Some More Thinking about Key Performance Indicators for Web Analytics
Web Analytics Key Performance Indicators (KPI’s) are critical for breaking through the dataglut spewing forth from your web analytics tool. I mean there’s a just a ton of data in web analytics, and the majority of it tends not to be very useful or applicable for improving your business performance. While it’s wonderful to have a tool that lets you cut, cross, and slice loads of data every which way but loose, its can be a real challenge to frame the data or put it in context in a way that helps your business optimize the web site. That’s why I like KPI’s - they identify meaningful, business-focused relationships in your analytics data. By understanding KPI drivers, setting expectations for KPI performance, and analyzing your KPI’s toward defined goals for those KPI’s, you increase understanding of data, alleviate data confusion, and provide focus for the usage of your web analytics tool.
For those of you who don’t have a KPI strategy or who are just getting into analytics, an easy way to understand a KPI is to consider the example of when you are driving somewhere and trying to get there within a certain period of time. If your goals is drive 60 miles (kilometers, my European friends) in exactly 60 minutes, you know that you need to drive 60 miles per hour (or KPH). If you go faster, you will arrive early, if you go slower you won’t meet your goal and will arrive past your deadline. So as you travel along the road, you measure the KPI of your speed. That’s what is important to measure on your trip. Of course you may measure other KPI’s like the amount of fuel left or the miles you’ve traveled… those certainly may be KPI’s you measure. But you definitely don’t need to measure you compression ratio or oil pressure even though it’s available data from your car. In the same way, when you are looking at web analytics data, you don’t want to track everything, only those things that are important to your business performance toward goals.
Several activities can assist the creation of KPI’s. Here are a few of them:
- Determine the Business Strategy. Why is the company funding and developing an online mission? What is the strategy? KPI’s can help you figure out if it’s working. To find the KPI’s that will help, the web analyst should be asking the question how can web analytics be used to formulate, implement and evaluate cross-functional decisions that will enable an organization to achieve objectives? How will web analytics be used in the process of specifying the organization’s objectives, developing policies and plans to achieve these objectives, and allocating resources to implement the policies and plans to achieve the organization’s objectives?
- Define the Site’s Goals and why the Site Exists. I covered this in a post a few months ago. A understanding of why your site exists enables you to effectively use online metrics. You need to define the purpose of your site in order to create effective KPI’s. Once you’ve defined your site’s purpose, you are positioned to examine Web data in way that helps you determine whether your site delivers on its purpose — does it exist effectively? Create your KPI’s, identify goals for your KPI’s, and track your performance against those goals.
- Recognize Value Drivers. How does the business make money on the site? Monetization, in cases where profitability is important, influences what you should be measuring. If you run a media site, you probably make money from content consumption (the recency and frequency of content consumption), conversation (social media, such as contributions or comments), and conversion (the rate at which people complete certain value driving actions, like signing up for newsletters, rss feed, webcasts, print subscriptions, or downloading certain content types, like white papers). So you create goals for and measure KPI performance around those value drivers.
- Map Organizational Roles. Classify your organization into audiences for your KPI’s based what they do on your web site. You may create KPI’s around function or action of the actors who receive your KPI reports. Function defines the group that KPI’s are focused for, such as product development or editorial. Action defines what those people do on the site to make it successful. By understanding function and action of key actors on your sites, you gain insight into the type of data needed in KPI’s and the number of different KPI reports you may need to roll out.
- Understand the Customer. KPI’s purely focused on internal function and actions are important, they need to be customer focused. If you think measuring conversion is important, while your customers tend to come to your site for informational or non-transactional purposes and then go elsewhere to convert, you may be disconnected from the reality of why your site exists. Learn customer goals from VOC (voice of customer) data and by examining historic behavioral data of key segments. Make sure you don’t create KPI’s that are vain or inane. Instead create KPI’s that help you guide action internally so that your business meets the needs of your customers.
Framing your KPI development around the five bullet points I listed above will help you create KPI’s that assist your team in guiding business performance toward goals - while not forgetting to consider some of the core elements of online business: business strategy, site performance goals, value drivers, the human organization, and the customer.
Now segment, segment, segment your KPI’s!
Judah added the following ...
Hi Stephen,
Thank you for your comment! I’m glad to hear you enjoyed my preso at Unica MIS. Your question about “extracting the ROI from KPI’s” is a smart one. While I can’t say everyone who uses KPI’s understands the impact of their changes on ROI, it’s a good idea to attempt to estimate it. That’s one of the reasons why I recommended always tying KPI’s to the site’s value drivers. That way you are positioned to understand the impact of changes to the KPI on the bottom line. For example, if you track some sort of “conversion rate” as one of your KPI’s and it goes up one month, then an annotation to your KPI report might be something like “conversion rate” has gone from 2% to 4%. The 2% increase means that we sold an additional X number of products last month. Each product was worth Y. Thus the page optimizations we made that resulted in this lift yielded Z (i.e. X * Y) in revenue.
Judah

Stephen Dantu added the following ...
Hey Judah,
Thanks for the post. We are actually in the process of aligning on KPIs for the organization and this will definitely hep me.
I had a question on measuring ROI for the analytical tool, is it a practice to extract the ROI from the KPIs?
I also was at the Unica MIS last week and enjoyed your presentation.
Thx,
Stephen Dantu