Sign in or start a free trial to avail of this feature.
7. Designing Data Driven Charts
Data-driven charts are probably the easiest to build but you need to make sure that your visualisation matches your action title.
To explore more Kubicle data literacy subjects, please refer to our full library.
Lesson Goal (00:06)
The goal of this lesson is to learn how to create charts that validate our action titles.
Using Quantitative Data (00:32)
When validating an action title, you can use quantitative, or numeric data, you can use qualitative, or non-numeric data, or you can use a combination of both. Action titles that can be validated by quantitative data only are often the easiest to put on a slide. We simply take the relevant data and create a chart that visualizes the data for our audience.
For example, if we have an action title stating that one particular market is growing, we can validate this with a chart showing revenue growth over time for each region. We may also add a label to the chart, drawing attention to the relevant area.
Mixing Quantitative and Qualitative Data (02:00)
In many cases, our action title can only be validated using a mixture of quantitative and qualitative data. In this case, we can supply the qualitative information we have, and we may be able to generate quantitative insight by considering other, similar situations. For example, we may want to consider the likelihood of a pharmaceutical product getting regulatory approval. We can visualize the approval process, but we cannot definitively say whether the product will be approved. However, we can determine the probability of the product being approved, based on the success rates for similar products in the past. In this way, we add quantitative insight to a qualitative process visualization.
However, we should avoid being overly obsessed with quantitative data. For example, if we have little confidence in our calculated approval probability, we should simply not include it. It’s best not to include numbers just for the sake of it, as it is possible to derive value from qualitative data alone.
In the previous lesson, we identified a set of action titles to address various sub issues from our issue tree.
Each action title, we'll need a slide that shows the evidence for your point of view. So for example, if your action title says that a market is large and growing, you should show evidence that the market is indeed large and growing.
When validating your point of view, you will typically rely on either quantitative information, qualitative information, or a combination of both.
In this lesson, I'm going to focus on quantitative information, which should always be your first port of call. Normally I will begin by identifying the action titles that have been validated by quantitative information only as these are often the easiest to put on a slide.
We simply take the relevant data and visualize it in a way that makes our insights obvious to our audience. For example, let's look at market size. Here we have stated that in the action title, the market is large and growing, particularly due to demand from Asia. Prove this point of view, I've created the following chart of camera that shows revenue growth year on year for the past seven years, split by region. Clearly Asia is the growth engine in this market. If you want to emphasize the point, you can also add a call-out highlighting what percentage of growth Asia contributes. Let's now take a look at another example, the estimated cost to market and distribute the drug post launch. Given that the company has launched new drugs in the past, we know that they can estimate the cost of launching a new product with an acceptable level of accuracy.
So in my chart, I'll break down the total cost into its various components using a waterfall chart. I'll also highlight the largest cost segment that is mentioned in the action title. In most cases, your slides will not be so easy to design because the action title will have been generated from a combination of qualitative and quantitative information. For example, let's take a look at the likelihood of regulatory approval. In this chart I've graded a sequence of stages corresponding to multiple clinical trials. This particular drug is at the penultimate stage, while it's impossible to gauge exactly the probability of success in both stages, we can stead apply some quantitative insight by researching the likelihood of success for similar drugs that have got to this stage in the past. To research, we can deduce that this product has a 60% chance of getting through the next stage, and an 80% chance of getting to the final stage. And this way we have combined qualitative and quantitative information to backup our point of view. Well, it's important to always consider adding quantitative elements to your analysis, analysts often run into problems by including numbers that do not stand up to rigor and are quantification for quantification sake. Say in the previous example, we include these percentages but have a very limited conference in their accuracy. Management will often latch onto a number because they trust you even if you are not that confident in the number.
If your prediction or insight turns out to be incorrect, this can put you in a very difficult position. When faced with the possibility of relying only on qualitative information, it's critical not to cave in to the temptation to quantify everything. Value can be gained by relying only on qualitative information and leaving certain risks or issues as undefined or on quantifiable. I'll expand further on this topic of relying on only qualitative information in the next lesson.