorize them before we can arrive at the right tags to label them accurately.
Here’s a short guide to the six-step process of coding for Interaction Metrics Text Analysis:
- Substantiveness: Is the comment intelligible enough to be coded objectively?
- Class: Which department does the comment belong to?
- Type: Is the customer voicing a complaint or explaining why they want to keep existing processes unchanged?
- Sentiment: Is the customer frustrated or happy? Although this is the easiest category for researchers to determine, it can be tricky for many AI-based solutions when customers have mixed opinions — which they often do.
- Tag: What is the comment about? Although this is the most important part of the entire classification system, the preceding steps are necessary for the tag to have any meaning.
- Sub Tags: What particular products or services does this comment address?
So what’s the advantage to our rigorous Text Analysis system? It eliminates subjectivity.
When multiple analysts examine the data independently and arrive at the same conclusions, you have a replicable system in which subjectivity has been removed.
Ensuring objectivity fulfills a critical pillar of science and makes for sound customer feedback research.
In a Nutshell
In a nutshell, without Text Analysis, it’s impossible to get an objective reading of customers’ comments. With Text Analysis, you can make changes to your customer experience knowing that your actions are based on an accurate understanding of what your customers actually want and need.
Start Mining Your Data for Valuable Insights Today!
Contact us to see how Interaction Metrics’ Text Analysis can add value to your survey today. We cover a lot of ground in just 25 minutes. Get in touch today!