How should I run a survey to know the appeal of my product?

You should do both face-to-face questioning and surveys. The face-to-face (which you should do first) will help you get very rich, subjective, qualitative data (that a couple of people like feature X, because it will help them do Y instead of Z). If you're new to this I'd recommend a semi-structured interview - write down a list of all the points you want to cover, but go with the flow. Interesting things - new ideas - will pop out.

The face-to-face interviews will help you structure a survey, which you can use to help validate some of the ideas that have come out of the interviews. In theory, the surveys will give you quantitative information (Z% of people say they'd use feature X).

One big caveat though: surveys are extremely hard to do well (people aren't very good at articulating their intentions; it's hard to find a good sample). It's very common to run a survey, for people to say they'd love to buy a product, and then for nobody buy it.

I wouldn't worry about confidentiality. It's probably enough just to tell people you'd appreciate it if they didn't tell people about what you're doing. Your problem will be getting people to use your product, not stopping them from copying it. At first, anyway.

This is a branch of product management. A good place to find out more is Pragmatic Marketing's web site. They've got a bunch of resources, blogs etc. Or google 'product management'.

Neil's advice is good and covers most of your questions.

I'd just like to add a note to your question what you should ask in a survey. Personally, I wouldn't ask about prices and features if I were you -- at least, for face-to-face interviews.

Participants seldom know how much they would pay for something. They also tend to provide strategic answers: Their guesses are either too low (to influence your pricing decision) or too high (to raise the likeliness of seeing the product for real).

Most customers also seldom know what features they want or need. They also overestimate the amount of features and forget about the complexity it introduces.

In other words: Your clients want all possible features in a simple interface for the price of zero. (If you really want to go this route, research 'Conjoint analysis' as a survey method.)

Instead, I'd suggest you ask participants about their current work flow, any existing problems within that work flow, how much time they need for certain steps and how much it costs them to get these done.

If your application is already build, estimate how much time it takes a client to get it running and to keep it running.

Your advantages: First, you'll understand whether your prospects are aware about the problems you're trying to address. Second, you'll have detailed numbers on how much time your clients will save when using your application. Third, you'll have detailed numbers how much money they could save when using your application.

With these numbers you can easily calculate a price range for your application. They are also invaluable for any sales pitch you need to make in the future!

Nearly nothing beats the headline like: "Make US$ 26,145.34 more profit in just a few minutes..."

Hope that helps.

I couldn't agree with Neil more if he had paid me to :-)

The Pragmatic Marketing methodology is probably the best way to help gather and organize thoughts about a product.

In addition to surveys, you could look at doing a "day in the life" study of the target users of your product. (More info on that at Pragmatic Marketing, too.)

If you do want to do surveys for free, you can use Google Docs' forms to generate the survey and then send out links to it.

If you don't mind paying a little bit for it, you can use Ideascope which is a great tool based on the Pragmatic Marketing methodology.

You also have to realize that your ideas will be copied eventually if they solve problems that your customers have. Generally, it's not the specific idea that makes the difference, but the quality of execution on the idea.

Look at all of the other MP3 players out there. They all do basically the same thing, but the iPod still wins due to things like the quality of implementation, the ways that it solves the users problems, and Apple's excellent customer service.

Find which of your ideas resonate to solve real problems, execute on those well, provide excellent customer service and you'll be a winner. Just the fact that you want to engage your potential customers early in the process is a great first step.

Let me know what you end up doing and how it works out for you.

Joydeep, The answers above are incredibly insightful and dead on. However, I am going to approach this from a more tactical level.

(1) Is your product online? If so, consider UserTesting.com - one of the best services I've seen to gain insight into web features/survey questions. You have the option to specify your target demographic when requesting users to test. The 15-20 minute user videos are returned to you within 24 hours and are incredible (often painful) to watch! For $29/user, it is an amazing bargain. (I am not paid by UserTesting in any way, shape or form.)

(2) In addition to Google Forms, there are a number of free survey options out there, provided you have a list of people to send your online surveys to: surveymonkey.com (my favorite), zoozam, etc. Another option might be to consider LinkedIn and Facebook for a series of mini-polls, targeted to your market.

In any event, good luck!

JoyDeep , keep one thing in one mind. If your audience is based in India, you are not going to get good response on surveys. And not every one would take out time to respond to survey. For eg look at my site skill-guru

There are so many users who take the test and do not leave feedback or rate the test. It is so convenient for them because we ask for feedback and rating at end of test.

Now imagine an end user to open up mail and answer your questions , I do not think you would get correct feedback.

What I would suggest is launch your site , let people use it and allow them to give feedback at end of their surfing/shopping .

Step 1. Do a focus group and ask what your customers want/features added Step 2. Ask also your engineers what they want added Step 3. Structure the survey using a Likert scale (preferably even-numbered) and ask how important a feature/aspect of the product is (1=least important; 10=most important) Step 4. Do a Factor Analysis on the results - you should get a reduced number of factors Step 5. Based from Step 4. try to group the factors, and reflect the grouping in a new survey Step 6. Run another survey based on the factors from Step 4.

note: highers sample size is always better