Last week we took an in depth look at Market Research and how to hold your own census, and we promised to follow with how to analyse your results, so here it is!
In our discussion of the 5 steps to host your own census, the first step was to:
Define your problem We generally undertake Market Research to solve or shed more light on a problem e.g. “Why are our sales decreasing?” Or to prove or disprove a theory e.g We are losing sales because our prices are too high. So start by writing your problem down at the top of your page, as a nice big heading, everything from here has to help us solve that.
So dig out your piece of paper (which if its anything like ours now has coffee stains, phone messages and assorted flower doodles all over it) and refresh your memory as to why you went to all this trouble in the first place. What were you wanting to shed more light on? What question needed answering? Which theory needed proving? When it comes to analysis, this problem will again be the hero at the top of our page. By the time we finish mining the data we received, we want to feel informed and educated and ready to spring into action.
If you used Survey Monkey, take advntage of the inbuilt analysis tools or transfer your data to a spreadsheet. If you used email or a form, you will need to collate all your data into a useable format (If you are confident using a program such as Microsoft Excel or Mac Numbers, I would highly recommend it as this will allow you to evaluate your data and turn the responses into tables and chart,s bringing your findings to life visually).
When we designed our survey we looked into the importance of different question types. Now that we are ready to analyse, we need to consider how we asked the question, in order to decide how to analyse it:
- Qualitative answers are best typed as a list and read over a few times, before making generalisations, identifying themes and noting any key insights. When it comes to qualitative responses we have to think about weighting, not getting too caught up in what one person says and we need to think about responses as a whole in context of who our target market is this is. This is where our demographics help (more on that later…). With our qualitative data if we left an area for other comments, we would read over all responses and make generalisations like “most consumers think we are doing a good job and don’t need to change anything”, “when asked for any other websites they use often, a few responses mentioned eBay”
- Quantitative answers can be easily analysed by summing and averaging responses and also by comparing different demographic groups. The great thing about quantitative data is that you have set the boundaries of the response type in your inital question design, making collating the data and drilling down (moving from a top-level view to sub groups of respondents) much easier. If you click the image below, we have completed some examples of creating data tables to collate responses, and then using different charts to make the data come to life.
- The Demographic questions which we ask at the end of our survey put the responses into context. This allows us to break out each question into different groups, to look at how those groups differ and also discard responses from groups that we may not deem relevant to our survey. As an example of how the context will help ensure we make our decision based on relevant results, first consider the above graphic where we saw that 54% of all respondents liked our current layout a lot, initially this is exciting news, but, is it relevant? If we then look at the graphic below, and imagine our target market is Women aged 56-70, we can see how valuable splitting by demographic is, all of a sudden we are presented with the fact that of all women surveyed only 7% aged 56-70 like our layout a lot, compared with 20% aged 41-55 and 18% aged 26-40, this immediately puts up a red flag (where we thought we had a lovely green one). It seems as though our current layout is too young for our target market, we would then look at responses between these three groups to other questions to diagnose the specific issues, and then form an action plan.
Once you get started you may find analysing your data is a bit like falling down a rabbit hole. You start thinking of more and more questions and keep drilling down in different ways to get the answers. At the end you should hopefully have some very powerful insight to assist you with your initial problem, plus a whole lot of answers that you didn’t know you asked for.
So now you need to spring to action, but first a final thought:
- Reflect – take a moment to think about everything you have learned and consider are you looking at the good and the bad?
- Listen – what are the top 10 useful facts?
- Re-Confirm – clarify with your target market that you have understood ask another one or two questions let them know how valuable their input has been so far
- Act – take what your target market are telling you and implement and tell them you did it because of their input
- Start again – do it all again in 6 months. Look for the changes in responses and whether your action improved key areas
So how did your survey go? If you need more help with how to analyse, just let us know where you get stuck! Next week we will launch a 26 part blog, we are very excited about it… any ideas what it might be?