Cross Tabs and other marketing research techniques

Business is tough. And a company needs everything it can get in order to get a leg up on the competition. As a result, businesses are relying more and more on market research to better determine the direction their company will go in. The task of market research analysis, then, is to provide these companies with relevant, accurate, reliable, valid, and current information, often in the form of graphs, charts, or tables. Thankfully there is a wide variety of analysis techniques and formats that can be used to glean this information. Below is a list of a few market research techniques that help analysts compile information.

Frequency Distribution tells you how many people chose a particular response to any single question.

Survey Cross-Tab Analysis allows you to see how responses to one question work in conjunction and/or comparison to another. For example, you might want to know what percentage of purchasers of your product also purchased a competing product. This is a powerful method giving a maximum amount of flexibility to the analyst.

Average by Category lets you compare the average value response for different, specifically defined groups. For example, you could compare the average age for men and women, or for full time employees and part time employees.

Cross Tab Means combines the two methods above creating a more complex measurement. For example, you could see how overall satisfaction is rated by men who own a Toyota.

Segmentation lets you create groups of customers who share similar traits or behaviors. This allows marketers to zero in on specific customer groups and market directly to them.

Gap Analysis allows a business to see how large a gap there is between what it is currently offering and what its customers want. For example, a survey might ask customers of an auto repair chain to rate four different things: service, quality, value, and reliability. Survey results show that “service” is consistently rated lower than the other factors, so this is the area that offers the greatest opportunity for improvement.

From simple frequency distributions to the more complex gap analysis to the extraordinarily powerful cross tabulation, these methods allow analysts to glean important information about their products and customers that can help them guide the company to create new and better products and serve their customers’ needs better.

Is Customer Segmentation Analysis still relevant?

Today, it is possible, certainly in a telecommunications, banking, retail or other online environment, to collect a vast amount of information about your customers. The question is, what sort of survey analysis to apply to it. Additionally, is customer profiling and segmentation still a good approach to organizing this data and to then interacting with groups of customers?

Of course, there are many approaches to using your accumulated data, some of which may be better than segmentation in delivering sales or marketing returns, or improving other areas of customer relationship management.

Nowadays, companies such as Amazon, tailor their advertising to individuals searching for books online: a technique known as ‘mass personalization’.  And for many companies, this approach is gaining more emphasis than traditional techniques.

Amazon is frequently cited as an example of good personalization.  As an on-line retailer, they are well positioned to take advantage of such techniques, but there are remarkably few good examples of mass personalization in an off-line environment (I do not remember any direct mail in my mail-box, for example, which gave me the feeling that I was very targeted).

The truth is, a lot of the time, companies sit on a goldmine of information and fail to exploit the opportunities for powerful communications with their customer base, or the opportunity to create highly desirable products and services based on the information they have available.
Profiling and customer segmentation do work – the problem is, a lot of companies don’t use them effectively and so end up seeking the next big thing.

Personally, I do not believe that companies will totally move from a segmentation based approach to a mass personalization approach right away, but it is true that the targeting power that can be gained by combining traditional segmentation analysis (based on large volumes of historical data), with the real-time analysis of what a customer is doing on the Internet, plus where they are located (when they are using the Mobile Internet) cannot be under-estimated.

So it seems that both traditional segmentation and the newer mass personalization methodologies may have their benefits and while, for some companies customer segmentation may no longer be required, I feel safe in the knowledge that for many, it is still a valuable tool and one that will remain relevant for quite some time.

PAI would welcome the opportunity to demonstrate how PAI’s mTAB™ service can help you to establish and analyze your own customer segmentation of the marketplace.  Please visit the PAI Website to learn more about us, or to schedule a no-obligation demonstration.

Pivot Tables – what are the alternatives?

Creating concise, informative summaries out of huge lists of raw data is a common task and spreadsheets of old have long troubled our weary eyes and tired minds with endless rows and columns of unserviceable data. Digging deep into the data with a standard spreadsheet required knowledge of spreadsheet formulas, math skills, and a tedious mind. The powers that be created the pivot table to replace this unnecessary work, forecasters everywhere rejoiced and were glad.

Data spreadsheets can contain a lot of information and sometimes it may be difficult to obtain summarized information in a simple fashion. Let’s say you’ve compiled a long list of data – for example, sales figures. A pivot table is a spreadsheet tool that gathers cross tabulation data and allows the user to select exclusively desired information from the columns and rows such as individual products or a sales date. Not only will a pivot table highlight desired information, but it can easily “pivot” variables around its axis to gain a different perspective on the data you are working with. The spreadsheet is now in the power of your hands.

The pivot table you will be working with will look and function similarly to the spreadsheet template, but there are many more functions behind those columns and rows. Once data has been populated in the pivot table spreadsheet, it is time to run the data and view the numbers. Sums, averages, standard deviations, and counts are just a few functions the pivot table can produce instantly! Pivot tables make the data seem easy to understand and they can be used within any level of data mining: nominal, ordinal, interval or ratio. The pivot table will become a good friend in helping prepare for reports since its functionality in a spreadsheet allots for more research time, and perhaps, resting those weary eyes and tired minds.

Essentially, a pivot table is a cross-tabulation report – an analysis type that survey research analysts have long known the benefits of. The few differences between the pivot table and a cross-tabulation report are seen in the pivot table’s deficiency of research power. The pivot table establishes an interdependent relationship between the two tables of values but does not identify casual relationships. Also, the pivot table’s power is indicative to the survey size. For example, a 20 question survey with a sample population of 100 can be effectively analyzed with pivot tables within a spreadsheet but when sample population and survey size increases over time or between markets then a more powerful, specialized survey analysis tool is required.

Analyzing survey results: filtering with multiple response questions

Filtering is one of the most common and useful means of analyzing survey data, focusing the analysis of survey results to a particular subgroup of survey respondents, such as first time purchasers or families with children under twelve.

Analyzing survey results with filtering can become complicated when multiple response (i.e. check all that apply) questions are included within the filter criteria.  Consider the multiple response survey question illustrated below:

Now consider defining filter criteria with your survey analysis software, checking the McDonald’s and Wendy’s restaurant checkboxes as shown above.

The survey analyst needs to understand that the subgroup of respondents passing the filter criteria may have visited other, and even potentially all of the other, quick service restaurants. A more accurate definition of our example filter would be that the subgroup of respondents visited AT LEAST McDonalds or Wendy’s.

Survey analysis software should equally support identifying respondents that indicated McDonalds and Wendy’s as their ONLY visits, without the need to construct complicated filter criteria that lists all of the individual restaurants.

Software tools well suited for the analysis of survey data will support more complicated filtering criteria such as “visited at least two Mexican food themed restaurants”, “visited at least one seafood themed restaurant and only seafood themed restaurants”, “visited only burger restaurants” and “visited McDonalds and Wendy’s but not Burger King”.

PAI’s mTAB survey analysis software has built in support for multiple response questions, which greatly simplifies the analysis of survey data containing this common survey questionnaire type.

When multiple response questions are selected for respondent filtering, mTAB automatically exposes the multiple response question filtering options that facilitate the example analyses listed above.  For a more detailed illustration of mTAB’s multiple response filtering, please visit the mTAB knowledge base article respondent filtering with multiple response questions.

Which Data Mining Application to use?

It seems that no matter what industry it is, everyone is trying to find the best means to analyze and draw actionable conclusions from their data. Data mining has grown to be the cornerstone of modern-day business operations. By clinical definition, data mining is defined as “the process of extracting patterns from data.” This definition is far from precise, but it does a great job of illustrating how tough the analysis process can be. In many cases, it is labor intense and time consuming, as if you were actually using a pick and axe.

So which data mining application should you choose? The most powerful? The most user-friendly interface? The most extensive features?
The best data mining application is the application that best fits both your analytical needs and the skill set of your employees. Analytical needs can vary based on the level of analysis you desire and the type of data being utilized. For example, the needs of analysts working with social media data, financial reporting, and survey research data are all different. Applications may provide differences in analysis features (statistics, verbatim reporting, etc.), file size limits, and source data flexibility.

Simultaneously, your employees need to be able to use the application effectively. One of the chief concerns in implementing any new system in the workplace is the amount of time required to adopt and productively use the software. Systems with user-friendly interfaces an

d intuitive functionality can ease the adoption process, while complicated programs may inhibit worker productivity, and in some cases, result in poor or erroneous analysis. And, as employees turn to telecommuting and team-based work, applications that can incorporate cloud co
In short, a multitude of software packages are available for data mining. Each have their own strengths and weaknesses, but all that really matters is which package best suits your objectives. And in an environment where speed, accuracy, and insights are key to success, why not let the software do the heavy lifting?mputing or collaborative capabilities have an advantage going forward.