Data doesn’t tell a story. Users who interpret data do. The way you set up your analytics determines the way you will interpret your data, which should be one of the most important things you do with your website.
Here is a quick outline for reference to get you started, but keep reading to get the meat off the bone.
- What do I count as pageviews?
- What do I count as interactions with my site?
- What are things users can do that provide value?
- Business Uses Interpretation
- What sites work together to bring my users their onsite experience?
- How different is my mobile experience? Do I have an app?
- What are the most important ways for me to look at people who come to my site?
- How many business units will be viewing this data? Do I need to keep some things separate?
- Different teams have different goals even within your company.
- Different parts of your site may require users to interact differently. --!
You Need Analysis to Tell the Story
Companies have come to rely heavily on tools such as Google Analytics to measure the effectiveness of their online properties for some time now. Despite all the advances we’ve made in our ability to track, there is still one fundamental flaw in many uses of the system: People don’t plan accordingly. One essential truth I’ve found in data analysis is that data doesn’t tell a story, users who interpret data tell a story. Because of this, data analysis planning cannot simply be to install some vanilla instance on your site and try to measure what is going on.
There are three major factors to consider when setting up a plan:
- Business Uses / Interpretation
Considering these three major factors in your plan will help you implement something out of the gates that will be meaningful, and easier to report to decision makers.
Business Uses & Data Interpretation
The core questions you set out to answer are:
- How will my business use this data to understand value x?
- How will we need to interpret user behavior to understand what creates value x?
Think of all the things you ask your user to do on your site before they perform an action that provides value to your business. Do they watch a video, do they open accordions for information? Are they forced through a certain pathway, can that pathway break? There are hundreds of considerations to make when thinking about how a user will engage your site. It’s important to determine these things as early as possible because this will also impact how you interpret and report on things like time on site, bounce rate, engaging content etc.
Determining who will view the data is also important. This will tell you what views you want to create and the customizations you’ll need in each one to help analysts interpret the behaviors that are occurring. Creating these focused views will allow your team to avoid pitfalls in their data, and help them see what is most important to their efforts.
Within creating these unique views, advanced segments, content groups, and channel groupings will be essential to giving data sets in Google Analytics meaning to each analyst. The truer the channel groupings are to how you acquire traffic, the easier they are to understand. The truer custom hit metrics like bounce rate and avg. session duration are to real user interaction, the more value you get out of your measurement. Remember, it’s people that create your insights not unpersonalized data sets that sound like gobbledygook to some folks.
When you’ve concluded all of the business based things you need in place, it’s time to start tagging. I always recommend using a tag management system like Google Tag Manager, but if you’re not comfortable with that you can work with your developer.
Step 1 is to talk to the person(s) who built your app or website. Tell them all of the business based things you want to track and ask them how those things work on the site. Ask them of any nuances in the code or any opportunities to track things at scale. If you haven’t built the site yet, plan for tracking in advance. Talk with the developers about things you want to be able to track and they’ll create scalable methods for you.
Determining when to fire virtual pageviews, interaction event hits, user timings, or other customized tagging will be essential for analysts to have a more accurate understanding of how the business intends to understand user behavior. Tagging all of these facets of your site is absolutely essential to getting comprehensible data and providing your business accurate information.
Set up a tagging strategy between your marketing and development teams, and communicate. You will find that developers are not always considering what is best for future measurement, and marketing teams are not always considering how different features will be tracked.
Last but definitely not least: What do you want your app or site to do for your business? What do your teams want users to do on the site? What’s the objective of different marketing tactics? This is the final major component because it defines how you will report the value of your engagements.
Goal values are crucial for sophisticated multi-channel measurement. If you’re not thinking about multi-channel measurement, or multi-channel attribution, you need to be. Optimization is about creating the most efficient methods to acquire revenues. You can’t optimize your business, your marketing, or your work if you don’t understand the value of your goals. These goals and their values give your analysts context and qualifiers for all the other data in analytics. They are the critical element of the data interpretation.