TURF Analysis helps analyze estimates of media or market potential and devising optimal communication and placement strategies using multiple Choice/Multiple Answer question types.
'TURF' is a statistical model that can be used to answer questions like, Where should we place ads to reach the widest possible audience? AND - What kind of market-share will we gain if we add a new line to our model? It was originally devised for analysis of media campaigns and has been expanded to apply to product, line and distribution analysis.
With QuestionPro any Multiple Choice/Multiple Answer question can be analyzed using TURF. The TURF Simulator calculates Optimal configurations for maximizing reach. Reach or Coverage is defined as the proportion of the audience (target group) that chooses a particular option.
TURF stands for Total Unduplicated Reach and Frequency Analysis. Reach means how many people you contact, and frequency refers to how often you are in contact with the people you reach. It is a statistical model that can be used to answer questions like:
QuestionPro conducts surveys on a regular basis. One of these surveys had to do with people’s experience using airlines. One of the questions had to do with the level of satisfaction with various aspects of a flight experience. According to the basic analysis, the top three items people were least satisfied with were: oversize baggage fees, checked baggage fees and handling care (damage, lost luggage).
At first glance, it looks like focusing on those three areas would be the best use of our time. A TURF analysis can help us determine if this is truly the case.
To start the TURF analysis,
Click the question for which you want to run a TURF analysis. You can now select from give analysis options: Top/Down Reach, Analysis, Simulator, Price Modeling, and Line Optimization.
This shows you how many potential clients you would reach if you focused on any one particular item. To use this, simply click Calculate Top/Down Reach. In this example, focusing on improving customer experience with checked baggage fees, you would reach 47% of the total population. Please note that this is based on the number of people surveyed. You will need to be sure that your sample size is large enough to be able to make inferences about the total population you are studying.
Analysis lets you mix and match up options to determine what kind of reach you would have if you focused on those particular combinations. To use this, check the boxes next to the options you want to include in the calculation, then click Calculate Unduplicated Reach.
The simulator lets you choose how many items you want to focus on, then simulates the reach, cost, and count (based on the number of responses to your survey) for every combination of choices possible. In this example, respondents had seven choices. We ran an analysis on every combination of up to five choices to determine the best possible mix of items to focus on for the maximum reach. We only show the top three combinations here (a total of 21 were calculated).
To use this, select the number of choices you want to include in the simulator using the drop-down. Then click Simulate Choices.
Price modeling is just as it sounds. You enter the budget you are willing to spend overall and the tolerance you have (wiggle room) within that budget. For our example, we said we had a budget of 100 USD and a tolerance of 10%. You can also assign weights rather than prices. Click Simulate Price Modeling to calculate the every possible option group based on the budget desired, starting with the lowest budget threshold (budget – % tolerance) and working up to the highest budget amount. Note: If you go back to the simulator, you will find that the option sets here are now calculated with pricing.
Line Optimization calculates the group of items that will meet your minimum reach desired at the lowest cost. To use this, simply input your minimum reach desired, adjust pricing if desired (if prices were assigned in Price Modeling, they will carry over to this tab), and click Simulate Line Optimization.