Sorting & Visualizing Data To Ask Better Questions

Visualizing data is something that gets lost in the rush of sorting out marketing initiatives. If you’re working in marketing, you probably recognize this scenario:

(internal) Client: “Can you give me some insights about …?”

Marketeer: “Sure, I’ll get back to you a.s.a.p.”

You, the marketeer, send a few moments later an update with insights. Some screenshots rapidly dropped into an e-mail with some remarks and maybe, if you’re feeling keen at the moment, some advice to get some actions out of those insights.

Nothing wrong with this scenario. Although… You should take more time to analyze things more properly, for several reason. One of the most important reasons is that you’re not taking use of the moment to ask better questions.

Sure, you have to answer the question of your client, but NOW is the moment to ask for a little bit more time to analyze things even deeper and uncover what hasn’t been seen. It’s time to uncover your unknown competence as a Business or Marketing Analyst. Well, a simple start to get to that title ;)

Allow me to help you with this example below, with only one table with simple data and some Google Spreadsheet visualizations.

Example: Top 10 Best Sold Car Brands, In Belgium, In 2014

For this example, I’m using a simple table I found in a car magazine on the web. Nothing more, nothing less.

Here’s the unfiltered, unsorted data:

Original source: http://www.vroom.be/nl/autonieuws/top-10-meest-verkochte-automerken-2014

The Power Of Conditional Formatting

Colors are essential in giving meaning to numbers or segments. So… next up. Color those values.

As an analyst, these colors give you a new way of thinking about that top 10. Some questions that could be asked, thanks to the help of some simple colors:

• Which car brands are in the top 20?
• How much marketshare do they take?
• Are there brands that didn’t sell any new cars? Is that even possible?
• …

Very generic questions. Nothing spectacular.

If you’re part of the government, let’s say your the minister of transport, other questions could arise with this simple table of data:

• What’s the share of new versus second-hand cards?
• Is there a trend that can be spotted, for instance, is there a shift from big(ger) cars to smaller ones or the other way around?
• Do people change from a Diesel engine to a Petrol engine or the other way around?
• Is the car really still king of the road for small distances? Or is there a shift towards public transport or even the electrical bike?
• How about carpooling? Is there a connection between more new cars on the road and carpooling? Are more people single riders or are they carpooling?
• …

The questions to be asked are endless. The data to be collected to get answers as well.

The Power Of Sorting Data

Sorting data can give extra insights.

A whole different sight, huh?

Hyundai is killing it. Mercedes is doing very well, Audi and BMW as well. Volkswagen (pré software scandal) is losing but they did sell the most cars in Belgium. Although Renault sold +44.000 cars, 5k less than Volkswagen, they loose 3.37% according to 2013.

Makes you think and put things in perspective, doesn’t it?

The Power Of Visualizing Data

Visualizing data is an art, and unfortunately, is time consuming. That’s mainly why lots of marketeers and people in the marketing analytics field aren’t using visualization enough. But… the power of visualization is not the be underestimated!

The art of visualization is in the storytelling.

Making beautiful visuals is awesome but you don’t need spectacular visuals to tell a story. Telling the story is more vital than the design of the visuals, if you’re using the _right_ visuals!

You want buy-in on C-level? Use visuals & tell a story. Extra tip: talk in budget or revenue to be gained. You’ll get their attention for sure ;)

Visualizing Total Cars Sold

Sorted by number of sold cars (what we all see with the naked eye in numbers):

Visualizing Delta 2013

Same numbers, different sorting. Visualized, it has more impact at first impression:

In this case, it is more interesting to sort the table. This way you’ll get a more comprehensible visual than the one here above. This one is much more readable. Depending on the message you want to bring, a positive one or a negative one, you could sort the numbers the other way around to make your point. In this example, I’m using negative to positive delta.

Some extra tips on making this graph more meaningful, if the software allows you to do so:

• Use red for negative values
• Use orange for in between values (you set the range)
• Use green for positive values

You could even use different colors per brand. You could highlight the car brand you’re focusing on.

If you have time to do this, this is the moment when the designer comes into the picture to help you.

Questions you can ask with this graph:

• How many old models from a certain car brand get replaced by a new car of the same car brand?
• These new cars, are they the owners first car or are they a second car for the family?
• Does a certain age group of people choose for a particular car brand or type of car?
• Does a certain age group of people choose to drive longer with the same car? (maybe Opel has that problem? Good cars, low sex appeal / reason to buy a new one)
• Are some cars of certain car brands more durable than others? Or the other way around, that it could impact sales or even word of mouth (you should buy this amazing car / you shouldn’t buy this shit car)?
• Are there specific models that are doing amazingly better than other models? (the Volkswagen Golf for instance with the 7th edition already)
• …

The list goes on if you’re willing to think broad and big enough. Data helps you to see an other angle. An other angle to make you think.

Other questions you can ask, if you’re questioning marketing or even resources:

• What’s the impact/share of marketing on these numbers?
• Is there a difference in budget allocation, channels or even timing that could have made a difference?
• What strategy is each car brand using? Is there a difference per model of car?
• …

Lets get into even more into this angle with the next visuals. Now it gets really interesting!

Visualizing Market Share Per Country

Interesting to look at the data this way, huh? Just adding an extra column to the data gives an extra dimension to it. Adding more data can help you to get more insights or even another angle to look at the numbers.

Wondering and thinking about these now:

• Do German cars take the lead in Belgium because they already out numbered the other cars (people tend to look at what other people bought in their surroundings)
• Has the price of the French cars something to do with their suddenly succes? Or their new and more beautiful design (they tend to seem cheaper + they did a really nice redesign of their lines)
• Has the image of French cars a part in losing market share (they tend to have the image to be bad cars)
• How do the sales shift from one brand to another? Which brand takes over the market share from which brand?
• Are people aware that Hyundai actually is a South Korean car brand? Or are they mixing Hyundai up with other Japanese brands. (Toyota has a very strong reputation)
• …

This visual below is even more interesting, combining countries and delta into one visual:

You can ask some interesting questions with this single visual:

• Does logistics play a huge part in pricing and sales?
• Do German and French car brands have more representatives and dealers in our country than other brands in the top 10 best sold cars for 2014?
• Do German and French car brands more marketing budget focused on surrounding countries?
• Do they sell, in proportion, as much cars in their own country as ours? And how is that marketing budget allocated?
• Do we see the same thing with Ford & Hyundai? Do they focus on Europe with the same proportion as Volkswagen, Mercedes and BMW do?
• What we see on this map if we would also plot the next 10 best sold car brands
• Do we trust car brand of surrounding countries more than countries like South Korea or the United States?
• …

Conclusion

Although this might not blow your mind and these simple tricks might not seem like rocket science, they aren’t always used to unleash the power of the data you have. You have get some time to generate buy-in by formatting, sorting and visualizing data and with those simple techniques, ask the question others didn’t ask.

Be the genius. Spot the things others didn’t or can’t see. Invest in working with data. Data can take down walls (digital transformation) by asking the right questions. Asking questions will reveal that there is so much more context to take in account.

It could be the start of getting time to collect more data and analyze more properly to help shape the business you’re in, even more.

Originally published at driesbultynck.com.