How to Design and Conduct Original Market Research: Goals, Tools & Validity


Create engaging content for free

You know what they say: If you want something done right, you have to do it yourself. Chances are, if you want the highest quality original data about your target market, you’ll have to collect it.

Visage_MarketResearchFINAL-01

To find helpful, useful data, you need to know a little about designing and conducting market research. Don’t feel intimidated – you’ve been doing market research all along. Ever listen to a customer complaint? That’s market research in its most basic form.
Market research has many purposes and many ways to collect data; the data collection method you choose depends on what information you want

Identify opportunities within different populations.

You might try conducting focus groups and interviewing customers to find out what they like, don’t like, need and wish they had. Then follow-up with a pilot program that offers these features and test how it does.

Define your value proposition.

This is a challenge to many marketers, but you don’t have to do it alone. Organize a focus group of people within your target market and tell them your ideas, what makes your product unique, and how you’d like your product or company to be perceived. Listen and create an action plan considering their feedback.

Find out if your marketing strategies are effective.

Analytics has made it easier than ever to track marketing campaign efficacy. The red button got 108 clicks; the blue button only had 20. But when you need to know more about why one appealed to users more than the other (advertisements, landing pages, logos, color schemes), you might want to just ask. Interview, send questionnaires over email, or request users to fill out an exit survey.

Have you noticed a trend? All of these research suggestions begin with a specific goal. Define what you want to figure out with your market research first, but be prepared for your research to reveal unexpected answers.

Helpful Survey Tools

Surveys, interviews, questionnaires, and focus groups are all examples of primary research: data gathered straight from the horse’s mouth, so to speak. Often, the answers are only as good as your questions, which makes these methods anything but foolproof.

Visage_MarketResearchFINAL-03

For interviews and focus groups, open-ended questions may prove the most useful, like “What are the most important features you look for when deciding whether to purchase this product/service?”

Surveys and questionnaires generally benefit from “Do you agree with this statement on a scale of 1-10?”

One of the easiest, and cheapest (it’s free) survey tools is SurveyMonkey. Hubspot uses SurveyMonkey to create content strategy surveys that ask questions like “How often would you want to receive information from our company?” and “When you share information about companies and the products or services they offer, which of the following do you use? [select all that apply].”

AYTM (Ask Your Target Market) lets users create surveys and send them to both their own lists, and the AYTM panel of 4.5 million people (after you choose targeting criteria to narrow down the survey takers). SurveyMonkey also has a consumer panel with hundreds of thousands of survey respondents, but that costs extra. Instantly offers similar services with variations on how many people are in their panels, and how specific you can be with your targeting criteria.

GutCheck lets you conduct 30-minute video interviews with its 3.5 million U.S. members. Choose age, income, and other criteria to make sure you’re talking to the right people.

Check Your Research Methods

The validity of your research boils down to this: Does your research effectively measure what it claims to measure? If you have doubts, take some time to look at the results.

The second question you should ask is: Are your results reliable? As in, can they be applied to a wider group than just those who took part in the survey? You want your data gathering methods to give you consistent, repeatable results. This is where having an appropriate sample size comes in. Creative Research Systems offers a sample size calculator to help determine how many people need to take part in a survey or interview to get results that accurately reflect your target population.

Market research can be complicated and technical, but it doesn’t have to be. At the end of the day, whether you ask your customers a question over the phone, or ask 100,000 target prospects that question in a survey, you’ll have gained valuable knowledge that can help your marketing efforts improve and your company thrive.

Now let’s get into the nitty-gritty details of how to design a market research survey to yield the most helpful results.

Any research will yield results – but without an expertly crafted survey, those results can be:

  • Useless
  • Misleading
  • Downright wrong

Unless you know a few simple strategies for how to design your surveys.

Ask Yourself the Following Questions:

Where will this survey live?

Will you publish your survey on your website, or email it out to your contacts? Understand that these two locations are populated with different types of people. Your email list is made up of people who have already opted in. They’ll have a little, or a lot, of experience with your product or service, making their feedback very valuable, and they’re more likely to be invested in helping your company improve.

Your website survey might pop up for new visitors who aren’t even sure if they’re in the right place much less whether they want to take time out of their busy day to help you out. There are ways to automate pop-up surveys so they only appear to new visitors, helping you glean valuable information on their first impressions.

As you can see, each population has something different and valuable to offer. But don’t confuse the two.

One more variable: Understand that the people who choose to take your survey are a self-selecting group. This means that your results are only true for the type of customer who is willing to take your survey in the first place.

How Will it be Constructed?

Some estimate that half of survey respondents don’t complete the questionnaire because they become annoyed at how it’s constructed. It can be extremely irritating if a company asks x, y and z, but fail to ask the question that allows the respondent to give the feedback they’d really like to give. The fixes for both of these problems are simple:

Be brief. Be flexible.

 

How to Write the Survey

Now it comes down to writing the survey questions and organizing their order.

  1. Ask the easiest questions first.
  2. Group questions on the same topic together (to avoid confusion).
  3. If you’re asking a more involved question, place it towards the middle.
  4. Ask the personal questions at the end.
  5. Always include an “any other thoughts?” form in case you missed an issue that should be addressed.
  6. Be mindful of wording.

Avoid Biased Results

When crafting the survey, phrase each question to avoid biased results. For example:

  • “What do you like about…?” pushes for a positive answer, when the truthful answer might be an emphatic “nothing! Ugh!”
  • “How much do you think … will decrease this year?” assumes something will decrease, when it might increase.

Another way to avoid this issue is to ask respondents to choose a number between 1-10 for how much they agree or disagree. Bonus: You can then turn this proprietary data into into charts or infographics for future presentations.

Avoid the No. 1 Survey Mistake

Don’t overcomplicate questions with too many variables. If you make this mistake, your results will be skewed and lose their value.

Visage_MarketResearchFINAL-02

For example: “How important is friendly, fast customer service?” combines two important factors that can skew results. For more meaningful and actionable responses, include two questions:

  • “How important is friendly customer service?”
  • “How many minutes would you be willing to wait on hold until you speak with a customer service representative?”

At the end of the day, honest results are helpful results. Allow those taking your survey to answer the questions and give their honest response.

Comments