Survey data analysis – Text Analytics for Net Promoter Surveys

As a consumer insights manager at a national retailer, you are responsible for understanding the wants and needs of your customers.  You’ve followed the sage advice of the Net Promoter Score (NPS) experts, and you’ve implemented an online NPS survey capturing “Would recommend..” that allows customers to leave open-ended comments.

In our prior posts we’d discussed how to join internal “structured” data such as retail location, transaction date/time, payment type, check amount, etc. to your survey results, segmenting the NPS scores to identify  the customer segments requiring the most attention.  But how do you make sense of a 100,000+ open-ended customer comments?  

You can segment your open-ended or “unstructured” customer comments using your “structured” data in the same manner as your NPS scores.  For example, we could focus our review of customer comments to the customers paying with a credit card within the Eastern region reporting an NPS score of 50 or less. Unfortunately, national retailers will likely find that they still have hundreds to thousands of comments to review even within very focused segments.

We can use text analytics to help make sense of survey comments without having to read every individual comment. Term frequency (TF), or the percentage in which a selected term occurs, helps gain an understanding of the important concepts underlying the comments, just like you would use the mean of a series of numbers to gain an understanding of a numerical series.  Term frequencies can be conveniently analyzed using a selectable tag cloud graphic as we’d illustrated in a previous post.

When TF analysis is combined with segmenting the comments using our “structured” data segments, we can compare the term frequencies between various segments to gain an understanding of the relative importance of key concepts within each segment.

We can further refine term frequency analysis by normalizing our TF terms using inverted document frequency (IDF) analysis. TF-IDF analysis goes a step beyond TF analysis by considering both the frequency of occurrence of a term with a individual comment (TF) as well as term’s frequency of occurrence within either all comments or a segment of comments.

We can then use TF-IDF analysis as the starting point for determining the relative similarity of comments. This additional step allows us to call out a small set of comments, for example, the top 20 comments, that are the most representative of the larger group. These top 20 comments serve as an “executive summary” of the overall set of comments, summarizing the key topics and issues without the need to individually review the entire series of comments.

We would welcome the opportunity to help you to reveal the hidden insights within your NPS survey results.  Please follow us on Twitter http://www.twitter.com/mTAB to receive updates to our ongoing survey analysis series as well as more information pertaining to our mTAB survey analysis service.

Develop Customer Retention Strategies from Analysis of Survey Data

Challenging economic times combined with easy access to the web have made it more difficult for businesses to gain and retain customers.  According to numerous studies, the cost of a new customer acquisition is ten times more than retaining one. Now more than ever, it is important to understand how to develop long-term relationships with your customers.

When it comes to bottom line figures, the thriving companies are the ones reporting high customer satisfaction and customer retention rates. There is no denying the direct correlation between satisfaction levels reported on customer surveys and retention rates. How you analyze survey data will play an important role in the development of effective customer retention strategies.

Strategy from Survey Data Analysis

The companies that are succeeding constantly analyze their client relationships and develop strategies based on identified trends.  A prime example is identifying factors that lead to customer attrition.  By learning the shared characteristics of clients you have lost, you can identify those most likely to leave next, and develop techniques directly aimed at retaining these “high risk” customers. The ability to focus your marketing strategies appropriately to different client groups can increase your retention rate and the lifetime value of each customer.

A vigilant program of survey data analysis can give a company the edge it needs to understand how customers respond to various factors in their business relationship. Fortunately, advances in survey and data analysis technologies have made it easier to scrutinize the wealth of information needed to develop future strategies.

Qualitative Research – Tips on how to use Unstructured Research Data

In my previous blog article ‘mTAB Understands Qualitative Research Needs!’  I discussed the need, or advantages, to utilizing qualitative, or unstructured, data when reviewing research results; now I want to help you learn how to implement this suggestion.

I have always visualized structured response data in my imagination as a city with sky-rise buildings and lots of traffic; there are mathematical equations creating a functioning city and it works as well as we need it to. In my imaginary city I visualize unstructured data as a reservoir of gold in the underground tunnels that the city is built on. Burgeoning cities have a tendency to expand, and expansion will generally follow the basis of the original model, but improvements are made by utilizing the underground reservoir.

Yes, fanciful, I know. Keep following me, structured response data gives you an answer to exactly what you’ve asked, but how did you decide what to ask? Have you based your survey response questions on your desires, or your consumer’s desires? Unstructured data allows for responses that are unsuspected and can lead to great improvements.

Let’s use an example: Car manufacturing. The survey asks for demographics, model preferences, etc. Generally an unstructured data response will be positioned as a means of gathering more information about a previous structured data question; e.g. ‘why did you answer yes to this question?’ This is brilliant because the survey is looking to dig deeper into reasoning, however, the full potential of unstructured data is not being utilized.

To harness the power of unstructured data one must think outside of their normal understandings. If you work in the automobile industry you immediately take for granted the need for automobiles. This assumption is the basis for the rest of your survey. Now, if you realize your previously held assumptions about your field and bring them into question, you may find that reservoir! Survey respondents don’t want to write a book while filling out your survey, so you’ve got to choose your open-ended non structured questions wisely. Some examples may be: why do you drive a car instead of taking mass transit? If money was not an issue what car would you buy, and why? What is the most important aspect of a car in your opinion? If you’ve ever considered fuel-alternative transportation, what factors concerned you? Are there any questions that you would want to ask us?

Asking your customers questions that you haven’t assumed answers to will result in responses that you couldn’t have anticipated, and insight you may have overlooked. By incorporating unstructured data responses in your survey you may just find the seedling of an idea that will greatly improve your product, and also give your customers the satisfaction of knowing that you’re interested in their thoughts.

Please be sure to follow my blog posts regarding qualitative survey analysis.  Keep an eye out as well for text analytic posts that we’re planning to run in the near future.

Correspondence Analysis – what does it all mean?

One of the most commonly used forms of Brand mapping is correspondence analysis. Brand maps are often used to illustrate customers’ images of the market by placing products and attributes together on a map together on a graph. This allows close interpretation of company perceptions with a variety of product and service attributes that are closest to them on the map. If products are placed close to each other on the graph, it means they have a similar image or profile in the market. This helps to inform strategy, like where to take the client’s product in the future.

Some questions that can be answered with a correspondence analysis are:

- What attributes does the brand own?
- What attributes do competitors own?
- Are there gaps in the market that may be filled by the client brand?
- How should the brand be positioned to be both relevant to the market and differentiated from the competition?

Correspondence analysis is an exploratory data analytic technique that allows rows and columns of a cross-tabulation (two-way and multi-way) to be displayed as points in two-dimensional space. The information from the cross-tab, as seen in the example below, is usually collected using simple multi-coded grid questions or semantic rating scales (typically 5 or 10 point scales).

It reduces a complicated set of data to a graphical display that is immediately and easily interpretable. The graph contains a point for each row and each column of the cross-tab. Rows with similar patterns of counts produce points that are close together, and columns with similar patterns of counts also produce points close together.

Running a correspondence analysis on the above cross-tab yields the following graphical display:

In this example, McDonald’s is more synonymous with Low Prices. Pizza Hut and El Pollo Loco can be associated with Overall satisfaction, as well as Taste of the Food and Quality of Food for the Money.

Correspondence analysis is a valuable tool that can be applied to many situations.