Gajus - Fotolia
The benefits of CIO dashboards and tips on how to build them
CIO dashboards can be a vital tool for assessing metrics in real time to gain insight on IT performance and support better decision-making. Here, we dive into how to build one.
Back in the last century, New York City Mayor Ed Koch was famous for asking New Yorkers, "How'm I doing?" The point of a CIO dashboard is to ensure that today's CIOs never need to ask that question.
However, it's easier said than done. A CIO's job is, by definition, to manage complexity -- every piece of software and component of an infrastructure needs to work together harmoniously to meet business goals. Yet, while every detail is important, some are essential.
With a CIO dashboard, the CIO and other business executives can assess which metrics are critical and provide insight on how IT is performing against them. It can be one of the most important tools that a CIO has in their portfolio but designing one requires careful attention to the metrics, design, tooling and data sources.
So, how should a CIO go about building a dashboard? There are three fundamental areas to focus on:
- What to measure
- How to display it
- What tools to use to generate it
1. Metrics that matter
The first step is to define and select the metrics that align with the technology strategy, which in turn aligns with the business drivers.
Specifically, CIOs should start by listing the organization's top business drivers and identifying the technology principles that enable the organization to deliver against these business drivers.
For example, one of the business drivers might be agility -- the ability to roll out new services to customers quickly and effectively.
Two technology principles that can enable agility are virtualization and migration to cloud. Metrics that matter are then the percentage of infrastructure that's virtualized and the percentage of workloads that are migrated to cloud.
Drilldowns for each of these metrics could provide additional details such as the percentage of workloads that are migrated to SaaS, IaaS and PaaS.
A challenge when selecting these metrics is the natural pressure to report on things that are easy to measure or are superficially important but not considered critical.
For instance, CIOs typically find themselves under pressure to reduce the costs of their department year over year, therefore they may want to report on "IT spend per employee" or something similar. It's a fairly easy metric to capture -- divide the CIO's budget by the number of employees -- yet it's not necessarily the right one.
Spending less on technology year over year really shouldn't be a business goal unless the company wishes to return to the days of computing via abacus. And while it makes sense to spend on technology to reduce overall costs in the company, that cost reduction may be difficult for the CIO to capture. Moving employees to a remote environment, for example, may dramatically save the company money in real estate costs even if the IT spend goes up. Yet the CIO may not be privy to those costs.
Similarly, system or network uptime and availability are easy-to-capture data points, but it's important to consider whether they merit being exposed at a high level on the dashboard. If the CIO was specifically brought in to address a chronic uptime and availability issue, or it's been a source of concern to senior management, then it possibly makes sense.In regard to projects, CIOs -- particularly ones with software development backgrounds --- are tempted to measure success based on the completing of the project. However, while it's important, the dashboard shouldn't consist solely of project metrics.In sum, good metrics are:
- aligned with technology principles and business drivers;
- based on data that the IT department has access to; and
- focused on critical issues rather than easy-to-measure metrics.
2. Design fundamentals
The next step is understanding how to display these metrics on the CIO dashboard. Many volumes have been written on how to display information, including The Visual Display of Quantitative Information, a seminal book written by Edward Tufte, who also offers workshops periodically, and Information Dashboard Design: Displaying Data for At-a-Glance Monitoring by Stephen Few.
In a nutshell, an effective information display on a CIO dashboard requires:
Simplicity. There shouldn't be design elements such as color or icons that do not transmit information.
Clarity. Information should be easy to discern.
Prioritization. The most important information should be most prominently located on the dashboard.
3. Dashboard tools and technologies
There's a wide range of software for developing dashboards, with much of it cloud-based. A plethora of free and open source dashboard tools promise low-code and no-code dashboard creation. There are also paid products such as AnyChart, Screenful and Infogram, among others. The IT team also has the option of designing its own.
There's no right answer to which tool to deploy -- for many organizations, a homegrown dashboard likely makes the most sense, if only because it's perfectly customized to the organization and IT teams typically have the expertise to write and maintain such apps. Off-the-shelf offerings have the advantage of offloading support and maintenance, if that's a concern.
More critical than the dashboarding software itself are the data sources. It's important to select sources that are credible and up to date. Many times, there isn't a single canonical data source -- in such a case, the team needs to invest some effort in developing one or integrating multiple data sources to form a canonical source.