Getting The Credit Your Analytics Efforts Deserve
The population of search marketers out there (you’re a lovely people) relies heavily on analytics tools to capture and report on success, and most of the time, there isn’t a safety net with your analytics tool—you pretty much have to capture the data the right way the first time around. You have to measure paid search campaigns, keep that information separate from natural search data, figure out how to attribute valuable actions like newsletter signups and purchases to each effort, and of course try to keep track of all sorts of minutiae that will allow you to dig deeper and analyze on-site behavior when you find the time. And this isn’t always so easy.
If you’ve gone to a sports bar for lunch lately, you’ve undoubtedly read some pretty funny closed captioning on the live news or sports channels running in the background. I can just imagine these poor guys typing at a frenzied pace, trying feverishly to keep up with everything being said. Your web analytics people aren’t so different when they’re trying to record the whole show unfolding on your site, and at times you probably wonder if they caught everything.
What causes issues?
Analytics implementations are often slapped together pretty hastily. One of the things that your analytics implementation person can struggle with is keeping track of new content and making sure that analytics is properly set up on new pages. Web analytics people aren’t usually as involved as they should be in the content creation process, so when they’re asked to make sure “it all works,” they often have to scramble to get the basics in place, hoping they will have time to set things up the way they really should be. And if you’re a paid search marketer creating great landing pages or an SEO working on developing a whole buffet of new content, please get the analytics guys involved early. Who knows? They may even have some good ideas or be able to shine some light on history that will help you out.
I’ve probably worked on about 100+ implementations of both Omniture SiteCatalyst and Google Analytics, and I have literally seen only one single, solitary good implementation… ever. And, unfortunately, I usually find out that there were a number of real-world issues that lead to that point (it would actually be reassuring to know that the implementer just didn’t know what they were doing; it’s always much harder to know that competent people couldn’t succeed because of crummy circumstances). So, let’s try to do what we can to help the analytics peeps out, because at the end of the day (I still really hate it when people say that), we really rely on them to illuminate the effect of this PPC and SEO work.
Beyond creating new content, you may be implementing redirects to support an SEO effort and improve your URLs or structure (or creating vanity URLs for other campaigns, etc.), and there is a lot of confusion around how analytics tools will react to these changes. I wrote a comprehensive guide to how Google Analytics and Omniture SiteCatalyst treat 301 (and 302) redirects, so if that’s something you’ve been trying to work out, please go take a look and fire off any questions you may have.
Redirects are especially touchy for a few reasons, two of which are pretty important:
- They can cause issues with traffic sources: if your paid campaign parameters aren’t retained post-redirect, your referral data may have some major issues. For one thing, your paid search traffic may be getting attributed to natural search! Yikes. Read that other post for more details.
- If you have 302 redirects in place, or 301s that have not yet been digested by the engines (the engines may still link to the old pages; new ones are not yet indexed/in SERPs), you may be really confused to see some landing pages for some keywords. If you are comparing the URLs in the SERPs (or a ranking report) to the landing page URLs you’re seeing in your analytics tool, these babies are not going to line up, and may give you a false impression of which pages are generating traffic for which keywords.
Some of the documentation from Omniture and Google is a little misleading or outdated when it comes to redirects, so be careful. Google has one misleading help page, while Omniture has one great page and two very misleading, contradictory pages—I can’t link them because you’ll need credentials to view them. Unfortunately, this confuses a lot of people. If you are an Omniture client and want to discuss it in more detail, send me a reply on twitter (@evanlapointe) and we’ll see if we can get things figured out (and read my other post first for some tips that might help you diagnose). Also, Avinash Kaushik covers these issues very well in his books, and the dudes at Omniture who are on twitter seem to be super sharp guys if you want to talk to them directly. Hopefully, they’ll find this post and put their info in the comments.
Some things you can do to make sure you’re getting the credit you deserve:
Put your analytics person on speed dial. It’s going to be helpful for you to know who is responsible for analytics implementation in your organization, and if you have a huge site, who else might be responsible for implementation in other site sections, subdomains, third-party shopping carts, sites recorded under the same suite/profile, etc. Talk through some of the issues you’ve seen with tracking in the past. And try to get friendly enough to where this person calls you once a week or so to give you some good news or ammunition for your boss or next meeting. It’ll be worth it.
If your analytics tool allows them, use annotations as much as humanly possible. Mark dates that redirects go up. Mark dates when products or content go live. Mark anything you think will ever be remotely important, because when your data spikes, everyone’s going to go running around to figure out why. This will save everyone a lot of time and energy.
Collaborate. Sit down with the usability, design, copy and IT people, along with analytics (or if you’re responsible for all of these things for your company, have a nice cup of coffee and talk to yourself for a while) and discuss what you really want to accomplish past simple traffic generation. What do you want people to do first? Then what? And then? Get micro. Talk about how you can all learn and iterate based on what you learn. Get on the same page and who knows what’ll happen.
Some opinions expressed in this article may be those of a guest author and not necessarily Search Engine Land. Staff authors are listed here.
(Some images used under license from Shutterstock.com.)
Analytics news and expert advice every Thursday.