Here’s a quick overview of how to do your own market research surveys to help you design and market your product.

Online product surveys can be an extremely cost-effective way of getting early-stage feedback. Questionnaires can be highly structured for statistical analysis (quantitative results), or open text responses (qualitative). Given the speed and affordability, there’s very little reason not to do surveys when developing a new consumer product.

Should you do your own product surveys?

If you have the time and inclination, then at least attempting your own survey is a great way to learn more about your market and the product R&D process.

The main arguments against going the DIY route include:

1. Misleading results. Inadvertent biases in your questionnaire design or survey sampling (respondents) might actually point your product development in the wrong direction.
2. Low trust. Giving potential investors and partners confidence in your product idea is a major benefit of surveys conducted by trusted professionals. Self-run surveys may be viewed with some suspicion.
3. Takes time. Conducting proper surveys is normally time-consuming, and more so when you’re still learning how. As an entrepreneur, your time may be put to better use elsewhere.

But since you’re an innovation entrepreneur, it’s nearly certain that you’ll do it yourself anyway! So here’s how you can get the best results on your own, doing marketing research surveys “from scratch”.

Survey result presentation

Even if you do an incredible job with the survey, others may not trust the results if the presentation is lacking. So if you intend to share your results, we recommend putting in the extra effort to make your final report look professional.

Product survey software

Assuming you’re doing an online survey (usually recommended for speed and cost reasons), you’re going to need some survey software. SurveyMonkey (rebranded as Momentive in 2021) is the best-known, and a good choice because you might be okay with a free account if your questionnaire is simple (few questions, no conditions or branching).

Product questionnaire design

We could write an entire book on this (and indeed, many books have been!), but suffice it to say: it’s challenging. A few mistakes we often see:

• Unclear questions, which may be interpreted in different ways, especially if your sample spans geographic regions and ethnicities.
• Unintentionally biased or leading questions.
• Bad response scales. Much research has gone into creating reliable response options.
• Too many, or too difficult questions cause respondent fatigue and “satisficing.”
• Pointless questions that, although interesting, are not actionable for your purpose.

It’s a wonderful idea to review your questionnaire live (in person or virtually) with the first few respondents to ensure they are interpreting your questions properly.

Respondent recruiting

Choosing a database or respondent pool is an important choice. Depending on your targeting requirements, some are better than others. Ask Your Target Market has an easy self-service interface with good targeting options. SurveyMonkey also has its own audience panel system.

Product survey sampling

The process of selecting your respondents from the entire population of interest is called sampling, and it’s easy to introduce errors here in the form of “sampling bias”. For example, if you post your survey on social media and respondents are able to choose to complete your questionnaire or not, then “self-selection bias” is occurring. What’s the difference between those who participated and those who didn’t, and is it important to your results?

In some ways you know your sample won’t be fully “representative” of the general population, such as how all respondents to your online survey must have a computer, which excludes those without computers/smartphones or internet.

Product survey analysis

After you get the results, you’ll need to analyze them. Software like SPSS is standard for large or complex surveys, but Excel is usually fine too. If you’re comfortable with Excel and statistics this will be no problem, however, it’s easy to make mistakes that no one will catch. It’s a great idea to get a second set of eyes on your analysis unless you’re very familiar with the process.

Product survey benchmarking

When your results show that 18% of respondents like your product–is that good or bad? A major challenge with doing your own market research surveys is that you don’t have benchmarks for other products, so it’s hard to interpret your own results.

One solution is to create your own benchmarks, such as by testing your product alongside other products that you know are “good” and “bad”, in what’s normally called “concept testing”. While this adds cost and complexity, there’s no easy way around this problem.

Product survey interpretation

The purpose of all product surveys is to take intelligent action based on accurate information, but which action to take isn’t always clear.

For example, if you asked respondents about their preferred product color choice and 30% said “blue”, should you make it available in blue? Maybe blue is just the favorite color of 30% of your respondents? Maybe they’d be almost as happy if it was green?

Pricing is another major challenge: respondents will always want a lower price, but would they still buy it at a higher one? One solution is seeking the advice of an experienced researcher to design and interpret your pricing questions. Another solution is using a product survey system that uses standard questions and analyses to automatically benchmark and interpret results.

Survey result presentation

Even if you do an incredible job with the survey, others may not trust the results if the presentation is lacking. So if you intend to share your results, we recommend putting in the extra effort to make your final report look professional.