Surveys are a great way to find out what your customers think and want. Knowing your customers’ attitudes to, and perceptions of, your product and services is vital to the future direction of your company.
Similarly, you can ask very specific questions about particular products – maybe a new product you have just launched or are thinking about launching.
Or perhaps you want to find out what clients think of your staff and your customer service?
There are three main things to think about when sending out a survey:
What are you going to say? Ask yourself:
Timing is crucial and can make or What do you change?
What do you leave the same?
Why are you sending this survey?
What do you want to find out?
Who are you sending it to?
Who are your targets? Why?
This last question will relate back to your initial strategy. Why are you doing this?
When should I send the survey?
Break your survey’s success.
What are you going to say?
The information gleaned from a survey is only as good as the questions it asks. What this means is, something that might at first seem relatively straightforward, actually requires a lot of thought and also knowledge about your product/service and your customers. The questions and how you word them – are your survey.
Tips for wording your survey
Keep sentences short – long, waffling sentences and questions will confuse the respondent and lead to ambiguous answers and/or ones that do not answer the question properly. Keep copy (number of words) to a minimum. Less is more. Your customers are very busy people, just like you. They do not have time to spend reading – and filling in lengthy surveys.
Tip: Once you have written your survey, go back over it and see which words you can chop out. You will be surprised how many words are a double up and just how many words you can get rid of. Use simple, easy-to-understand language. Imagine you are speaking to a high school student who knows nothing about your product or business.
You cannot assume your customers have an in-depth knowledge of your product or service, even if they are a customer. If they become confused, due to too much ‘technical’ information, they will stop doing the survey.
Targeting your survey, Who should you send it to?
Targeting your survey to the right people is vital to the results of the survey and its overall success. You might, for example, not want to send it to all your customers. Say you are a restaurant and you have designed a new, fast-turnaround lunch menu to attract the business market during their lunch breaks. You want to find out what kind of ‘quick and easy’ meals they would be tempted by and also whether they would be interested in pre-ordering via a dedicated email address, to save time.
In this instance, you would be best to send your survey to those people who have already been to the restaurant for lunch.
Tip: Try and build your database to suit the kinds of research you will be undertaking further down the track. So, for example, when you gather customers’ contact details, also ask them whether they are primarily interested in lunch, dinner or both?
You might also want to capture where they live – this is often most conveniently done by asking customers for their post code – and whether they eat out often (more than once a week), not so often (every 2-3 months) or ‘only on special occasions’.
By segmenting your database by ‘key criteria’ in the first instance, your surveys will become more targeted and hence the results will be more useful.
At the end of your survey, you can ask respondents to give you basic information about them that will help you build a picture of your customers or your potential customers. Demographic information enables you to interpret your survey results by market sectors. Information usually asked for in surveys includes age, gender, location and, quite often, salary level. Use ranges for age, salary etc.
For example, are you:
18 – 25 years old?
26 – 40 years old?
41 – 55 years old?
56 – 70 years old?
Targeting non customers
You might also want to think about sending surveys to non customers, for example, if you have a new product you feel might appeal to a new market segment. The fact you are considering launching such a product means you should have done some research into the types of people that will consider buying it. If you don’t have a market, there’s no point launching the product!
To test the reaction to a new product, you might want to send it to a totally new database (as well as your current customers of course). You can build your own database – the restaurant mentioned above, for example, could go through the local phone book and capture all the email addresses in the advertisements of businesses in the vicinity of the restaurant – or you can buy in databases (see Echoplus’s help line if you would like us to source a mailing list for you).
Timing of surveys
There are two things to think about when it comes to the timings of email-based surveys. The first is the timing of the campaign in relation to other activity and the second is timing in relation to the specific target customer.
When you decide to send out the survey will depend on your overall marketing strategy and what it is you want to find out. Say, for example, you are a florist who wants to find out which products your customers like most, when they prefer to shop, how often they shop and how much they spend each visit. It would make sense to send this survey out in plenty of time before your next buying cycle. If, for example, you go to two big trade shows a year, use the information captured from the survey to inform your buying decisions.
This is one good reason to have an overall 12-month marketing strategy – so major decisions, many of which are cyclical or seasonal, tie in with each other and help to create a synergistic effect for your business. Another influencer of timings might be new legislation. Say, for example, you are an accounting firm and new legislation is coming into effect that you think will impact on your clients. You would be best to send out the survey - asking clients what they think - in plenty of time before the legislation touches down, so you can be prepared and ready to help those customers affected.
Timings – customer focused
The other aspect as regards timings is when do you actually send out the survey to best suit your customers? Should you send it on a weekday or the weekend? Morning or afternoon? Or perhaps after hours? There has been much research done on this kind of thing but sadly the jury is still out! Some surveys, for example, claim Thursdays and Fridays are better than the start of the week for business clients. Maybe by the end of the week workers feel they deserve more ‘down time’?
The main thing to consider is your own targets – Who are they? When would they be most receptive to your survey? When is it a logical time to send it? If, for example, you are a sailing holiday company asking specific questions about a proposed new destination, you would send out the survey in plenty of time before your customers tend to go on holiday. This is another example of diagnostics helping your survey – the more information you have on your customers (e.g. their preferred holiday dates) the more targeted the results.
Chatting to your customers can help – the results are not scientific but quite often what a handful of customers say about your products permeates to many of your customers. Informal feedback can often give you a starting point for what to ask customers in your survey.
Qualitative versus quantitative surveys
You will no doubt have heard the words qualitative and quantitative when it comes to surveys. In summary, here are some of the main points of difference:
Qualitative is where you ask respondents more in-depth questions about how they feel, what they think. Questions that require longer, more wordy responses. Good for more in-depth surveys where you are trying to measure perceptions, attitudes, opinions. You can ask ‘open ended’ questions – questions that require the respondent to formulate their own answers – or multiple choice answers (you can still gauge attitudes and perceptions but have to word the questions very carefully).
Tip: Never ask questions that can be answered by either yes or no as this tells you nothing unless this is all you needs to know e.g. Would you try this new lunch menu if it was offered at an all-in price of $19.95 yes/no?
Qualitative samples (the number of people surveyed) tend to be small – the value is in the answers, not the numbers. Qualitative research needs careful interpretation of the answers as they can be subjective, emotive and sometimes undecipherable!
Quantitative research requires that a minimum number of respondents be surveyed in order to make the results ‘statistically significant’. In other words, if you ask 5 people something and 4 out of the 5 say they like it, this is not statistically significant (the next five people asked could just as easily say they did not like it). However, if you ask 100 people the same question and 80 out of 100 people say they like it, you would have asked enough people to give you reason to think that perhaps you are onto a winner. You have questioned a statistically significant sample of people. The number of people you need to survey to give you that assurance varies – there are some very complex sums you can do to work it out! – but for a small business 100 to 200 people would be a good start.
If you think you’re on to something – or want to ‘drill down’ a bit more on a specific aspect of your product or a specific question – you can re-send another survey specifically to those people who have said, for example, ‘yes, I would be interested in a new lunch menu offered at $19.95’.
Testing your survey
Testing your survey is one of the single most important things you can do to ensure success. What you think is a straightforward question might cause respondents all kinds of difficulty due to the way it is worded (see tips for wording your survey). Ambiguous wording, technical jargon, acronyms – all of these will put people off. Just because you understand industry jargon does not mean your customers will (or should!). You can do a cheap and quick test by sending your survey to five friends or acquaintances. Ask them to complete the survey as if they were a customer and then give you feedback. You will be surprised at which areas they stumble over!
If you are happy with the survey after your initial test but want to really firm it up, send it to 20-30 of your regulars (customers who will ‘forgive’ you a few mistakes) before sending it to your entire database. The great thing is testing is free! It just takes time and patience and the willingness to change when you receive feedback
Gone are the days when busy people filled in surveys for nothing! Give your respondents a reward for taking the time to fill in your survey – thank them (you should thank them anyway and give them a reward). Rewards can be anything and everything, within reason, but clever marketers will ensure their reward is relevant and appealing to their target audience. Think carefully about what it is you are trying to achieve. With the restaurant example, a voucher for a free meal would probably work. If you are an accountant, is a free tax return quite as appealing?
Money (in the form of coupons), money-off, discounts – all work if a meaningful amount is offered. Alternatively, you could offer respondents a chance to go into a draw for something ‘big’ such as a holiday or cruise (why not consider partnering with a holiday operator if relevant to your industry?)
Using surveys as a PR tool
Survey results make great material for media releases. The results could be relevant to other industry members and also customers – the ‘man in the street’. Indeed, many companies survey customers for this specific reason. You don’t have to release all the information – just the response to a particular question or questions. You could consider adding a slightly controversial question to the end of your survey just to give you an ‘angle’ for a media release.
For example, a restaurant could ask diners whether they agree with new council regulations on pavement dining. If you decide to go down this track, be sure to report your survey findings faithfully and be prepared to back your release up with facts, should the journalist call. For technical information on creating and building your databases see Echoplus’s "Technical How-to on Databases".