Know-How Exchange : Research/Metrics

I think the co is right! As to your questions:

1. % of client population to survey to get a meaningful sample size: it depends how different clients are from each other. If they are very similar, you need to survey only a few...question is how to figure this out:
a) you can identify meaningful segments from previous experience: make sure you sample a few at least from each segment, stop surveying when no new information come in
b) you have to find this out during current market research: start with clients that look most different on paper (eg company size, type of products ordered,...) and compare their responses.

Statistics does offer a way to calculate the required sample size if you have an idea about how different clients are (the variance sigma) and how far off the sample answer may be from your client population answer (the error margin).

2. Yes: you can dig much deeper in face-to-face and hence are likely to learn more from each respondent. However, you will still need to cover at least 2 clients from each key segment

3. How large is your budget? If you have sufficient resources, investigate all segments of some importance to you. If you have few resources, and your top clients differ from eachother, it may make more sense to survey each of them

Good luck!

There is a helpful sample size, confidence interval and confidence level calculator here: https://www.surveysystem.com/sscalc.htm which you can use for free.

It's a good idea to run some formal research to find out from clients what they want and think of your services instead of relying solely on sales feedback.

But before you start, agree and write down a set of objectives for the market research. "What customers want and what they think of your services" is a very generic question and is liable to lead to some bland answers. The questions need to be more specific than that.

Begin with the end in mind, and you'll construct research that is actually helpful - that can lead to change and action. Bland research usually sees management patting themselves on the back, or else it gets buried in a filing system never to see the light of day. Ask tough, specific questions, if you want to be able to make tough, specific decisions.

If you want help to frame a research project, there's plenty of people here who can assist, just click through on profile names to contact experts directly. Alternatively, use the link at right to Under "Hire an Expert" to Post a Project and experts will bid for your work.

Hope that helps.

ChrisB