Liz Janik
by Liz Janik
Are you comfortable with audience research? If you answered “no,” then you likely haven’t had any training on how to buy and use research effectively, yet you need information to successfully position and market your station.
Research, in general, has a ‘bad’ reputation. There are several factors contributing to this cynicism. Some of it comes from a lack of knowledge about what research can, and cannot, do. Then, there are all those variables that affect the quality of a research study: There can be problems with the design of the research. The wording of the questions may include multiple factors. The recruiting of respondents might be compromised in an effort to bring down costs.
Once you have the data, how do you ensure the analysis and comprehension is complete and accurate? There can be further concerns in applying the findings. Even though it takes time to absorb and evaluate research information, I have witnessed major format decisions made within a few minutes, based solely on the researcher’s “top-line” findings. Or the research is ignored; many a research study has been filed away after the formal presentation, without further review or analysis – or comprehension. People simply don’t know how to read the research or apply the findings.
In addition, there are inherent high costs in interviewing the large, random sample of respondents needed to produce good quantitative research. This can lead to risky price-cutting steps from ‘eager’ research suppliers. These are just a few of the reasons why research may not deliver the results that are possible.
Valid, statistically reliable research will give you the information edge to precisely predict the potential size of the audience for a potential format. It will provide direction on exactly what music and information to program to attract listeners. Research can also help determine the most effective station image, marketing message and which media to use.
Here is a quick over view of some research basics, starting with picking the right research for the job.
Choosing the Right Tool
First, you need to have some understanding about simple research methodology. Using the wrong research tool can result in lost time and money. In a worst-case scenario, “bad” research can destroy a station’s position, resulting in a potential loss of millions of dollars.
In simple terms, there are two basic kinds of research tools to choose from: 1) Qualitative Studies and 2) Quantitative Studies.
Both are useful tools, but ONLY Quantitative studies can be safely applied to the market or audience as a whole. Quantitative studies use a large enough sample of respondents to be able to get statistically valid results. These findings can be safely applied to the market. Here’s a quick review of the difference between the methodologies of qualitative and quantitative research:
The Qualitative Method
Qualitative research is a great way to get insights into how people think and feel about a product. The most commonly used qualitative study is the focus group.
A focus group is the ideal tool to gain a fresh perspective on a station’s programming and marketing. Listeners are objective, and have insights about issues that might never have to come to mind for radio people. But as interesting as those insights might be, it is very dangerous to use qualitative research to make changes to your radio station.
The key point is reliability. How safe is this information to apply to the market? The tiny sample of a few dozen respondents, cannot produce statistically reliable findings, and is therefore unsafe to apply. The choice to use focus groups is often based on cost. A quantitative study usually requires a sample size of several hundred participants, so it is always perceived as a more expensive study
Without understanding the critical lack of statistical reliability of focus groups, it’s easier to pick the lower priced research. Perhaps with the misguided thinking that some research is better than no research. Unfortunately, not all research suppliers will advise you of the difference in reliability between the two research methods.
The insights into the consumers’ thoughts and feelings that you’ll get from qualitative research are very valuable when used to design the questionnaire for the more statistically accurate, quantitative study.
The Quantitative Method
This research tool reliably measures what your market (or audience) thinks about an issue. But you need a large sample of respondents in order to safely apply the findings and estimate numbers about the market. Quantitative studies range in price from $25,000 to $50,000 depending on market conditions.
‘Reliability’ indicates that this information is ‘safe’ to apply to the market as a whole. A quantitative study should deliver the same results 19 out of 20 times. Another indicator of reliability is the ‘margin of error’ which should be low number. Reliability is communally expressed as a ‘margin of error.’
Margin of Error
There is no point looking at any research findings, unless the margin of error is included in the report. ‘Margin of error’, is expressed as “plus or minus ‘x’ percentage points”. A study would be considered reliable if It had a margin of error within a range of ‘plus or minus 3%’. The lower the ‘margin of error’, the more reliable (and safer) the information is.
To illustrate: the research findings for a new format option might determine that 12% of the respondents strongly agree that they want to hear this new format. If the margin of error was determined to be “plus or minus 3%”, then the findings would indicate that the new format would attract between 9% to 15% of the market.
Sample Size
Statistically reliable information is based on an appropriately sized sample of randomly selected respondents. In very general terms, a sample of 600 respondents will give valid, reliable results in most markets. The mathematical formula is basically static for markets between 100,000 and several million people. This seems confusing to non-math heads, as the expectation is that the larger the market, the more respondents would be needed. Suffice to say, the way those mathematical algorithms work, small and medium markets need as large as sample as a major market.
‘Random’ Sample
Respondents for a research study can be solicited in a variety of ways. The highest level of accuracy for a total market study is achieved by using a ‘randomly’ selected group of respondents, whose age and sex profile matches the demographics of the market. Random sampling is more expensive to achieve, but the reliability and over-all success for a total market study depends on it.
There are limited occasions when a station might use a list of listeners as a resource for respondents. This can be done under the right conditions. How the list was prepared/sourced and what type of information is needed would be considered.
The Quantitative Research Danger Zone
It’s important to know that every time you divide a group of people into even smaller groups, you diminish the accuracy of the numbers as applied to the total market. The sample size of the smaller group can become too small to reliably project – just like the qualitative focus groups. When you are looking at a sub-group of 50 respondents, the margin of error has become significantly larger. Caution should be used when looking at sub-group of less than 50 respondents.
In fact when you subdivide the sample so there are fewer than 30 people in a certain cell, the margin of error will be extremely high. The data has about a 50/50 chance of being accurate. You would get the same degree of reliability if you flipped a coin!
The Power of Information
Research can open the door to fresh new thinking about formats and listeners, but it is only as good as all the people who work with it. Good research has led to many break throughs in radio programming, marketing and sales. The power of reliable information gives managers the courage and confidence to be original, and ultimately successful.
Liz Janik is a ‘format architect.’ She helps managers get the information edge. 905-454-3965 or liz@lizjanik.com.
