The SEED approach to data visualization: explore

Ah, springtime! The warmer weather, sunshine, blooming flowers. It is the perfect time to break out of your comfort zone and explore – your data, that is!

As you may remember from the last blog post, we introduced you to SEED – a structured approach to implementing game-changing data visualization and storytelling in Market Research. Over a few weeks, we are explaining the process through a 4-part blog series.

The SEED approach is a challenge (you have tasks to complete!), and in the end, what you create will be a guide to building your own data visualization strategy. Each blog post will focus on a letter/activity in SEED. Be sure to complete the step before moving onto the next letter/activity to ensure your success.

Today we’re focusing on the first letter “E” for “explore” – as in “explore your data.” For many people, data is an intimidating thing. In its raw form, it’s simply a mess. But if we get this right, we can turn raw, messy data into visualized insights that will set your research apart in your stakeholders’ mind. So, let’s get started.

The first step in exploring data is to create a map for your explorations. Start by figuring out what data you already have available. Make a list. This list should contain both your structured and your unstructured data. Don’t skimp on the data either – if it’s available, add it to your list. No excuses that it’s not formatted correctly, or it’s too hard to get, or unstructured data is a pain!

Now that you have the initial list of data. Consider what value each bit of data is likely to provide or what insights may be hidden within that data. For example, while you are likely going into any research project with a specific objective in mind, consider what variables could be hiding your next ah-ha moment. At this stage, you are the scientist hypothesizing what new and exciting information or context you may be able to uncover. Write all this down in the same list. Here’s a simplified example:

• Consumer’s mobile device type: Could this potentially reveal that Android users are more likely to buy the product than iPhone users? Why would that be, and what other data do we have (or can we collect) that we can use to determine the reason?
• Price of the product: Can we get away with a higher price? If we combine this with other data, might we find that gender or generation impact consumer’s willingness to buy at the current price? Do people consistently buy product x with another product?

Get creative with your hypotheses!

Before you know it, voila – you have just created the map for your exploration. Now that you have a clear picture in your mind of what value the data may be able to provide. Next, like any good explorative scientist, we need to test these hypotheses. This is where the analytics will come in.

Of course, now you’re likely thinking, ‘that’s great advice for the structured data, but you’ve forgotten the unstructured data you insisted we include.’ Ah, but I haven’t forgotten! Here, text analytics is going to be your best friend! Manually sifting through and coding unstructured responses can be painful. But with text analytics, you can automate the entire process, categorize responses, and assess sentiment with ease. Best of all, while your hypothesis from the previous step is likely fantastic – automation can help you mine for concepts you may have failed to consider!

Once you have finished exploring your data for all the wonderful possibilities it holds, you’ll be ready to move on to the next step! If you aren’t sure that you’ve covered all necessary elements, don’t have the resources, or are daunted by the plethora of raw data – give us a call. We can help!