HOW TO WIN BIG DOING YOUR OWN
MARKET RESEARCH

by Dr. Kevin Nunley

index of articles

Thinking of starting a new business? Adding to the business you already have? Introducing a new product or service? Wouldn't it be nice to have a good idea of how successful you'll be before you even start? That's the money-making edge that smart market research can give you.

You've probably heard that market research is expensive, only something that big companies can afford. That's partly true. Even a relatively modest research program can eat up several hundred thousand dollars in a hurry. But market research doesn't have to so complicated only expensive consultants can figure it out. Here are some very simple ideas and tools for getting a pretty good idea of where you stand--BEFORE you shell out big bucks for marketing and advertising.

FINDING THE RIGHT PEOPLE TO RESEARCH

Most products and services have next to no chance of success without a good, strong marketing program to promote them. No small business has the budget to do a saturation ad campaign that attempts to reach everyone. There's no need to. Your business, no matter what you're selling, likely only needs to get the attention of a select group of good prospects.

Market research techniques help you get the factual information you need about your target audience and the effectiveness of your message. If you are wondering which new product to offer, market research can poll people who have bought from you in the last six months, people who bought once but never bought again, and people who usually buy from your competitors. This will give you a very good idea of how well your new product will fair once it's introduced.

GETTING A GOOD SAMPLE

While research can be very complicated, there are a number of simple techniques that will deliver fairly reliable results. Quantitative research methods provide statistical information. A carefully chosen scientific sample is studied as a representation of the larger public. In other words, 50 people are chosen. If we've chosen them with an eye to good sampling methods, those 50 will closely represent everyone else who is included in the target audience.

This can be a bit trickier than it seems. You've surely seen the market research person with a clipboard standing in the mall asking people if they'd like to take a survey. Would this be a good sample of the entire community? No. Not everyone goes to the mall. A great many people, including people who don't own a car, people who live a long way from the mall, and older folks ho don't leave home often rarely or never go to the mall. Mall shoppers may be inordinately young, or more affluent than the rest of the population. It wouldn't be accurate to assume that mall shoppers represent the entire community. However, the market research person in the mall might get a very good picture of what the mall-shopping community is like.

The most common kind of quantitative research (the kind that provides you with numbers and percentages) is the telephone survey. It's fairly accurate to go through the telephone directory, calling every eleventh person listed. The problem here is that not everybody has a telephone and a great many people have unpublished numbers. This will reduce the accuracy of your findings, although you will still be way ahead of the mall survey.

The best solution is to use a computer program which gives you at random all possible telephone numbers in your area. These programs can be purchased, and most university communication departments have them for student use. Perhaps a student can print you a randomized list of telephone numbers.

You can also mail surveys to homes and businesses, or visit them in person, through this method. A sample of locations, be it homes or offices, can be gotten by first choosing areas at random, then blocks at random, then homes on those blocks at random. You could roll dice to determine which locations are picked. All this keeps personal opinions out of the research.

You can do surveys with randomly chosen email addresses provided those in the study have given their permission to be contacted.

Most research is based on simple statistics. No higher math is required. You can do just about everything with a simple calculator and advice from your junior high-aged child. If you want further information about scientific sampling and the statistics you can perform on your sample results, please consult one of the great many books on research. It's a subject that has remained largely unchanged for the past 50 years, so an old tattered volume in the used book store or at the public library will do just fine.

STEPS TO A GOOD QUESTIONNAIRE

You want your questions to be carefully written so that they do not confuse or suggest "correct" answers to the respondent. Here are some general guidelines:

1. Make sure your questions are clear and easily understood.

2. Keep questions short. People in a hurry won't take time to understand a long and unclear question.

3. Questions must be in sync with the purpose of the research. If the question is irrelevant to what the survey is trying to study, leave it out.

4. Don't ask questions that can be broken down into two or more questions. For example, "Do you think the mayor is dishonest and a poor financial planner?"

That is really two different questions. Be wary when the word "and" appears in a question.

5. Stay away from biased words. For example, "Do you eat a healthy breakfast or just have a Big Gulp at 7-11?" The word "just" prejudices the answer by suggesting that the Big Gulp is less worthy than the healthy breakfast.

6. Avoid leading questions. "Like most New Yorkers, do you drink coffee every morning?" Watch for a hidden premise showing up in questions. Remember, the goal is to accurately determine what the respondent thinks, even if it isn't what you wish they would say. The purpose of research is to find out which of your ideas are wrong.

7. Leave out questions that require very detailed answers.

8. Avoid questions that may embarrass the respondent. Many people don't like to give their age, and most won't tell you how much money they earn. A better way is to give the respondent a broad category that they can identify with without giving away sensitive information. "Are you between 18-24 years old, 25-49 years old?, etc." Additionally, research carries with it a certain authority that will make your ideas more persuasive to others.

Kevin Nunley provides marketing advice and copy writing for businesses and organizations. Read all his money-saving marketing tips at http://DrNunley.com/. Reach him from his site via email.