How to Develop an Analytics Dashboard

You see one every time you sit in a car: a dashboard. It’s not the most exciting part of an automobile, but it is essential for information you need, from how fast you are going to how much gas you have in your tank. You monitor it to feel confident everything is functioning as it should.

A business can be as complex as the workings of a car; and for that reason, your business should monitor operations with a dashboard. However, while a car comes with a built-in dashboard, your organization will have to build its own analytics dashboard. Selecting the right “gauges” for that dashboard can provide quick information to maintain marketing and financial goals and to take action.

Here are some steps you can follow to set up a dashboard to meet your business needs:

1. Start with an operations overview: Initial conversations with managers will reveal what they believe are essential department measurements as well as the frequency required to monitor them. The measurements, called key performance indicators, reflect business performance and are most often ratios or rates to describe performance trends. Eric Peterson wrote an e-book, The Big Book of Key Performance Indicators, that suggests KPIs for different operations.

2. Identify the way measurements are displayed and create a dashboard: Creating a dashboard in a web analytics solution can be a straightforward affair; it increases in complexity with the number of people, purposes, and marketing tools involved. The dashboard should be one page long. It summarizes the story behind the data, the needed actions, and the impact to the company if the actions are left unaddressed. Standard metrics in web analytics solutions do not match KPIs out of the box. For instance, a customer service department may not know how a website reflects its effort to reduce wait times. But reviewing visits and time on site on a web page containing customer service content can indicate how the site matches up to the department operation. The key is to display metrics with data trends that closely reflect company goals.

When establishing a dashboard, analysts and business leaders should consider the time required to pull data together. Dashboard figures may be a mesh of details from multiple sources, such as social media tools and databases. Enterprise-level analytics tools such as Webtrends and Omniture can incorporate such sources easily, but in many cases collecting data can require multiple steps and more than one person. A dashboard should also permit downloads into CSV format or a spreadsheet as well as the ability to easily create graphs from the input data.

3. Schedule dashboard reviews with marketing managers: Managers periodically reviewing the results will ensure follow-up on necessary actions. Determine who should receive automated e-mail about changes in metric performance. Ultimately the analyst provides interpretation of the data to identify the wins and lessons learned. One consideration will become clear as a team proceeds with dashboard reviews: Dashboard metrics require modification in organization strategy. Important factors at the beginning of the year can be unimportant by year’s end. Therefore, the analyst should be kept aware of company changes that can affect results.

A few books give great dashboard examples and comprehensive measurement considerations. Dennis Mortensen, an entrepreneur behind IndexTools, the solution that became Yahoo Web Analytics, includes dashboard examples in his book, Yahoo Web Analytics. Web Analytics 2.0 by Avinash Kaushik provides an overview of how a dashboard impacts a company, while Performance Marketing with Google Analytics by Caleb Whitmore, Justin Cutroni, and Sebastian Tonkin offers ideas on how analytics works with marketing media. 

Pierre DeBois is the founder of Zimana, a consultancy providing strategic analysis to small and midsize businesses that rely on Web analytics data.