For years businesses have worked with and focused on desktop dashboards but given the increased mobility and rise in the use of tablets versus PCs in recent years the mobile dashboard has come into the spotlight. Continuer la lecture de Mobile dashboards and efficiency
10 common mistakes in Dashboard Design
1. Not thinking of the story telling
‘What story do I want to tell?’ is the number one question you have to ask yourself when creating a dashboard. Not thinking of the message you want to convey will ultimately result in an irrelevant assembling of data and misrepresentation of your company’s business. Creating a dashboard is like writing a book, your book.
When a company comes to us for a marketing dashboard they do so to be able to get a simple, single glance view of their KPIs. Our job is to know that technology can be beautiful and simple and that to achieve that effect it needs to be relevant.
Now what exactly do we mean by relevance? There is a lot of data available to the client today and it is very tempting to put the entire gamut of information on the dashboard. It is little wonder that most dashboards we see are bright, shiny toys that the clients rarely use.
The other part of relevance is played out in how we represent this information on the dashboard. Oftentimes designers fail to recognise that the dashboard is a tool. In order for a dashboard to function to its full capacity, we need to know where to use numbers and which data to visualise.
A basic guide to data visualisations:
NUMBERS or VERBAL LANGUAGE –
Numbers and verbal language can be very effective in communicating the message when used correctly. Not only does it take far less space to write a single value as compared to a graph, but it also serves to keep the focus on the message itself.
Best Use: When a single value needs to be communicated.
A bar graph shows relationships between different data series related to each other and make a very simple yet strong statement. They are the perfect vehicles to emphasise individual values due to the weight of each bar and thus make comparisons easy.
Best Use: When displaying discrete items across nominal or ordinal scales.
Line graphs show the overall shape of a series of value thus making them extremely good for emphasising the movement or trend of data over a period of time.
Best Use: When showing the movement of data across a related series is when we turn to line graphs.
Heat maps are a visual display where quantitative values are displayed as a variation in colour in proportion to their measurement. While their most common recognisable form is as a geographical map, they can actually be used in all kinds of displays including a tabular format.
Best Use: To display a large quantity of data in a very limited space.
Tree maps are used to display hierarchical data using rectangles nestled into one another to display the relations. In addition to that they can be used to display 2 quantitative variables by the size and colour of rectangles.
Best Use: To spot particular trends of interest in a large amount of data.
Bubbles are the perfect way to display data along three variables where the X and Y-axes are used for two variables and the size of the bubbles represent the third. When divided into four quadrants they can even be used to categorise the bubbles.
Best Use: To analyse for patterns/correlations between metrics.
The leaderboard displays a metric over a dimension in a way that allows you to easily categorise each dimension value’s “performance” in terms of the metric value.
Best Use: To compare the performance over a set of values and compare them to a total or average whole.
A battle is a simple way to represent a 2 to 6 metrics over one or two dimensions. With Battle charts we can often choose the dimensions we want to pit against each other thus making them very versatile.
Best Use: To easily spot cause and affect along with give an overview of comparisons between two entities (example two countries or two schools).
A modernised version of the pie chart, a sunburst emphasises segment lengths (in percentages) in relations to each other as “parts of a whole.” Each segment can be clicked upon to see a further breakdown of data thus giving it a hierarchical coding ability.
Best Use: Best used with a single dimension (possibly hierarchic) and one metric.
PIE CHARTS –
Pie Charts are a special mention here since they are an exception in that they are the most inappropriately used by people. Even when they are used correctly there rarely is an instance where they are the best option to be used.
Best Use: As far as dashboards are concerned these charts should preferably be avoided.
This brief article is definitely not a comprehensive guide on data visualisation. Our goal here was to highlight and talk about the most common ones that we use at Captain Dash to create marketing dashboards. If you want to know more about various graphs head to the visualization lair on the Captain Dash website and play with some graphs and models to understand them better.
In closing, the biggest thing we will stress on is that simplicity is your best friend on the road to an elegant dashboard.
Stay tuned for our next post in this series, which is about the role intuition, plays in dashboard design.
Find the previous post in this series about Do’s and Dont’s of Dashboard Design here.
Color may seem an activity reserved for comparing curtain samples or choosing between pairs of shoes. In terms of data, science and research, you may think that color hardly factors into the equation. If you want to be professional you use black and white. If you want to be « fun » you use bright colors. Right?
If you have recently joined our Community and are new to the world of Business Analytics, you may have noticed the explosion in interest in Dashboard Applications (such as ours). Most people in the industry know that collecting valuable Data is paramount to measuring a company’s performance and helping you make sound decisions to meet long-term goals. But is this sufficient enough going forward? Not exactly. Continuer la lecture de Dashboards and Reporting: Separating the Traditional from the Modern