About three weeks ago, I wrote a post about the necessity of knowing your audience. We ourselves are currently investigating our audience. We asked our newsletter readers to fill out a questionnaire. After we have collected the results of our survey, I started analyzing the results. It has been a great week for me! In this post, I would like to present some of our results and try to translate it to some practical tips for you all to use.
Read more: How to analyze your audience »
What did we want to know?
Our audience has grown rapidly during the last few years. Yoast began as a WordPress blog, but is currently much more than just a blog. Few years ago Joost was writing and working alone in a small room in our house, while we currently have a nice office and over 10 employees. That made me wonder: who is our audience nowadays? Do we still appeal to a technical (nerdy) group of people? Are our customers mainly developers? Or is not all of our audience that technically skilled? And which products appeal to our audiences?
My first analysis of our audience focused on whether or not there exist different groups within our audience. My hypotheses was that we would at least have two types of visitors on Yoast.com: a technical ‘nerdy’ group and a less technically skilled group. The aim of my analysis was to investigate whether or not these different groups (intend) to buy different products.
How did we do our audience analysis?
We collected the data for our audience analysis via an online survey for which we invited all of our newsletter readers to participate. Our final sample consists of 2796 individuals. We asked all kinds of questions, including age and sex, intentions to buy our products and questions about the extent to which individuals liked Yoast. We also measured the level of experience with internet and WordPress by asking people to grade the extent to which they agreed to the following statements:
- I have a high level of knowledge about WordPress;
- I am an experienced WordPress User;
- I am a skilled Web Developer and I am capable of maintaining my own website.
A total of the scores on these statements was calculated (resulting in a range from 4 through 16). As people score higher on this scale (or nerdiness as we like to call it) they’re more experienced. This measurement of experience is important in our analyses, because it is one of the variables on which we assume groups within our audience to differ.
|mean internet experience||mean age||% women||N|
|group 1 : the women||11,4||45,9||100||600|
|group 2: the nerds||14,6||36,0||0||666|
|group 3: the young, male, non-nerds||10,5||32,6||0||615|
|group 4: the old, male, non -nerds||11,1||54,1||0||910|
Simple alternative to a cluster analysis
I can imagine that most of you do not consider it fun to figure out the complicated statistics of a cluster analysis. Although our solutions are in fact very elegant, it remains a random grouping. A cluster analysis is just one way to identify groups, and you could just as well use another (more simple) method. Dividing groups on the base of sex, education, age or experience could well do the trick. Decide which variable is most important to differentiate your groups upon (in our case, this would have been internet experience). Then create a number of groups on the base of the scores on this variable.
After computing the scale of experience, we executed a so-called cluster analysis. This is hardcore statistics indeed. You can try this at home using SPSS (expensive, but nice) or R (open source, and nice). A cluster analysis results in the identification of a number of different groups of individuals who display similar scores on the variables of your choice. We identified four groups based on sex, age and the level of experience.
Table 1 shows that the four groups score differently on variables used in this cluster analysis.
Our first group consists of women, with an (for our sample) average level of internet experience. The second group consists of men with a very high score on internet experience. We will call them the nerds. Then the cluster analysis revealed two other groups, both containing solely men. One group contains young men with a relatively low level of internet experience and the last group contains older men with an average level of internet experience.
What did we find out?
On the base of sex, age and experience, we can differentiate between four distinct groups within our total audience. Interesting. But quite useless as well. We have to investigate a bit more. Do these different audiences also buy different products?
|WP SEO Premium||Video SEO||Local SEO||Theme||Website review|
|group 1 : the women||37||9||11||20||11|
|group 2: the nerds||50||19||26||17||14|
|group 3: the young, male, non-nerds||43||15||21||19||19|
|group 4: the old, male, non -nerds||39||12||16||15||12|
Analyzing intended shopping behavior yourself
If you want to study whether different groups have different shopping intentions, you should add a question about intended shopping to your questionnaire. Offering a limited number of answering possibilities makes analyzing your results much easier, especially if your sample is large. Subsequently you calculate the number of people who intent to buy a certain product per group. Make sure you use percentages instead of raw frequencies. Percentages make it possible to compare between groups (even if group sizes differ). Making a table like table 2 in this post will create a clear overview of your results.
As you can read from table 2, the different groups within our sample do have different preferences. The nerds have a higher intention to buy one of our SEO plugins than the other groups. However, the highest intention to buy a Yoast Theme is found among the group of women. Website reviews appear most popular among the young, male, non-nerds. Of course, some differences are small. These results nevertheless show that different groups within our audience prefer different products. We are now looking into possible strategies to optimize our sales. How can we better address women at our theme page? How can we reach young, male, non-nerds with our reviews? Conclusion: we are not done investigating…
Why could an audience analysis be important for you?
Knowing your audience means that you can anticipate upon the needs of your audience. And this in turn can increase your sales. Knowing the preferences of your public allows you to adapt your marketing and sales strategies. You should consider doing similar research if you don’t know your target audience well. Even if you think you do, the results of research like this may well surprise you!
Keep reading: How to analyze your audience »