Rank Checking Lives On
Every now and again you'll come across someone that says search engine rankings aren't important. What most of these people are really trying to say is that rankings make a poor measure for success. I agree that other measures such as organic traffic or conversions are more appropriate. However, I think that checking rankings regularly is still important.
Why Check Keyword Rankings?
One of the best things about rankings is that they can give you a glimpse into what is and isn't working with your SEO efforts. The trick is to check rankings deep enough so that you can see what keywords are moving up long before traffic from these keywords materializes.
Another useful thing about rankings is that they can help with determining the cause behind decreases or increases in traffic. Without additional information, assessing the health of traffic levels can be quite difficult since there may be things like seasonal factors at play. Throw in a broad increase or decrease in rankings and you've got another piece of the puzzle to help determine the root cause.
More Is Not Necessarily Better
People involved in the SEO game often over simplify rankings by showing how many keywords are up and how many are down. Comparisons may be against some baseline in the past or just the previous month. Either way, this basic analysis is flawed. Rankings should include a weighting to provide true insight into progress. A weighting can be any of the following (and probably a bunch I haven't thought of):
- By potential relative traffic i.e. using search frequency data from WordTracker or KeywordDiscovery. Yes these numbers aren't 100% accurate, but in relative terms they're good.
- By buying lifecycle phases i.e. if a keyword is transactional, it should have a greater weight than on informational one. This is difficult to measure, but maybe you could use the data from MSN Lab's Commercial Intent tool as a guide. And of course, this assumes that you want transactional traffic.
- By CPC cost i.e. the greater the CPC cost, the more “savings” there are from increased rankings of that keyword. Again, this could be a difficult measure to calculate and it does rely on a well-optized and active PPC campaign from which you can extract data for analysis.
- By number of terms i.e. a two-word phrase is likely more valuable than a five-word phrase. The fewer the terms, the greater the weight. This is similar to weighting by expected traffic since shorter phrases generally have more traffic. However, this approach may prove to be simpler.
The beauty of applying weightings is that the list of keywords can be quite long which means you can keep track of the long-tail. At the same time, the weighting keeps you from applying too much importance to low-value keywords.
Rank Checker Shortcomings
Rank checking has gone on for years and automated solutions are quite mature now. The official word from search engines is that they don't want you automatically querying their databases, but in practice if you set these rank checking programs to behave nicely you shouldn't have any problems. Sadly it seems that all of the ranking checkers are failing to keep up with the times.
I say this because they all seem to be turning a blind eye to significant developments such as Google Universal. Ideally, rank checkers would recognize any type of result and make note of the position. For example, a number 1 ranking in the “regular” results is a lot less valuable when it is preceded by 10 map links that occupy most of the page so I'd want to know there were local listings. Or wouldn't it be nice if ranking checkers could keep an eye on Google's sitelinks for you? I think so. I suspect that the first vendor to recognize these needs of the SEO industry may grab some additional market share.
And so I suspect that I'll be watching rankings for the foreseeable future with an eye to making sure they provide value. As for those that want to dump rankings, I say go ahead. I'd rather people had less information than I do 🙂
Mostly agree here. Checking rankings does help you sort through the ups and downs that may be associated with demand changes. For instance - my traffic increases 50% over three months but none of my rankings really move up or down. I then realize its b/c the site is in a seasonal business, and that time of year is when demand picks up. If you just track organic traffic you'll think your site is being better positioned when it might really just be staying even.
I agree with the unimportance of the PR. If the site ranks well in the results, the PR is neglible.
Just want to follow up on Nate's comment about PR. It is not important! What it represents is. Please remember to make this distinction... the quality of your inbound links as recognized by Google are important. The PR calculator on the toolbar is not.
You're right. More isn't necessarily better. But page rank is so important, especially if you're in a competitive market.
This is all true, plus there's one more reason -- with Personalized Search becoming more widespread, my view of a site's ranking for a given term might differ from yours just by virtue of the fact that I am logged into a different account, or querying from a different IP address.
A hybrid calculation of raw traffic combined with conversion ought to get you much closer to the truth.
Great post Marios ; I think the key point that resonates with me is that rank checking is a way to tell if your SEO investments are working. I am working with a new rank-checking site called RankAngel that is working hard to accommodate to make sure that the ranking results returned match the human experience very closely. It also allows you to set up projects that then graph your results with a specific keyword phrase(s) over time - a really useful way of communicating trends etc.
You're right about personalization throwing a wrench into the machine. However, the right ranking software would be introducing the same bias during every run and would hopefully still provide useful relative measures i.e. improvements over time even if the actual rankings differ when a user searches.
Nice post Marios. Ranking reports do get torn apart nowadays, but if you tie those rankings directly to another more important metric like sales viewing these reports can still be of benefit.
The main problem with ranking reports if that personalization has and will greatly change what you see comparable to what I see ranking in the top 10 of Google. The disparity is more evident in rankings on page 2-3-4-5.