Is there a better way, and can we imagine life beyond rankings? I’m going to argue that rankings still have an important role, but I think we also have to broaden our view and consider more diverse metrics, including search traffic, aggregate rankings, and traffic diversity.
The Problem with Metrics
When it comes to analytics, we have a very bad habit – we accept the status quo with total devotion until someone finds a better metric, which then becomes the new status quo, shaming and deriding all who dare to hang on to the old metric. The history of web analytics looks something like this, if I’m over-simplifying:
We moved (roughly) from hits to pageviews to visitors to unique visitors to conversion metrics, discarding the old metrics with each new step. Have our analytics improved? Absolutely, but each improvement doesn’t invalidate the steps that came before it. Knowing how many unique visitors you received is great, but you still need to know how many pages they viewed and how long they spent on your site. Tracking leads/conversions is excellent, but how big was the pool those conversions came from? Are you failing to convert, or are you just not getting enough people to your site?
So, how do we elevate our approach without looking down on the metrics “below” us? Here are four groups of metrics I think you need to be aware of, to both move beyond rankings and, at the same time, learn to appreciate them even more.
(1) Overall Search Traffic
What’s the end result of ranking? Hopefully, it’s organic search traffic to your site. So, let’s start there. The best part of overall search traffic is that it basically combines three other metrics:
Again, forgive some over-simplification – it’s not really additive, but the total traffic picture factors in keyword volume, ranking, and even click-through rate (CTR). Overall traffic can’t paint the whole picture (see my first point), but it does have some clear benefits:
- Traffic is actual, whereas ranking is potential
- Overall traffic is easy to report and understand
- You can quickly identify trends over time
I still believe that all three of those pieces above (volume, ranking, CTR) do matter individually, but overall traffic is a great top-down view of your SEO success, and it’s oddly overlooked.
(2) Aggregate Rankings
One of the tricks with rankings is that any one keyword may only be a tiny piece of the puzzle, and that individual ranking will naturally fluctuate even on a good day. So, why not look at rankings over a group of keywords? AJ Kohn did an excellent post on this idea – what he called “rank indexes”.
The math takes some work, and there are many ways to create and calculate a rank index, but the concept is simple enough. Think of it like a stock index – instead of tracking individual keywords, you create a group of similar keywords and track how they collectively rise and fall. This could be an average ranking, or a weighted average, or you could even transform it into a dollar amount or 1-100 scale.
What I love about rank indexes is that they use large numbers to overcome the unavoidable noise in individual rankings. They’re also great for communicating to a C-level audience. It’s easy to create themed rank indexes (think of them like sector funds). Here are a few ideas:
- Branded keywords vs. Non-branded
- Head keywords vs. Long-tail
- Intent (Navigational, Transactional, Informational)
Better yet, create more than one rank index. Each of them can present a quick but unique view of your overall performance and could be relevant to different teams or executives.
(3) Keyword Diversity
Here’s a simple question we don’t ask often enough in online marketing – how many keywords drove traffic to our site? While ranking is a measure of SEO effectiveness, keyword diversity is a measure of SEO risk.
Consider two major Google algorithm updates over the past two years. The first is Panda, which still confuses a lot of people. Much of what Panda hit was so-called “thin” content, which naturally means those sites lost long-tail traffic. In many cases, those sites lost not only overall traffic, but keyword diversity. By looking at which keywords were lost (or which types of keywords), you have a much clearer path to recovery.
Another recent update was the Exact-Match Domain (EMD) update, and the gradual decline of low-quality EMDs in general. If you rely too heavily on an EMD to rank for a few “head” keywords, keyword diversity (or lack thereof) can help you assess your risk. It’s easy to look at huge search volume and pat yourself on the back, but if that volume is only coming from a dozen keyword variations, you could lose a lot from a relatively small algorithm change.
(4) Landing Page Diversity
A healthy site with diverse traffic sources should naturally draw search traffic to a variety of pages. In the past, I wouldn’t have thought to track landing pages generating traffic separately from keywords, but then along came this:
Yes, that’s a real site, where (not provided) accounts for almost 67% (2/3) of all keywords reported in Google Analytics (not my definition of “single digits”, but that’s another blog post). You know what Google’s change didn’t impact, though – your ability to measure which pages that traffic landed on.
Rachael Gerson has an excellent post on using landing pages to partially reclaim (not provided) - it’s not a simple procedure, but if your numbers look anything like my screenshot, it’s worth doing. The best part is that landing pages can be tracked by virtually any type of analytics software, and so it’s unlikely Google or Bing could take away this data.
Why Rankings Still Matter
With all of these options, why should we still track rankings? One answer is that our clients expect it, right or wrong – rankings have what people who study measurement call “face validity”. In other words, they make a certain sense at face value. When we type a search into Google, we believe what we see.
Obviously, there are problems with that perception, but it’s still a reality we face as an industry. There’s a better reason, though, and one I don’t see going away for quite a while. Let’s revisit our traffic “equation”, but this time without rankings:
What would happen if you suffered a serious organic search traffic loss? In other words, how would you diagnose that loss? Keep in mind that potential volume is an estimate at best (if you use Google AdWords data, like most of the industry) and CTR is little more than an educated guess for organic search. In other words, these are important pieces of the theoretical puzzle, but we’re lousy at measuring them in real life.
So, what’s left? In a word: rankings. When your organic traffic shifts, whether it’s up or down, rankings tell you where the shift happened. You can try to use traffic for individual keywords, but the (not provided) problem is making drawing conclusions from any single keyword much more difficult, especially over a span of weeks or months. If you know rankings fell hard for a single keyword or group of keywords, you’ve got a shot at solving the problem. If all you can do is walk in and tell your boss that traffic fell, you’ve got a shot at unemployment.
Where Do We Go from Here?
We can sit around complaining about the loss of data and let Google corral us into a narrower set of metrics, or we can get creative and become better marketers. You don’t have to put all of these ideas into play at all at once, but I challenge you to pick one and implement it over the next six months. You’ll broaden your perspective, and you may just find yourself with new weapons and a competitive advantage when the next tools shake-up or (not provided) comes along. There is life beyond rankings, but first you have to stop playing dead.
Dr. Peter Meyers is a cognitive psychologist and resident Marketing Scientist at SEOmoz. His latest obsession is hunting the algorithm to find out what makes Google tick.