Ask the SEO Guru

No wonder so many who call themselves SEO’s are worried about their future what with all the talk this past few months of personalized results, bounce rates and changing search parameters. There is no question the organic search results landscape is changing and behavioral metrics analysis is playing some part in that change. But, as online marketers, what does that really mean and what can we do about it?

Understanding a concept such as behavioral metrics analysis isn’t too tough. Tracking and analyzing which people do what, when and how often to try to determine their future response to a specific problem or opportunity. For our purposes that simply means which ad they are most likely to respond to when based on past historical behavior.

BUT, that’s the concept as it’s spoken of in SEO circles.  What is the academic definition? Well here is how Wikipedia defines it:

Search results

From Wikipedia, the free encyclopedia

You searched for behavioral metrics analysis (all pages starting with "behavioral metrics analysis" | all pages that link to "behavioral metrics analysis")

Jump to: navigation, search

No article title matches

http://en.wikipedia.org/wiki/Special:Search?search=behavioral+metrics+analysis&go=Go

HMMM, maybe this will help clear it up:

On Behavioral Metric for Probabilistic Systems: Definition and Approximation Algorithm
Taolue Chen; Tingting Han; Jian Lu
Fuzzy Systems and Knowledge Discovery, 2007. FSKD 2007. Fourth International Conference on
Volume 2, Issue , 24-27 Aug. 2007 Page(s):21 - 25
Digital Object Identifier   10.1109/FSKD.2007.426
Summary:In this paper, we consider the behavioral pseudometrics for probabilistic systems. The model we are interested in is probabilistic automata, which are based on state transition systems and make a clear distinction between probabilistic and nondeterministic choices. The pseudometrics are defined as the greatest fixpoint of a monotonic functional on the complete lattice of state metrics. A distinguished characteristic of this pseudometric lies in that it does not discount the future, which addresses some algorithmic challenges to compute the distance of two states in the model. We solve this problem by providing an approximation algorithm: up to any desired degree of accuracy epsiv, the distance can be approximated to within epsiv in time exponential in the size of the model and logarithmic in 1/epsiv. A key ingredient of our algorithm is to express a pseudometric being a post-fixpoint as the elementary sentence over real closed fields, which allows us to exploit Tar ski's decision procedure, together with binary search to approximate the behavioral distance.

Or maybe not.

I first recall hearing about behavioral metric analysis a long time ago and if memory serves it was an article published by NASA talking about trying to design space capsules in such a way as to be able to predict the humans reaction to specific events that were identical to, or at least resembled, previous events. Hence the probabilistic bit.

While I have no references to list, I’m pretty sure that most of the research and application of the concept is not in the field of space exploration but rather advertising and marketing.

There can be as many metrics I suppose as there are people wanting to track them, especially if you start blurring the lines of behavioral metrics with demographics. Again, a simple concept such as a person who did this before will likely do it again, becomes very complex when attempting to digitalize it and then un-digitalize it into something a human could take action upon.

But as marketers, usually with limited time and resources as in the case of small online promotion firms, we need to focus on only the ones that increase our convertible traffic and/or make us more money. So from there it comes down to:

#1. What can we get within the boundaries of what can we afford?

#2. What of what we can get can we do something about within the boundaries of what we can afford?

There are certainly times when knowing things like the time of the visit , geo-location, IP address, search history, where a prospect spends the majority of their time, who do they talk to about what and how often can come in pretty handy in making a feature, benefit, call to action presentation. It’s not difficult to see how any one of those things or any combination of those things could influence how you would display your presentation in the hopes of getting the most desired responses.

Then comes the demographics which some would argue are even more important than the behavioral metrics. Where they live, how much money they make, what religion they are, what political affiliations they have, education level, what kind of car they drive, how much credit they have etc, etc,  etc.

BUT LET’S GET REAL:

As close as it appears anyone can tell, there are approx. 273 pages per domain on average

http://www.boutell.com/newfaq/misc/sizeofweb.html

Now regardless of how many pages YOU have, take a quick look at how many unique referrers you had this past week.

I picked one of ours at random and checked. That domain last week had 1438 unique referrers from 970 unique IP’s. That particular domain has 114 pages. So that gives an average of 12.6 unique referrers per page within one week.

SOOOOO, if the average unique referrers is 12.6  per page and the average domain has 273 pages, that would be 3440 unique referrers to try to figure out what to say to get the maximum conversions for each referrer. Now if you try to analyze just where the referrer came from geographically and it only took you 4 seconds to do that, you’d be looking at 13,760 seconds or 229.3 hours. Considering Mother Nature only allows each of us a total of 168 hours per week, that wouldn’t leave much time for eating, sleeping or bumping uglies with the old life partner now would it? Even foregoing all those creature comforts, you’d still be getting behind more  than 60 hours a week. YIKES!

My point is simply that if you lack the bandwidth, technical ability and financial wherewithal to automate the entire analytical process based on selected criteria, then pay someone,(even if it’s you), to read the reports and THEN modify pages to enable you to split test a vast variety of displayed content, you ain’t gonna make it brother!

If you do lack any of those things, then join the club. Pretty much all of us that have not gone public lack at least some of those resources. So what are our options?

#1. Throw as much bullshit content and bullshit links at as many pages as our auto generator will belch out and be satisfied with a .02% conversion ratio, (which in this guru’s opinion is the major problem with the web today. The old, “if one is good, then a million must be better philosophy).

#2. Be satisfied with building a web business at the speed of height, never really knowing what to change when or where, but always believing that if you just keep adding quality, original content,(whatever the hell that is), somehow you will magically win the day and eventually get financially rewarded for all your hard work and devotion,(hahahahahaha. Sometimes I crack myself up).

#3. Eliminate the noise. Forget about all the stuff that consumes too much time or too much money and pinpoint what you can get fast and cheap and then focus on that to double your conversions. Once you have doubled your conversion ratio, THEN figure out your margin and go get more of the same traffic. 

If you go for #3, all you really have to do is 2 simple things. Know what the referrer was, (the search term or where they came from such as a forum post, a blog, a newsletter, email, etc.), and what language they speak, (and even that is not as important as the referrers). Then show them what they expected to see when they got to your page.

The 3 Second Back Button Boogie

That’s it. I call it, eliminating the 3 second back button boogie.

Just by following this simple process, I have increased specific pages conversions from single digits to over 70% !

All you have to do is track your referrers so you know what your prospect was looking for and then show them that SPECIFIC content within 3 seconds of them hitting your page. That alone can increase your conversions dramatically.

I’m not suggesting cloaking. I’m suggesting simple ad delivery or content management. The same thing major sites do when they know what IP you are coming from. Besides, when it comes to me running a legal business for my benefit and the benefit of my clients, guck foogle and the horse they rode in on!

**************************************************************************

>disclaimer<

Guck foogle is a fictional phrase devised in the twisted mind of a fictional character and any association to any past or present actual persons or entities either living or dead, is unfortunate.

****************************************************************************************************

In my next post, I intend to discuss exactly what the 3 second back button boogie is and how to avoid it using a unique philosophy and a cool little tool I’m going to soon release as a free open source project.

Peach Y’all

Da Guru

Some pretty cool reference reading

Sphinn Discussion

If you kids don’t quit jumpin on that bed I’m gonna git my belt and come in there and give you something to jump about!