1. Analytics and measurement shine light on what’s important, but both come with an electricity bill
This is one of the best ways to think of the value of measurement and analytics. They can provide light where it’s otherwise dark and give you an alternative to make guesses about what is happening in the dark.
This must be why when we have discussions about analytics for our employee portals or other solutions, I often hear: “We should track and measure everything!” I think this is great but it reminds me of an excited kid in a toy store that wants to get everything without thinking about the craziness of actually doing it.
Yet measurements come with a cost. Many–if not most –measurements have no meaning or purpose. They are going to be like that toy you throw into the basket, pay for, and then at best play with for a few minutes.
In the case of web analytics, there’s cost in implementing extra tags, quality assurance, maintaining and auditing those tags, identifying and creating the reports, and making sure the right people get those reports. In other words, there is a bill attached to all the metrics you collect and report on. Measuring everything all the time is like paying for electricity 24/7 in every room in your house, regardless of whether or not you have the lights on.
What does this mean?
Limit yourself to the metrics that will have business value and purpose to shine light on things that will impact decisions.
2. “Data-driven” over-states the role of data
I like the spirit behind the term “data-driven.” I think it’s about striving for objectivity and making decisions based on facts versus personal biases and opinions.
Aside from using “data-driven” to describe Google's self-driving car (I don't think you can get more data-driven than that), the term bugs me because people should be making in the vast majority of business and design decisions.
It doesn’t matter how much data you have or how reliable it is, data doesn’t give good answers without good questions and it's still humans who are asking the most strategic and important questions. Data also doesn't know when you should be asking “why?” or where to go to get the answer to “why?” It also can’t tell you when and how to act on what it is telling you.
What does this mean?
Your data can only be as good as the people using it.
3. Everything can be measured, but...
It is true that everything can be measured, but if you have any experience doing it, you know how hard it can be to actually do this effectively. Here are a few things to keep in mind:
There needs to be alignment on common goalsIf you’re having trouble finding good metrics, there’s a great chance everyone isn’t super clear on the goals both for your group or business as a whole, as well as the goals of particular activities.
Imagine if I think we should drive our car to go to New York and you think we should take it Los Angeles. Or I think our human resources team should focus on recruiting and you think we should spend more time on employee experience. We’re never going to get any high-value metrics if we can’t agree on the direction in which to go.
What does this mean?
If you’re doing measurement, you’re going to need a consultative process to ensure everyone is clear on common goals. If your group or business is not aligned around goals, measurement is going to be hard.
Be open to some subjectivity
Some measurements will have some hard numbers behind them. It’s not hard to see when you spent a certain amount of dollars on this online campaign and got a certain number of conversions, which resulted in a certain amount of dollars.
A colleague and I were talking about measuring the outcome of things that don’t easily translate into a dollar value or other number the other day. I asked him how you’d measure the value of a new office design and he said, “you’d ask people.”
Since so much of what you want to measure has to do with impact on people, this is what a lot of metrics come down to. Of course these aren’t perfect measures–and of course surveying people takes effort–but it can often be good enough to improve decision-making
What does this mean?
The hardest part of measurement is usually not figuring out how to measure something, but identifying what to measure and defining it in a meaningful way. Once you do that, even if it’s not perfect, there will almost always be some way to measure it.
4. You need to know your analytics tools
I’ve been surprised at the number of times I’ve seen web analytics tools implemented without anyone around who is well-versed in the capabilities of those tools. If you must make the choice, it is better to have someone that understands the needs of the business and the solution or service you are measuring without in-depth knowledge of tools over someone who knows all about the tool but nothing about the business side of things.
That person who understands the needs of the business and is responsible for reporting on measurement should start learning about whichever analytics tool you have as soon as possible. If you don’t get to know your tool, you are going to miss out on all kinds of opportunities to capture the right data and create useful insights and reports.
What does this mean?
You need to make sure your analysts know how to maximize the use of your analytics tools.
5. Done well, analytics and measurement can be the place where people come to get help to make good decisions
So, with all these challenges and costs with analytics and measurement, why should you do it?
At the risk of being a little simplistic, much of what is good in life and business is due to good decisions. By extension, bad results are often the result of bad decisions.The whole purpose of measurement and analytics is to provide information and insights that will help us make more good decisions and less bad ones.
The possibilities are truly endless here. To give just a few examples, analytics and measurement can help me decide which parts of my employee or customer portal I should invest in, which parts I should discontinue, or even who I should train to use the portal. If I’m a user experience designer, they can help me make decisions about how I structure and design specific components and the overall solution. If I’m in communications, analytics can help me make decisions about how I allocate my time on different channels.
What does this mean?
Understand the decisions people in your organization need to make and design your analytics and measurement solutions to help them make better decisions.
To learn more about creating measurement and analytics plans for your intranet, check out this video.
