DIY market research

A survey of your customers can be a powerful tool to help you make decisions. Here are some tips to help you do your own research, or to prepare yourself well before enlisting help from a professional. Just the process of thinking through these questions will help you learn more about your target market.

1. What are your goals?

Take a lot of time to think about this one. What do you want to achieve by doing this survey? Think higher then just the answers to the questions.
Do you want information that will help you tailor your offers, re-design your product, improve your store layout, segment your market, target your advertising.  Do you want to know how to pitch your pricing? Are you going to revamp your graphic image or rewrite all your brochures?

A good way to work on this is to decide what you are going to change as a result of knowing what your customers think and how they say they will behave.
For example, if your customers tell you that they find your shop difficult to access, can you make alterations to meet their preferences??  If you can’t make a change, then it is generally a waste of time asking a question about that aspect.

2. What are you trying to find out?

Be very specific. Write down a long list of all the things you would like to know.
For example, you may want to know about:

  • The profile of the people who buy your products – their age, sex, life stage, lifestyle, family, occupation, income, what magazines and books they read, or radio stations they listen to (ie. where you might catch their attention if you advertised)
  • What customers like about your products – in this case remember to be very specific when you ask the question so you know which product and which features they are talking about in their responses
  • Why people don’t buy your products – their reason may be nothing to do with whether they like it or not
  • Where else people go to  shop for similar products, and what they think about your competitors
  • How they perceive your prices – and what prices they see as value for money (or you may want to know if  your customers feel that high prices indicate that the product is better quality or more desirable for another reason).

Keep focused on ‘value’. With all of these topics you are really wanting to know how your customer‘values’ you and your offering, how ‘valued’ they feel, and how you might increase the ‘value’ of your product or service in their minds. This information will help your staff to recognise their own ‘value’– they will enjoy the positive feedback and feel motivated by the obvious opportunities to do better.

When you have a comprehensive list of questions, analyse each one, ask yourself whether the answers you may receive will help you to achieve your goal.
Delete the questions for which you can’t change anything, and take out those that give you ‘interesting’ answers rather than answers that will drive your decision-making.

Who will you ask?

Who will be your sample? Are you going to put a questionnaire on your reception desk, send one to every customer, or telephone a representative few?

Do you know how many people you need to survey in order to gain a statistically representative sample on which you can base your decision-making? To work this out you will need to identify how many different ‘groups’ of customers you have and whether you want to compare the responses of these different ‘segments’.

Survey too few and you won’t be able to rely on the answers representing the views of all, and too many adds cost and time to your project.

What method of research will you use?

Here are some choices from most expensive to least expensive:

  • Personal Interview – face-to-face interviewing – sometimes conducted in a mall or on the street. This can include a ‘quick question’ while a customer is instore.
  • Telephone – probably one of the most popular methods but getting very difficult in today’s world, even with a customer list.
  • Mail – inexpensive, but fewer people will bother to respond.
  • Web based – can be fun, quick, enhanced by ‘rewards’ for responding, but there is no way to check that the respondent is telling the truth.

Plan your research carefully

Most professional researchers will begin with qualitative research – focus groups or in-depth unstructured interviews – to establish the issues that are important to a small number of typical respondents, so that the questionnaire can be designed efficiently.

It really is a good idea to get someone from outside your organisation to moderate your focus groups and do your interviews, because your customers will usually be somewhat reluctant to tell you their criticism, and they may not be happy to reveal too much of themselves to you either.  But they will love telling a stranger their opinions, their problems, and their recommendations for putting things right!!!

Once you know who you’re surveying and the type of method you will use, develop a time line, a summary of how long it will take from designing the survey to analysing the data.

Then do a cost estimate. You might break down cost by each step involved. Include things like postage, time, and paying someone to sort out the results, (cost it out, even if you are going to do it yourself – as your own time is a cost to your business).

Design the survey.

Your research professional will be able to help you design your questionnaire too – this is a critical part of your research. It is very easy to write questions that could be read incorrectly, or that suggest the ‘correct’ answer – and even easier to unwittingly write questions that really confuse the interviewee (and sometimes the interviewer!) Beware of words that have double meanings, long lists of alternatives to choose from or rank in importance, and yes/no answers where the next question depends on the previous one – eg if no go to Q 13, if yes go to Q12. It pays to get help!

Keep it short. Warn people how many questions are in the survey.  You can do this in the introduction or the letter/e-mail inviting them to take it. Knowing it’s short will encourage them to participate.  You may want to offer a reward to each respondent – a small gift or discount will increase the uptake.

Pre-test your survey ‘instrument’ (the questionnaire, usually!)

Pre-testing will help you determine whether the survey is easy to understand, if people are able to fill it out, and other problems that may occur. Rewrite the survey if you need to.

It is really tempting to skip this part, but it is possibly the most important step. You can waste your time, efforts and money but going to your customers with a questionnaire that has ambiguous or nonsense questions, or if the answers that people give to the questions are not clear. You will be surprised how something that makes perfect sense to you, and has very clear meaning in your mind, gives rise to confusion or an answer where you can see that the respondent interpreted the question differently from what you had in mind.

You can ask your friends and family to be your guinea-pigs for the pre-testing – or you can ask real customers. Start with about ten people, and keep re-writing until it all works well.  Plan to achieve half a dozen completed questionnaires where every single question is answered without a problem, before you do the actual survey.

Prepare your analysis tool

It pays to think about the collection and analysis format at this stage, or even before this, and to test this with the data you collect in the ‘pre-test’ stage. You will be surprised how some of your best ideas don’t actually work out that well in practice.

If you are going to analyse the results yourself, MS Excel is a good tool to use, especially if you are familiar and competent with it and can handle pivot tables, data sorting, and the myriad of useful facilities in that programme. If you can do the basics with Excel you can still make graphs with the click of a button, as long as your data is entered in the right format.

Don’t even think about counting the answers manually – unless you are asking just one or two questions. If you are not computer literate and able to do a really thorough analysis, get someone in.  But PLEASE get them involved BEFORE you finalise your questionnaire!

Do the actual survey.

Collect the data and put it into an organized format. If it is a mail survey, be sure to emphasise the ‘reply-by date.

Analyse and interpret the findings

If you’re using quantifiable information you can analyse with statistics. However, expect to spend some time learning how to do statistics.

Aim to show your results in tables and graphs so that you and others can see the results at a glance and compare factors – eg how many men answered a certain way compared to women.

Include a range of people in the interpretation of your results – sometimes people who have nothing at all to do with your business can see the meaning or relevance of a particular finding better than the people up close.

Be prepared to learn things you would rather not know, and to change things that you like the way they are!

Make some changes!!

Thank your customers, and if possible, tell them about your findings. You don’t have to reveal all the detail, but you can comment in your newsletter, or in a notice in your shop, that your recent customer survey told you ‘ this’ or ‘that’, so you are making a change in accordance with their feedback.

Those who have participated in the survey will be delighted to see that they helped bring about a change, and those who haven’t will respect you for asking for feedback and being prepared to meet their needs.