Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.

Thoughts on Quantifying Qualitative Data

928 vues

Publié le

Ways to quantify qualitative data to gain insights

Publié dans : Données & analyses
  • Login to see the comments

  • Soyez le premier à aimer ceci

Thoughts on Quantifying Qualitative Data

  1. 1. Thoughts on Quantifying Qualitative Data A Brief Overview by George Sloane
  2. 2. Data, Data, Data—We are Awash in Data  The real trick is using the data to better understand and improve our business  Most analysts are great with quantitative data, but qualitative data can be more challenging George Sloane 2
  3. 3. Quantitative and Qualitative Data  Quantitative data is relatively easy to:  Collect  Measure  Analyze  Manipulate  Chart  On the other hand, qualitative data is more difficult to handle. The thing is qualitative data can be like gold when seeking insights into products, customers, and markets  Balancing an analysis with a mix of both quantitative and qualitative data can be extremely powerful George Sloane 3
  4. 4. How do we Transform Qualitative Data into a Usable “Quantified” Format?  The simple answer is to find a way to group and standardize the data into a number  Let’s look at some simple examples . . . George Sloane 4
  5. 5. Example #1: Company or Product Media Perception  We want to know how a company is perceived in the media. The media publishes articles and reports about the company and its products-- we want a solid way to measure that media perception  What if we took a timeframe, like a fiscal quarter, performed a literature scan of the media and used a point scoring system to each article or report in a specific timeframe? A positive media report is +1, a neutral media report is a 0, and a negative media report is a -1. Next, we add up the scores and look to see if we have a positive or negative score. Furthermore, we could identify trends by looking back several quarters. Additionally, performing exercise on competitors over the same time periods would provide further valuable insights  Don’t want to use a scoring system? Consider a ratio—positive to negative reports in specific timeframes over a time series. Either way, we quantified qualitative data and developed a way to compare ourselves in an apples-to-apples fashion against ourselves or competitors that we can track and chart. The beauty is that number of articles doesn’t matter much*—we want the net “mood” that we can measure  *One word of advice: Common sense dictates that this approach relies upon having several articles to rate. Clearly, one or two articles per period aren’t enough for a good “read” George Sloane 5
  6. 6. Example #2: Stakeholder Influence  We want to compare stakeholder influence in a group of publicly traded companies. One approach would be to look at the ratio of inside to outside directors of the board of directors at a given point in time and in a series of given points in time. Inside directors are officers of the company. An inside director might also include a major shareholder. Outside officers are not connected to the company other than their role as a director. Again, we can then compare one company to another on a level basis George Sloane 6
  7. 7. Example #3: Sales Force Focus  If we want to look at customer satisfaction by opinions, we once again need to code our data to translate it into a quantitative format for analysis. Assume we asked a simple satisfaction question at the end of a transaction. We could use happy face, neutral face, and sad face graphics tied to coded words from a response to a survey question where answers choices were exceeded expectations, met expectations, or disappointed George Sloane 7
  8. 8. Example #4: Simple Customer Satisfaction Data  What if we wanted to know what the sales force focused on? Let’s assume management is concerned about client/customer revenue attrition. It would be helpful to identify the sales force’s focus, beyond their expected, but natural attraction to the high commission sales. What if we gathered sales data and assigned an “A” to a new customer relationship sale, a “B” to additional new business from an existing customer, a “C” to a repeat sale equal to the last period measured, and a “D” to reduced business from an existing customer, and a “F” to a lost customer? We would than be able to grade our sales force or even individual sales team members with a frequency distribution  If we used number grades (A weighted average GPA), instead of letter grades, we might allow for those sales people who excel at pure rainmaking and those that succeed at squeezing more from what we already have—both types of sales team members can provide for success  Furthermore, using these techniques, we would see those sales people that need new incentives and adjustments to the way they are compensated to avoid highest commission sales, at the expense of what is needed for long-term organization success based upon margins George Sloane 8
  9. 9. Conclusion  In all of the examples, we translated qualitative data into a quantitative format so we could chart or graph it, compare it to our own results, or compare it to the results of competitors. It helps make objective data a bit more subjective.  Granted, these examples are overly simplified, but the basic approach can be applied to more complex business questions. These approaches should provide a framework to get started.  Finally, this approach is not designed to discount qualitative data in its pure form, it has real value. Rather to allow the qualitative data to be visualized in a format (such as a Dashboard report) that might be more understandable to individuals without the opportunity to sift through the raw data. In many cases, qualitative data can provide soft insights beyond the capture of hard numbers. George Sloane 9
  10. 10. Thank You! This Slide Concludes This Review  For More Information, Please Contact George Sloane at:  AnalyticDesignGuy.com website  LinkedIn  Twitter  Facebook  Google+  Slideshare  Goodreads • Gmail: gsloane.business@gmail.com • Phone: (203) 981-4488 George Sloane 10