Very few of us could live without analytics. The data informs (or should inform) nearly every decision we make as SEOs, but all too often we don't use the information effectively enough, or fail to take into account enough information to put us on the right course of action.
Here are three ways that most analytics analysis can be improved, or used in ways you might not have thought of.
Many people’s first, and sometimes only, port of call within analytics is to look at search traffic and keyword traffic volumes. These are great places to start, but certainly don't tell the whole story.
Next time you log into analytics, try going to the top landing pages area instead. Date range comparisons for landing pages can be far more telling when analyzing long tail search volume changes then keyword reports.
Where you may see many keywords jump from 20-30 visits a month without giving it much notice, a landing page whose traffic goes from 2,000 to 3,000 a month is more likely to jump out at you. Once you know where you're doing something right, you can start to apply those changes throughout your site.
Landing page reports are also invaluable for diagnosing technical issues, with everything from meta no-index to analytics tag removal showing up far more visibly in landing pages then it does on keyword reports. Looking beyond keywords shouldn't ever be overlooked.
Don't Trust the Traffic Volumes
So you've been working with a client for 6 months or so, and you've been steadily optimizing the site, and seeing some results in the rankings (yes I said rankings, and this is exactly why you want to track them), so you know what you've been doing has been working, but you're not seeing the results in search. Usually in this situation the client starts to get antsy, and starts questioning the value of SEO. Don't despair though, chances are that analytics isn’t telling you the whole story.
Year-on-year (YoY) data is crucial here, as is keeping accurate records of changes made on the site. Start comparing previous year’s traffic for at least 3 months prior to you starting work on the site, and look at the patterns. You may find that where seasonality causes a drop in traffic, your work, having maintained volumes is actually a boost for the client.
On occasion, though, without YoY data available, or when seasonality doesn't seem to be the cause, you need an external data source for further information. Export your traffic volumes and Google Insights data for the same period, and compare them side by side in a chart.
Especially in this economically difficult time, what seems like a traffic flat line or drop can actually be a huge improvement compared to customer trends for the same period. Your customers may struggle to see it, but there is no arguing with fewer people searching and buying.
Custom variables are Google’s name for them, but the functionality exists in pretty much every analytics suite out there. No matter what type of site you are working on, custom variables can mean the difference between lumping all your customers into one big pile, and truly understanding user behavior.
The most commonly overlooked example is using custom variables to identify previous purchasers. Do people who have previously bought from your site enter through different keywords, do they stay longer, browse more items, are they more or less likely to buy and if they do, how much do they spend compared to first time shoppers?
This data will help you to target your resources to maximize return (if repeat shoppers spend more and are more likely to enter through a particular keyword theme, putting more effort into ranking for those terms may provide higher returns, even if the overall visitor level or search volume is lower), and also improve your customer journey (if repeat shoppers aren’t logging in or creating accounts, bingo there's a problem).
But repeat customers aren't the only group you may want to analyze with custom variables. Create a visitor level variable to track people who have ever visited from social media and you can start to assess your social media impact beyond direct sales.
Have a membership area? Start tracking how members interact with your site even when they're not logged in. Check how often people who have commented or interacted with your site visit compared to other users, and if they click more ads.
The depth of data from custom variables can make a striking change to the way you work and the way you think about your visitors, turning that amorphous blob into distinct behavioral groups.
Twitter Canada MD Kirstine Stewart to Keynote Toronto
ClickZ Live Toronto (May 14-16) is a new event addressing the rapidly changing landscape that digital marketers face. The agenda focuses on customer engagement and attaining maximum ROI through online marketing efforts across paid, owned & earned media. Register now and save!