Many of us are predisposed to believe good news. When you’re doing SEO, believing too quickly can, at the very least, lead to embarrassment, which is why I decided a long time ago to question even good results. In many cases, all it takes is waiting for a while to see if an improved ranking or a spike in traffic is an anomaly or a new trend. Patience will protect you in many situations, but what if a change is longer lasting and you’re due to provide a status report?
The first thing I like to do is look for data that verifies what I’m seeing. For example, if there’s an increase in traffic, there’s a possibility that there’s a corresponding increase in rankings. While not always foolproof, if I see improved rankings I feel more confident that the increase in traffic is SEO related. The problem is that rankings don’t always improve in a way that’s detectable, e.g., a small improvement in a lot of long-tail terms can drive measurable traffic increases while the ranking changes themselves are difficult to spot.
The next thing I look at is what my counterparts on the paid side are doing. An unexplained change in organic traffic can be due to a problem on the paid search (either someone has made a mistake or some tracking software has “misbehaved”). One challenge with paid search is that every ad URL has to be properly tagged or the traffic coming through that URL could be improperly attributed to organic. And because the paid search folks can get most of their data from programs like Google AdWords, they often don’t know that a mistake has “broken” the tracking in other packages like Google Analytics.
The good news is that finding such an error is pretty easy. Just switch over to the paid search reports (in Google Analytics you get here by clicking on Traffic Sources -> Sources -> Search -> Paid) and look for unusual changes. For example, if organic traffic went up because of a paid search tracking issue, then paid search traffic should show a corresponding decline since a single visit won’t be attributed to both sources.
Figure 1: A big drop in paid search traffic could mean there’s a tracking problem
Sometimes the problem isn’t as obvious as the above so you need to dig a little more. Often, the next data I’ll look at is the organic keywords driving traffic. If I see any traffic-driving head terms that I know the site doesn’t rank well for, then this is likely incorrectly attributed paid search traffic. You’ll need to do a little more digging of course since it’s conceivable that there’s actually been an increase in search volume for these keywords and you’re just benefiting from this increase.
There is one other thing I check that does more often than not point to a tracking issue though. For this I examine the landing page URLs (in Google Analytics go to Traffic Sources -> Sources -> Search -> Organic -> Landing Pages). Once in this report, apply a search filter to look for a question mark in landing page URLs. If you’ve mostly got search engine-friendly URLs, then you should be able to pick out URLs that include paid search tracking parameters that, for whatever reason, have been classified as organic. If you do see such URLs, head over to Google and search for the kinds of URLs you’re seeing to make sure they’re not indexed. If the URLs aren’t indexed, but they are receiving traffic, you’ve got some solid support to request that the paid search folks kick off an investigation.
There are a lot of moving parts in the search space and good news can easily be the result of bad news. Don’t inadvertently undermine your credibility by jumping the gun, but rather look for an explanation for the good news. There really is no downside to a short delay since if your investigation validates the good news you’ll still have the good news to share.
Video content has skyrocketed over the past few years, and therefore it’s time to examine how adding SEO to your videos can impact rankings.
SEO and search marketing are a vital part of any marketing strategy, linking together channels like social media, content marketing and offline advertising.