Audience Dialogue

Participative Marketing for Local Radio, Chapter 2
Audience concepts: section 1

2.1 Sources of data

Two kinds of data are useful for radio stations' data about the population, and data about audiences.

Data about the population

Almost every country has a government department in charge of statistics, often called a Department of Statistics, or Census Bureau. Every 5 to 10 years, that department will do a census of the entire population, and publish volumes with detailed tables of results.

In some countries, other government departments also publish statistical data. For example, education departments often publish statistics about education levels of the population. This is often done annually, so tends to be less out of date than census data.

Some organizations, such as NGOs, often collect this statistical data from various sources, and republish it in a handy format.

It's almost certain that some data about the population will be available for your coverage area.

Data about audiences

Radio (along with TV) has a unique need for audience research. All other types of organization automatically find out if they have customers, because they get income from sales. But for radio and TV (without audience research) you have no way of knowing how large your audience might be.

Other types of industry measure their audiences (or customers) from sales data, but for broadcasters, audience sizes can come only from surveys. It takes a lot of skill to do a survey well, and produce unbiased information. Even then, a survey can cover only a small proportion of the population, so there is always some sampling error involved. The smaller the sample, the larger the sampling error. Despite these limitations, a skilfully-done survey provides unbiased information. It is more trustworthy than spontaneous audience feedback.

For example, using the number of phone calls you receive as a guide to program popularity could be very misleading, because the availability of phones to different listener groups at different times of week may be the main reason why listeners call in. Another reason might be that some programs often broadcast your phone number, while others never do.

You can obtain audience data in three ways:

1. Buy a syndicated survey (done by a market research company for all stations in a coverage area).

2. Get a research company to do a survey for your station.

3. Do your own survey.

If a syndicated survey exists in your area, that's usually the best option to choose. It's relatively cheap, because a number of stations share the cost. Potential advertisers are likely to believe the results, because the survey wasn't commissioned by any one station.

Option 2, a survey done specially for your station, can give you much more detailed information than a syndicated survey, and can be very helpful for improving your programming. However it's not so useful for selling advertising (unless done by a very reputable research organization) because advertisers may believe you have influenced the results. This is by far the most expensive option of the three. However, if a syndicated survey exists, but excludes your station, or its methods are biased against your station, it can be worthwhile to commission your own survey - if you can afford it.

Option 3 is tempting: to do your own survey. It can be cheap, because most of the cost of a survey goes on interviewers' time, and if you already have the staff your additional cost may be quite low. If you've never done a survey before, I should warn you that it's much more difficult than you might expect. You need a highly skilled organizer, and also a high level of computer skills and perhaps special software. Another problem with doing your own survey is its likely low credibility with advertisers (no matter how thoroughly the survey is done).

Do-it-yourself audience research is not so good for accurately estimating the size of the audience, but it's excellent for getting detailed information to help you improve the programs. If you're interested, see my book Know Your Audience, which gives detailed instructions for most types of audience research - including several methods that are simpler and more useful than standard surveys.


If you like this article why not subscribe to our free Participative Marketing for Local Radio mailing list to receive periodic articles and updates on relevant topics by email, refer the signup form below. Simply enter your email address then click subscribe.  Your email address will only be used by Audience Dialogue to communicate related topics, will not be provided to any other party and you can unsubscribe easily at any time.  A sample newsletter is available to view here.