Message Testing

Message Testing is the technique of trying out alternative messages on an audience sample, before you use the messages publicly.

How You Do It

First, you write several alternative messages. Second, you identify a sample audience group, generally called a focus group. Third, you assemble the group in person, show them your messages, ask for their reactions, and analyze the results. Alternatively, you can assemble a virtual or online focus group.

There are many available services and tools, which you can find via the web. In addition, there are detailed instructions for managing virtual focus groups (for example, here) and in-person focus groups (for example, here).

It's worth spending some time surfing, because you'll be more likely to find services, tools, or information appropriate to your specific needs and situation.


This is a powerful navigational tool, because one message can be many times more effective than another. In practice, it is common for one message to pull two, three or four times more responses than another message. In other words, this tool can give your program phenomenal leverage. It can multiply your profitability.

Remember, however, that this tool can't help you measure your profitability, as evaluative tools can. The ultimate measurement system that you assemble will have to include a combination of navigational and evaluative tools.


As a marketing metric, message testing has three major weaknesses:

  1. It takes a chunk of time to do it. And unfortunately, this chunk of time must be taken from the period when management is getting eager to roll out the program. You often have to fight for the time to test. If your management is not marketing-literate, you may lose that fight. (However, if you surf the web for tools, you will see that some time-saving tools are available.)
  2. It can be expensive. (But, again, some tools can help you reduce or contain the costs.)
  3. You have to be careful how you select and invite the participants, how you frame the questions you ask them, and how you analyze their responses. Doing it wrong can skew the results, often to a great extent.

Generally speaking, it's a good practice to enlist the help of a research expert, because there are pitfalls here that may not be obvious to the non-expert; for example, Nonresponse Bias and Dishonest Answers.  Either of these pitfalls can significantly distort you test results.

So, if you want to do message testing, you probably should hire an expert. 

Return from Message Testing to Measuring What They Thought