It’s even more than looking at metrics to see where site performance excels or declines. The most important step, and perhaps the biggest hurdle, in analytics is in the implementation on existing sites. Especially on big sites. What I’m talking about specifically is goal implementation on a live site while maintaining the integrity and continuity of data.
There is a Difference
It seems simple enough; you track the funnel of each goal, URL by URL, and drop them in. Most times it’s never that simple. I’ve implemented analytics on both sites: big corporate sites and small and small-ish sites. No two are ever alike and no two ever do things the same way. There’s definitely a difference between implementing analytics on live big sites and on live small sites.
Small Site Goal Implementation Planning
With small sites, there’s a very good chance the site was never tracking anything to begin with. No analytics period. And, if they did have analytics in place, chances are the only thing they ever saw was traffic volumes because either they had never implemented tracking on goal conversions, or they had no conversion to speak of. In this instance, it’s easy. You implement the goals where there were none before and let analytics go to work.
And, even if the site was tracking goals prior, the chances of the having to track more than one or two goals (outside of creating on-click events or virtual page views) is slim. So preparation will be limited and one-to-one goal match-ups are easy to accomplish, as well as the maintaining the continuity of data on the site. At a cautious level for small sites, you may want to dump all the data as far back as it goes in case something does go wrong. In a worst case scenario, you can piece the puzzle back together.
Large Site Goal Implementation Planning
It’s best to start off this section with being honest. I’ve screwed one up in the past. But it’s that screw up that led me to write this post initially, and was the catalyst to create a game plan of tracking implementation.
Large site goal implementation planning is a different animal all together. They’re complex with lots of spinning wheels and cogs. It’s a formidable task to be sure. But there are ways you can make this project easier on yourself and make the implementation more manageable. Based my own experience (both the failure and the successes) here’s how I like to go about breaking down the implementation:
- Take at least a year’s worth of data from the site for safe keeping. With big sites, a lot things can happen that are simply beyond your control. You might not be left holding the bag should something go awry, but having the data in case something does go wrong, is the only way to be able to stitch the puzzle together.
If you’re entering the game late and you don’t have time to grab that data, I’d suggest that you dive into the analytics themselves. Reverse Goal Path is a wonderful thing to track down all the destination URLs of all the past goals. Moreover, under the assumption new profiles haven’t been created, you can set date ranges to see that past activity and make record of it.
- Make sure you’re talking with the company point-person to make sure you have ALL the goals they want tracked.
Just looking through the site isn’t enough in this case. You may not find all the trackable goals on your own, and you may not know the client wants certain events to be tracked, or you may be tracking something they don’t care about (wasting valuable goals). It sounds like common sense, but when you get into time deadlines and pre-launch mode, things get missed. And the last thing anyone of us wants to do is miss critical trackables.
- Examine all your client’s current goals.
I’ve run into a couple times where clients unknowingly have had been “double-counting” goals. Or incorrectly tracking goals. The sole purpose of doing this is to leave no surprises for the client. When you implement anew, site conversion rates change (for better or worse). What you don’t want to happen is to see the conversion rate for a goal or two plummet and have no explanation. In this way you can prepare a client for what they may be likely to see due to double-counting.
- Map out current goals with new goals and match them up
This is the mistake a made. I didn’t do this. And it’s critical. Because many larger sites will be close to exhausting all 20 goals in a profile (if not utilizing a second profile for tracking), not making a one-to-one match ups where possible will crush the integrity of the data. While you may retain the continuity of the data, it’ll be worthless. Profiles live on forever, and as such, mis-matched goals can, and likely will, wipe out the data integrity.
- Finally, discuss with your client how they want the goals implemented. Do they want new profiles or use existing profiles
It’s a big decision that can affect how reporting is done. Communication is a key here. As the SEO/SEM you need to layout the ramifications of each option. On one hand, to create new profiles, the data continuity starts from the day you create it. On the other hand, if there isn’t much match-up between old goal tracking and new goal tracking, does it make sense to make the data mushy? Every situation will be different, so this is a conversation that should be had.
Hopefully this helps you start putting a game plan together when you are about to implement new goal tracking on both small and big sites. I’d love to hear your suggestions on how you game plan your analytics implementation.