Opinion Research, Done Right, Improves the Success Rate

Right now, how confident are you in your ability to accurately answer the following questions: What do your customers like best about the service you provide? What specifically could you do better to keep them happy?

How likely would they be to recommend you over one of your competitors? How likely are they to stay with you in the future? What do they think about a new service idea or product you are considering?

Information may be power, but it is good information that leads to good decisions. Some of that information may be readily available, but some of it lies hidden in the form of people's opinions. Opinion research is often the only way to uncover this kind of information. It is the tool that allows you to move beyond hunches and guesswork into reliable feedback directly from your customers.

Opinion research can be a competitive advantage

Research is an "early warning system" that can help you identify opportunities for a competitive advantage and avoid potential pitfalls. It is the resource that helps you sift through a dozen good ideas and determine the very best one to pursue to achieve your goals and objectives.

Engaging customers through research lets you test the impact of decisions before you actually make them. Before you spend a lot of money to introduce a new product, service, change the name of the company, or develop a new marketing strategy, you can see what the reaction from your key customers will be first. This prudent investment may save costly mistakes and will almost always give you ideas for improvements.

Lastly, research is the critical tool that lets you listen to the most important people in your business - your customers. The success of your business depends on how well you listen and respond and how you react in the marketplace. They might include the general public, current or former customers, potential new customers, employees, other businesses, even government leaders. Because these groups are vital to your success, researching their opinions should be very important to you.

Principles of good survey questions

  1. Questions should ask about first-hand experiences or perceptions.
  2. Questions should focus on one subject at a time.
  3. Questions should use words that ensure all respondents will answer the same questions.

Step 1: Identify objective of research

Failure to define clear objectives for your research is often the key reason why a research project does not live up to its strategic potential.

The first question is: What is the business objective behind the research?

Research should always be designed to meet a clearly defined and agreed upon business objective. If you can't answer the questions "why are we doing this?" and "how will the results help us improve our business?" you are not ready to launch a research project.

Setting a clear business objective ensures that each step of the process ties back to fulfilling these objectives.

Step 2: Define your audience
Who is the target audience to be surveyed?

Customer research is only as good as the people giving you their input. Defining the appropriate research participants is a critical step in making sure that the results represent the opinions and behaviors of the people you want to impact. Are they your current customers or potential customers? Are they top decision makers in the organization or front-line personnel who interact directly with your equipment or service people? Are some companies more important to study than others (say, for example, those that accounted for 80 percent of your revenue in the last few years)?

Step 3: What do you want them to do?

The questions that you ask need to focus on the areas where you have questions about your company's performance. Sometimes this is attitudinal - what do customers think about your business. Sometimes it is behavioral - getting customers to take a desired action like purchase more of your products, sign longer term contracts or tell others about your company. Defining what action you want customers to take is critical to designing the types of questions that they should be asked.

Customers wear different hats in the course of the day. At any given moment they could be making choices as a parent, an individual consumer, a purchasing agent or a corporate representative. Effective research needs to clearly define in what context a respondent should be participating in the research to make sure that the feedback is in line with the business objectives you have outlined.

Once you have made the commitment to conduct a survey and have a clearly defined set of answers to your questions, it's time to settle on an approach. Customer research generally falls into two methodological approaches, quantitative and qualitative.

Quantitative research

Quantitative research often takes the form of mail, telephone and, increasingly, Internet surveys. In this approach, research is about the "how many" and the "who" in research. That is, how many people (what percentage) feel a particular way?

Quantitative research generally involves large numbers of respondents with the goal of generating reliable and projectable results to a given population. For example, polls that you see reported on the evening news or in publications are almost always quantitative surveys.

Qualitative research

Qualitative research usually involves far fewer participants in a more intimate setting like focus groups and in-person interviews. Qualitative research is designed to understand the "why" of an issue, not the percent. Qualitative research seeks a better understanding of underlying emotions and thought processes of your customers.

Picking the best research method comes from having a clear understanding of your intentions. Often, quantitative and qualitative research approaches can be used together to give the most holistic understanding behind customer perceptions and decision making.

For example, a quantitative survey can be a great tool for measuring what percentage of customers rate your service a 7 on a 10-point scale. However, qualitative research is more effective at helping you understand what factors prevented them from rating you a 9 or what specifically you are doing that kept them from rating you a 5.

Will it be anonymous?

Before beginning any research effort you will need to make a decision about whether or not you are willing to offer your respondents confidentiality or anonymity.

Anonymity means that there is no way for you to ever link back a specific response to a given respondent. Sending a mail survey to a customer that they can mail back (so long as there is no unique code or identification number printed on the survey) would be anonymous.

Confidentiality means that though it is possible to link a respondent to a given response, you guarantee that you will not make any efforts to do so. If you were to contact a customer through the Internet to administer a survey, they are no longer anonymous, but you can offer that their responses will only be reported in aggregate with all other respondents keeping their individual responses confidential.

Anonymity also has its limits

Generally offering anonymity or confidentiality increases the likelihood that an individual will be willing to participate in a research effort. However, there may be times that you want to be able to follow up with a given respondent based on their answers to specific questions.

A case that often arises is when a customer gives exceptionally low performance ratings of a company or has had a particularly bad experience that is likely to cause them to seek out a new operator.

The temptation is to often try and follow up with that individual to get things back on track. However, if you have already made an offer of confidentiality, you'd be violating ethical (and perhaps legal) requirements if you did.

So, it is a trade-off. Confidentiality can generate higher participation rates in your research, but it comes at the cost of not being able to take targeted action to satisfy an unhappy customer.

Ask the right questions

Whether you are planning on a quantitative survey like an Internet or mail approach or a qualitative investigation like a focus group or a one-on-one interview, the general structure of the research itself should follow the same outline. Start with an appeal to your customers about why you are conducting the research and why their input is valuable. Something along the lines of "we are interviewing a select group of our customers and your open and honest feedback is critical to helping us continue to serve you better" lets a potential respondent know their feedback can have real benefits for them.

The first research questions should be general, nonthreatening, easy to answer and spark an interest in the rest of the survey. Starting out with "how much money do you plan on spending in the next year with our company" or "please list all the products you have purchased from us in the past five years" is a certain way to get a survey tossed in the waste basket or a polite "I'm not interested."

Questions should create interest

Instead, an initial question like, "What is your level of satisfaction with your vending service?" can be a way to immediately engage a respondent and generate interest in what questions may be coming next.

Once you've built some interest with the respondent, questioning can begin to move into the more meaty issues that are your primary interest. Three principles of good questions can help guide you to ensure reliable and valid responses.

1) Questions should ask about first-hand experiences or perceptions.

Asking customers about the perceptions of their supervisors or colleagues regarding your business opens responses to significant error. Questions should focus on the individual respondent's beliefs, attitudes and behaviors. Avoid hypothetical questions. Asking "if we provided healthy snack choices would this improve your overall health?" is not a question your respondents can reliably answer.

Questions should focus on one issue

2) Ask questions one at a time.

Compound or "double-barreled" questions muddle a respondent's ability to answer a question and your ability to interpret the results. "Overall, how would you rate our customer service and product offerings compared to other companies you have engaged in the past?" is an example of a double-barreled question that may leave a respondent with an inadequate means to answer the question, and you with difficulty interpreting the response.

3) Use words that ensure the respondents will be answering the same question.

All respondents must understand the question in the same way. Perhaps words or terms must be used that have meanings that amay not be shared, definitions should be provided to all respondents. "Within the past two months, how many contacts have you had with one of our sales representatives or service personnel?" provides respondents with a context for answering the question.

Know your audience demographics

After asking your main questions, a survey should always end with a series of demographic questions (job title, company size, length of time as a customer, etc.). Demographics can help to ensure your survey is representative of the population you are interested in. For example, if 75 percent of your customers are small businesses, 75 percent of your respondents should be as well. Second, demographic questions can help provide needed insights for your analysis.

For example, 20 percent of your total respondents may report being receptive to a new product idea. But when broken out demographically, you may find that the majority of those are your long-standing customers, indicating that newer customers might not be as familiar with your history of service and product offerings.

Include thank you in closing

The survey, regardless of the method, should always close with thanking the respondent for their time and insights.

Regardless of the general research approach you take or the specific survey questions you ask, customer research needs to be viewed as an exchange of services. Customers who take the time to participate in your research effort are giving you something of great value - insights that will make your business more effective in the marketplace. As a result, you need to consider what you are willing to give them in return.

Recognize their participation

Sometimes this exchange may simply be a clear and compelling statement early on that the research is going to be used to guide future decisions in your business. Sometimes that exchange may require a token of appreciation, like a financial incentive or discount on a future order. Generally the longer the time needed to participate in the research and the more important the respondent, the more substantial an incentive is required in order to be seen as a fair exchange in the eyes of your participants.

Opinion research, done right, can be a powerful planning tool for any size company.