Analytics news and expert advice every Thursday.
Analytics, PPC & SEO: So Happy Together
“Search is the world’s largest focus group.” A while ago, someone said this to me (I can’t remember who, but buy yourself an eclair if it’s you: you’ve earned it), and it struck me as pure brilliance. We tend to get so focused on the challenge of showing up in search results that we frequently forget that we’re bringing real people with real needs to our website, and unlike other traffic sources, these people have already told us exactly what they’re looking for. Searchers are an analyst’s dream because of this fact: in many cases, we know why they came to our site, and this helps us put their actions into some sort of context.
I decided to tap into my artistic side to illustrate a common problem in search marketing, but I have to forewarn you, this is my first comic, and I’m not much of an artist:
Google Analytics evangelist Avinash Kaushik has coined a phrase that goes something like this: “Bounce rate is when people get to your site, puke, and leave.” I love this, and I think we can build on it. For one thing, we have to ask ourselves if we’re setting ourselves up for this, well, puking. In many cases, it seems to me like we are.
SEO, in a vacuum
Search engine optimization is without a doubt one of the most effective forms of growing your audience for the average website. And it’s also incredibly complex, spanning technical issues, keyword research, keeping up with ever-changing search engines and a much much more that you can read about right here at Search Engine Land. And the goal of it? Gain rankings for key terms in the search engines.
The mostly correct assumption in most SEO campaigns I’ve seen is that by reaching the goal of higher rankings, traffic will follow. And with traffic, we get website activity. And with activity, conversions. And this is where it gets dangerous—when SEO turns into a trophy collection of rankings rather than an ROI-focused venture.
None of these assumptions are inherently wrong or bad. It’s the act of assuming that’s the problem. Web analytics offers us sufficient data to know which of these keywords are driving traffic to which landing pages, whether people stay, what they do, and whether they buy (or whatever your “conversion” is). This information should cause immediate action, but rarely does.
It seems like the big issue really boils down to the keyword research process. Most SEOs and their companies / clients will automatically focus on driving large volumes by optimizing their site for relevant keywords with the highest search frequency, when this really isn’t usually the best approach. While you may have significant volume, you can spend countless hours trying to figure out why Banana Republic doesn’t convert for the word “clothes” and never come up with an answer. But if you find out that users looking for “khaki pants” are landing on the home page and not converting, you can probably dig in and find out that there are no images of pants or links to khakis on that page. Head terms might captivate the imagination of the masses and deliver them to the site, but they’re hard to diagnose and understand, while mid- and tail terms provide more focused traffic that will help you optimize your site.
It’s going to be difficult, but learning to accept mid-tier and tail terms for SEO will offer benefits in the long run: you’ll be driving more conversion-prone and focused traffic to your site; it’ll be easier to understand user intent and alter the site to improve conversion and you’ll be building a wide foundation of SEO for when you do want to go after the sometimes soupy head terms later.
Am I saying you shouldn’t go for head terms? No. But if I had a head term converting at 1% and a dozen mid-tier terms at 5%, I’m betting I can generate more revenue, and sooner, with the mid-tier terms.
Pay per click campaigns
PPC campaigns have also proven themselves to be incredibly profitable for an array of businesses because of how easily the campaigns can be manipulated to increase performance. They have also been incredible money quagmires for some because it’s so easy to open the spending floodgates. A few weeks ago, I did a search for “facebook” and saw a paid ad for a new car. Looks like someone gave the keys to the interns.
One nice thing about PPC is that a form of analytics tool is built right into the process of managing the campaign. From one place, the campaign manager can see impression levels, clicks, average cost per click, conversions and average cost per conversion (and other things). It is immediately apparent what is working and what is not, right?
For instance: Do a thought experiment and change all of your paid search destination URLs to the “contact us” page (please don’t actually do this). Are the same things still working and not working? Why is that?
While this is an overly-obvious example, you can see that there are factors that determine performance beyond the inherent value of a keyword, and a good analytics person can help you determine what those factors are. Make sure that you’re looking at things like performance by landing page and funnels for paid search segments vs. other segments to see if your “winners” can win more and your “losers” even have the potential become winners. And remember that your ads aren’t the only things that can be changed. If a good landing page doesn’t exist or some tweaks need to be made, fix the site. Don’t feel like the site is off-limits; it plays the lead role in the performance of all marketing campaigns, and just like any lead actor it needs help with its lines every now and then.
Key: Asking what vs. why
So what we’re getting at here is looking at why, rather than what. SEOs and PPC managers can easily tell what keywords have high search volumes and should send a lot of traffic. They can later tell what keywords make you the most money. But do they know why? Knowing the what is only weakly squeezing the fruit. There is a lot more juice in there, and digging in with a good analytics package can help you extract every last (relevant) drop.
P.S. I wanted to quickly comment on my last post, Is Web Analytics Easy or Difficult?, where I said that Avinash Kaushik called web analytics “easy.” I’m afraid I may have given people the wrong impression: Avinash has never said that it’s easy, just that analytics is far more accessible than people think, and being afraid of it being “hard” is a principal reason that people never even crack the shell. So no, it’s not easy, but I think you’ll find it easier than you expect, a very reasonable approach to dig up some pretty cool and actionable ideas, even if you’re a total beginner. Good luck.
Some opinions expressed in this article may be those of a guest author and not necessarily Search Engine Land. Staff authors are listed here.