Ed Schipul » Search Engine Advertising

Google “God in Houston” and you won’t find a church

eschipul — Tue, 01 Jun 2010 19:55:25 +0000

In the process of explaining SEO (search engine optimization) over the years I frequently demonstrate that if you Google “God in Houston” the top results are not churches. Now I’m not talking about the local results that show the churches, but the actual search results below that that lists KSBJ as the top result for “God in Houston” when searched on Google. And the only paid search result is for “Houston Gold” – like the shiny stuff you make jewelry out of. Here is a screen shot:

From a technical perspective, this makes perfect sense. Because the largest churches in Houston do not mention the word “God” on their web sites. Yes really. Using a search engine keyword analyzer, a test of second.org shows the following.

Note the title is “Second Baptist Church, Houston, TX.” Thus they will likely rank for “Churches in Houston” but not for “God in Houston.” A simple fix would be to update the title to “Second Baptist Church, Serving God in Houston TX”.

I mention this because exactness of speech matters. It means that some of our largest churches have zero (0) possibility of being returned if a downtrodden person googles for them in the middle of the night. It means missed connections when a bible study group at a particular church might be the perfect connection for a fellow Houstonian. But we will never know because of a failure of exactness of speech.

On the flip side, a tip of the hat to Braeswood Assembly of God church which comes up for both the physical location and second natural ranking after KSBJ in the search results. And all because they mention the word “God” in their title.

So be specific. Be exact. And I’ll leave it to you to search for the ministers’ names – they rank a bit higher than God.

SEM Presentation from Barcamp Houston 2007

eschipul — Sat, 25 Aug 2007 21:43:02 +0000

Met Steven Evatt today at Barcamp Houston. Steven does SEO with the Houston Chronicle. This is our presentation on search engine marketing at the HTC Houston Barcamp.

Mahalo Greenhouse – user generated search pages

eschipul — Thu, 14 Jun 2007 11:21:38 +0000

For the search engine advertising folks in the house, put Mahalo on your radar. It is a manually created directory calling itself a search engine. They are now trying to be more inclusive as the top 10k search results really is a finite amount.

My take? Maybe mahalo will get DMOZ right as DMOZ has fallen into a group of non-responsive fiefdoms.

Viral Marketing Described as Link-Bait

eschipul — Fri, 11 Aug 2006 14:37:37 +0000

While a few of our team attended Search Engine Strategies in San Jose this week, I did not. But I have been reading the newsletters that come out on the sessions.

On session that is worth review is from SEOmoz blog on a formula for creating compelling content by Randfish. Content so compelling that others link to it and therefore help your search engine rank. It is told from the perspective of the web yet it should work off line just as well from a PR perspective.

Here is the link-bait presentation with a summary below (go visit the presentation for the REAL content):

Researching a Sector’s Link-Worthiness
Discovery of ‘Big’ Players in Your Field – Technorati and Google Trends with your keywords
Targeting YDDS (Yahoo!, Digg, Del.icio.us, Slashdot.org) (maybe also netscape.com)
Targeting Offline Media – hire PR expert
Selecting a Content Focus
Melding Branding and Viral Elements
Targeting Keywords/Search Traffic
Look at Examples of ‘Brilliant Ideas’
The Value of a ‘Web 2.0′ Look & Feel – "The right "look and feel" will earn links"
Elements that Encourage Linking
Pre-Launch Public Relations
Managing Launch Traffic – do not respond negatively
Continuing to Get Value from Linkbait – "Update the content with timely information."

Squidoo and SEO Spam- please no

eschipul — Wed, 02 Aug 2006 04:27:45 +0000

From this long tail post on Seth’s blog, it looks like SEO black hat bait to post lenses on squidoo. Please tell me it isn’t so? I have some friends headed to SES San Jose so I’ll post what I learn when they get back. Sigh.

For the non-serious geeks, squidoo lets people build a list of sites to be searched as a group, by topic. So you can add 10 sites related to ferrets and search like google, but only within your select sites. And of course you can share it. Seth’s blog post is pointing out that the lens itself is scoring well in other search engines, which creates an incentive for folks to cheat in the search engines using custom lenses. Just something to watch out for as I am sure google is already doing.

Perhaps there is a strategy application for this within public relations for the enlightened? But that is another post.

Microsoft AdCenter Gender Predictions for Demographic Marketing

eschipul — Wed, 28 Jun 2006 16:28:22 +0000

The demos of demographic profiling of search on adlab site of adcenter search technology. The image to the left demonstrates the end result; you get gender approximations based on search term. These are estimated from the behavior of the user I am guessing from total search patterns.

So, hypothetically if a searcher is obsessed with someone perhaps this indicates a male behavior. That profile data is tied back to search terms and presented to the marketing team as such. Interesting.

All kidding aside, I do see real value in demographic data when formulating strategy for clients. This does matter. Note that Google only released google trends after it was obvious that technorati and blogpulse visual reporting was skyrocketing in use. Perhaps this will lead to a similar disclosure from adwords or overture in the future?

Privacy? That is another matter.

Cost-Per-Action Advertising for Survey Promotion?

eschipul — Thu, 22 Jun 2006 16:15:59 +0000

I am reading a book on conducting surveys, so surveys are on the brain. Talking to Susan Saurage at the AAF conference last week may have contributed to my current mind set. So when I saw this post:

Google Launches ValueClick Killer
The Google AdSense team would like to invite you to test a feature that provides you with a new way to earn revenue from your website by hosting ads that are compensated based on a Cost-Per-Action [CPA] basis.

My first thought was – sure you could use that for affiliate marketing, but Cost-Per-Action would also be a great low cost way to drive completed surveys. I like the microwork concept from mturk but that is such a small subset of people. PPC for surveys is problematic because most people do NOT want to take a survey when seeking something. They want to find something at that point.
But I believe an image marketing type ad soliciting survey completion without any direct compensation to the respondent might produce a balanced result. File this under "to be tested."

The picture? That has nothing to do with this post. It is just the most ridiculous advertising I have seen lately. And a terrible brand name; how can you differentiate "nutrition on sale"? Uuuugh. Maybe they need some cost per action advertising.

HitTail – Web Analytics from a PR Firm?

eschipul — Fri, 16 Jun 2006 05:21:27 +0000

Via Marketing Shift a PR agency has created an SEO “long tail” optimization tool. From the Marketing Shift post:

A free program organizes “long tail” data about how people found your
site through natural search and suggests keywords to use for search
engine optimization.

and

PR Firm Connors Communications, which wrote the MyLongTail software (now located at HitTail.com),
is also touting the merits of blogging and public relations to create
content that is effective in garnering natural search instead of over
paying for paid search.

The HitTail site has an 8 minute demo (sorry, 8 minutes is a LOOOONG TIME!). From viewing the demo, it looks like Google Analytics with a clear visualization of more than your top 25 inbound search words.

UPDATE: Please see the comments section on the latest from HitTail!

Technically you can get this from any analytics package. Just look at
your keyword search terms and change it from “show 25″ to “show 100″.
In DeepMetrix you can export ALL of your referring keywords for a given site into Excel.

Yet that misses the point. Yes look at your keyword density calculations for your content, but only AFTER you have written the content. Add value first by writing good content. Don’t
look for long tail obscure keywords to target. At least not in my
opinion. This is just spam blogging with a PR person in between. Maybe
not that bad, but I find putting long tail terms before the content
akin to putting the cart before the horse.

Voter Fatigue Skews Search Engine Results for the Aristrocracy

eschipul — Tue, 16 May 2006 20:03:34 +0000

Voter fatigue, in political science parlance, refers to people literally getting tired of voting. Of if they vote, perhaps they vote the major ticket but block vote the rest because who the heck knows there the 2nd District Court is anyway, right?

David Berkowitz in today’s SearchInsider argues this phenomenon is coming to search. From the MediaInsider article "The Aristrocracy of Relevance" (reg required unfortunately)

the search engines and their hundreds of
millions of users will soon find themselves subjugating to a new ruling
class: the Aristocracy of Relevance. The whole search engine marketing
business was much easier when the search engines were the arbiters of
relevance.

and it continues

Now, we have tagging. With the engines’ new services, anyone can label
a search result as he or she sees fit, and the search engine recognizes
the categorization (within reason). The old model might be viewed as a
benevolent dictatorship or even as enlightened absolutism; the new
model is a democracy–power to the people. Yet most people won’t take
advantage of this power.

He goes on:

Considering that My Web and Co-Op will remain the playground of early
adopters for the foreseeable future, and only users who opt in to those
services will find their listings affected by other users, then only
the early adopters will notice the impact. Or, to put it into the
context at hand, an aristocracy is being created, but the aristocrats
are talking to themselves. They have their social club, and no one else
will notice. Only when the aristocrats’ recommendations take prominence
over the search engines’ own algorithmic rankings for all users, logged
in or not, will the Aristocracy of Relevance gain power.

Even if the Aristocracy gains power, it may wind up benefiting everyone.

Who? Like the French? Like the Dean Campaign?

OK, that seems harsh. And I really don’t know how it will play out. I do know when the Internet was young I would tell people about this cool thing called Prodigy and then Compuserve. And they would say "oh" but do nothing." Then I rambled on about email and they would say "oh" and most did nothing, only to sign up six months later and say "why didn’t you tell me about this??? And can you come configure my new modem?"

Now I ask if people are using Bloglines and RSS and I get blank stares or the particularly aggravating "I don’t have time." (huh? isn’t that the #$*(&# point? you don’t have TIME NOT TO!)

Maybe this is the flip side. Alpha geeks get to dictate to the majority?

Yet it still seems dangerous. I’m not on board yet and I think it is very important to get social software down to regular users to balance the Dean affect. I think distributed authoring is important.

Google and Accoona for “Perception of Public Relations” – Accoona no match!

eschipul — Tue, 28 Mar 2006 16:48:24 +0000

I received an email that made me curious about the difference between Accoona (which I had not
heard of) and Google (ya, someone told me about google once) on understanding the context of a search. From the Accoona site:

Accoona Artificial Intelligence is a Search Technology that understands
the meaning of search queries beyond the conventional method of
matching keywords.

and

Accoona’s Artificial Intelligence uses the meaning of words to get you better searches.

The email was from a PR agency so of course I searched on "perception of public relations" in both search engines.

Everyone loves an underdog, particularly an underdog victory. But alas, such is not the case with Accoona. Google pretty much trounced them in my very unscientific comparison. This is one search, one test, yada yada and I look forward to checking back with Accoona.

One thing Accoona can do immediately to improve results is kill all of the advertising at the top. Obviously the ad engine is not as advanced as the search engine part. As a user advertising at the top IS search results. Right hand ads are extra stuff to me. But note that even without the top 3 ad served results, google still produced more of what I was specifically looking for.

Perhaps pay some turkers to evaluate a larger quantity of results, hopefully in a double blind test? But I don’t think PR alone, perception or not, is going to raise Accoona above Google in my perception any time soon.