"If you haven't asked the question, how will you know the answer?" That old saying is a pithy summation of the rationale behind conducting a survey to find out information about your customers.

Survey research, however, is not as simple as asking questions: The right questions must be asked of the right people. You must first determine what information you want to collect, which then guides you to choosing which questions to ask, how to ask them, and of whom.

This article will discuss when conducting a survey is appropriate; it will offer some tips on survey design; and it will explore how to incorporate survey results into PR and marketing activities.

When a Survey Is Useful

Surveys are generally conducted for at least one of three reasons: to describe a population, to explain behavior or attitudes, or to explore or discover new, relevant topics.

Goal

Accordingly, the first step is to establish the goal of the survey: what you want to know—such as finding more about users, their perception of a product, or their requirements and the environment in which that product will operate.

The topic can be further narrowed to possible questions regarding who are potential users, how current customers feel about a certain product, or what the most pertinent issues are when deciding to purchase an item in this product category.

Sample and Process

Once you have decided on the goal, you can then decide on who are the best people to survey. You need to be aware of biases, such as those of people attending a tradeshow, stopping by your booth, and agreeing to take part in the survey. Perhaps a telephone survey conducted by an outside firm would find more accurate responses because it would be an independent study and it would be perceived by participants as more objective.

Obviously, time and budget issues must be taken into account, as well, and an outside-firm route is more expensive than the tradeshow option.

Some surveys may require collection of data over time (e.g., "How have attitudes toward the Cloud changed over the past year?"). In such a case, the respondents would need to complete the same or a similar survey multiple times if you're going to see whether attitudes or knowledge has changed and the reasons for the change.

There is also the question of how many people to survey and the number selected may impact the validity of results. While the US Census may count everyone in the country, it is too costly, time-consuming, and generally unnecessary to ask everyone in the market about your product. The quality of data gleaned from a more focused sample is more accurate because you are just asking, for example, those who are interested in, knowledgeable about, or targets for your product.

The goal is to find people to survey who represent a cross-section of the total population. If everyone were exactly the same, we'd only need to survey one person.

Without going into scientific and mathematical discussions on determining sample size, the key is to determine a high-enough confidence level and accuracy of representation.

Survey Design

Survey design discussions have filled several textbooks. The most important takeaway is that the questions should be clear and unambiguous to the respondent if they are to deliver responses that will be useful to those conducting the research.

For example, avoid questions such as Agree/Disagree with "Purchasing this product will enable my company to expand into new areas, garner additional revenues, and be more respected by employees." Respondents may not understand whether "new areas" means geographic areas or product lines; moreover, you will not know whether the agreement/disagreement has to do with new areas, revenues, or respect.

During an in-person or phone survey, the person administering the survey must neither offer any guidance on the questions nor offer any opinion on the responses given.

How to Incorporate Survey Results

Once the data has been gathered, what will be done with it?

As the saying goes, "no need to reinvent the wheel." In this case, that means once the data has been converted into conclusions and information, it can be used for PR and marketing objective, and disseminated and used in various formats and for various purposes, such as press releases, sales presentations, whitepapers, formal reports, webinars, infographics, etc.

The information can be divided by topic, as well, with one whitepaper covering motivation for migrating to new products or services, for example, and another discussing cost issues related to the migration.

Any text should include visuals such as charts and graphs that clearly display the most significant results.

On the marketing side, the conclusions ascertained can be used for such business decisions as product introductions, pricing, messaging, and branding. The information cannot and should not totally direct decisions, but it can be combined with overall company direction and market needs.

* * *

If designed properly and if clear in purpose, survey research is a valuable tool for gathering information about customers, attitudes, and market drivers and challenges.

When questions are designed properly and the respondents are carefully chosen to avoid biases or misrepresentation of the overall population, valuable results will follow. Goals and questions must be clear to avoid being led down the wrong path.

Whether you receive expected responses and trends—or not—conducting a survey can be informative and useful for your business, marketing, and PR strategies.