The DataHero Blog

Designing the Learning Curve

August 15th, 2013

Designing the Learning Curve

When we set out to build DataHero, our goal was to create a product that made data analysis simple and fun.  We wanted to make an application that empowered our users to find insights and make better decisions without requiring help from a business analyst or data scientist.

To do this, we continually iterate our design to create interactions that are simple enough for our most novice users while making sure these same interactions accelerate even our most advanced users.  Internally, we’ve called this design methodology:

Designing the Learning Curve

The Learning Curve

We know that we only have a few seconds in our first interaction with a customer to make an impact before the user presses back, changes tabs, or moves on forever.  By designing to the learning curve, we ensure that the most common tasks can be achieved in the least amount of time and provide a meaningful result as fast as possible.

Let me give a concrete example: Imagine that I have a bunch of stock market data and want to make a graph of the Russell 3000 Index Data over time.  Take a look below to contrast the steps required to create this visualization in Excel versus DataHero:

Excel:

Select the data to graph -> Select the data range -> Select the graph type -> Choose axis for each piece of data -> Result

DataHero:

Select the data to graph -> Result

By requiring three extra steps to get to the result, Excel users face a higher cognitive load and the barrier to a successful outcome is higher.  The result is a learning curve that is a little bit steeper in Excel because it’s asking the user to answer many intermediate questions before achieving a result.

DataHero Adjusted Close by Date

In DataHero, we immediately present the user with a result and then allow them to take the next necessary step or steps to refine their data.   Someone who is not yet an expert can quickly associate each step to a specific result and, more importantly, a feeling of accomplishment is created with each intermediate step.  Changing the data range, modifying the default chart type or swapping the axis can be done in subsequent steps if they are even necessary.  Each step produces a new visible result.  New users appreciate the simplicity and guidance while expert users can appreciate that no unnecessary intermediate steps are added to slow them down.

Besides making each action simple and understandable, we use our algorithms to guide users through questions where they may not know the best answer.  For example, what chart type to choose can be a difficult question to a novice user who is not a Tufte expert and doesn’t instinctively know that a continuous set of data over time would be best communicated as a line graph.

We designed the technology that drives DataHero to reflect the learning curve.  In DataHero, you can simply, “drag on your question” and it automatically chooses the chart type, axis for each data attribute, and even default grouping or aggregations where appropriate.  By relying on our underlying technology in our smart data decoder and chart magic, DataHero makes the time to first result fast and intuitive.

Our focus on designing the learning curve is one of the many things that makes DataHero easy and fun to use.  With each new feature we add to DataHero, we are constantly designing the learning curve for our users to make sure everyone is empowered to answer the questions that matter the most to them.

By Jeff Zabel

Create my Free DataHero Account

Get the fastest, easiest way to understand your data today.