Introduction to Analytics
SES is going strong; lots of people in the expo hall and some great info has been presented so far (continue to check the sessions we’ve covered by following the “SES SF 2012” tag on this post).
I’m excited to bring you this session about beginners analytics by Matt Bailey (@SiteLogic and @MattBaileySays). If you’re new to analytics, thenyou might be overwhelmed with what to track and how. Hopefully, some of your most-common questions will be answered in this session.
What is a visitor? A visitor is not necessarily a person. Visitors are devices. If you go to a link from a different device, then you are a different visitor.
The numbers are undependable. They are wrong.
It takes software to try and figure out who is a visitor. It depends on what software you use, too. From software to software, you will get different data.
Referrals are really good to know. Is our marketing drawing people to our site? But what we are really about is the conversion rate. Any goals you put in your analytics becomes a conversion.
People assume the information they see when they log into analytics is the most important data. It’s not. It’s generic and meant to work for every business. You must make the software work for your business.
“Cave Man” analytics generates reports like time on site, pageviews, etc. This leads to “Hamster Wheel” analytics — same numbers every month from one report to the next. When people have questions, we don’t have answers.
This is only reporting performance, not if you are meeting business goals. No context means it’s just data. Drop the stuff that takes your time and doesn’t improve your business. Don’t measure time on site if it doesn’t bring you money.
We have to get much more complicated. Analytics is about being complicated. What are the goals for the site? It should be what makes you money. Then, what makes you money not now, but will later?
When people come to your site, they have a mission. Sometimes your goals for them are not aligned with theirs. Users bounce all over the site when they try to find the information they need. You have to figure out what they want.
Ask questions — who are they, what do they want and what do they do? The more you rely on copying and pasting analytics into reports, the less questions you ask. You can transform your analytics strategy by asking:
- Where did visitors come from?
- What did they see?
- How did they react?
You must start building context. The first question starts to build this. You might have different groups of visitors, for example. You need to understand the intent behind these groups, and then measure appropriately.
This is segmentation. Segment visitors based on groups of information like search term, source or campaign. Find like denominators and create segments.
The more context, the more you can begin to tell a story. This equals buy-in, if you have to get it to do anything in your organization.
Start asking yourself these questions to build personas:
- What five types of customers come to your site?
- Why do they come to your website? Name five reasons.
In Google Analytics, think about what you want to segment people by. These are your dimensions. Use an “and” statement. It just means this keyword and this keyword is what you want to track.
The 3 Cs of analytics:
What is the same, what’s different? This is more data, more complication. But when you have this, you can draw better conclusions.
Segment acquisition (when you want to pinpoint something like best visitors): Which acquisition method provided the best visitors? Was it the link in an ad, a blog, social news, etc.?
He’s found that how people interact on the site is directly related to how they got there. For example, if you have a blog that people go to, then they might already be in favor of your content. This will cause them to interact differently.
Every site has an engagement pyramid:
Analytics is about psychology, not math. We observe behavior.
Finally, value is the cornerstone of analytics. When you don’t report value, you report activity only. This is the forgotten metric. In analytics, set up goals and set up how to track that goal. Then don’t forget to assign how much the action is worth. If you’re in the B2B space, for example, what’s the value of a lead to your organization?
Rankings reports are useless unless you are getting rankings with value. Look at keywords and the value they bring per-visit goals. These can create your actions plans. If you have long-tail keywords that have more value, you want to focus on those.
The point: everything should point back to your revenue. Marketing is about knowing what works, why and doing something about it.