I had heard about the Jeopardy Watson challenge before, but it didn’t sink in until I saw a recent post by Joe Hall and his point about the new ways of information retrieval that are upon us.

To gain a little insight into what is in store for humanity, check out the IBM YouTube post:

Click to View Watson Trailer

The Watson Jeopardy episode airs February 14th.

IBM’s Watson brings to mind a science fiction story by Terry Bisson:

from THEY’RE MADE OUT OF MEAT

“They’re made out of meat.”

“Meat?”

“Meat. They’re made out of meat.”

“Meat?”

“There’s no doubt about it. We picked up several from different parts of the planet, took them aboard our recon vessels, and probed them all the way through. They’re completely meat.”

“That’s impossible. What about the radio signals? The messages to the stars?”

“They use the radio waves to talk, but the signals don’t come from them. The signals come from machines.”

“So who made the machines? That’s who we want to contact.”

“They made the machines. That’s what I’m trying to tell you. Meat made the machines.”

“That’s ridiculous. How can meat make a machine? You’re asking me to believe in sentient meat?”

While watching The Family Guy, I had often wondered where Stewie Griffin’s catch phrase “What the Deuce?”  came from.

Stewie will often say his catch phrase at times of astonishment.

Turns out the “What the Deuce?” was once a euphemism for “What the Devil?”.

-

Using Google’s new Ngram viewer, we can see that the catch phrase began falling out of favor soon after the start of the 1920′s.

-

So what happened?

Why did the phrase “What the Deuce?” fall out of favor?

-

-

It may have been the loosening of restraint for certain topics or it may have been the overuse of the old phrase “What the Deuce?” which blunted the phrase’s emotional edge.

I can give no definite cause and effect.

Out with the old and in with the new.

Many approaches with pay-per-click advertising used today were first applied towards the beginning of the last century by advertising pioneer, Claude C. Hopkins.

In 1923, he published the classic “Scientific Advertising” emphasizing testing and measuring ad campaigns.

Below are some excerpts from his 1927 follow-up book, “My Life in Advertising“.

On Potential Customers:

People are seeking happiness, safety, beauty and contentment. Then show them the way.

All are seeking advantages, improvements, new ways to satisfy desires.

On Calls to Action:

People are dilatory. They defer action, then forget. Many an advertiser loses in that way most of his half-made converts. One cannot afford that.

Aim to get action.

Limited offers of every sort. Something to induce prompt action, to avoid procrastination, is always an important factor.

On Ad Headlines:

Another thing is to learn exactly what sort of headline most appeals. Again and again I multiplied results from an ad by eight or ten by a simple change in headline.

On Ad Content:

Give actual figures, state definite facts. Indefinite claims leave indefinite impressions, and most of them are weak

I have seen many an ad killed by a single unfortunate phrase, indicating ulterior desires which repel. Phrases like “Insist on this brand,” “Avoid imitations,” ”Look out for substitution.” Such appeals have no good effect, and they indicate a motive which buyers cannot sympathize.

Anything which suggests an effort to sell on lines other than merit and service is fatal.

On Capitalization:

Many put their (ad) display lines in all caps. They think they look more prominent. But all our reading is done in upper-and-lower case type. We are accustomed to that.

On Superlatives:

To say that something is “The best in the world” makes no impression whatever. That is an expected claim. The reader may not blame us for exaggeration, but we lose much of his respect. He naturally minimizes whatever else we may say.

On Image Ads:

I am prepared to believe that on some lines, like fruit and desserts, colored pictures may prove profitable. But I know of no line as yet where, on traced returns, they have warranted their extra costs.

On Coupons:

Countless tests have proved that coupons multiply returns. I have seen many tests made by mail-order houses, offering catalogs. Some ads had coupons; some did not. The difference in returns was enormous.

-

I certainly recommend perusing the free online editions and gleaning helpful tips for your own advertising pay-per-click campaigns.

During the recession,  did consumers hold back and  buy a less expensive turkey to serve for Thanksgiving dinner?

Google Trends indicates yes.

A search trend for “Butterball” shows a dip in 2008 after peaking in 2006.

Butterball turkeys are one of the more expensive turkey brands.  What other indicators does Google trends reveal?

On the timeline, we see a news snippet at point B “Butterball still poultry icon after sale“.  This was the point in 2006 when the Butterball brand was sold by ConAgra to a company called Carolina Turkeys. Carolina Turkeys was owned by Smithfield Foods and Maxwell Farms.

By the end of this timeline at point E in late 2010, Smithfield has sold its entire  49% stake in Butterball and  Carolina Turkeys company to use the proceeds to reduce debt.

A search trend for “recession”:

At the end of 2010,  will we still be talking turkey or will we all be butter off?

Pattern detection in Google Analytics is a lot easier with visualization tools. One of the neat visualization tools available to enhance your Google Analytics is JuiceKit that works through the Google Analytics API.

Very easy to setup. Just visit the Google App Gallery and login into your Google Analytics.

There are two tools available. A Referrer Flow and a Keyword Tree visualization.

The keyword visualization lets you see right away the most frequently used search terms and their performance.

There are a few parameters you can set like the size and color. In the example below, word size corresponds with the number of visits and color relates to the bounce rate. Other settings can include % new visitors or average time on site.

In the center of the tree, you will see a frequently used keyword. To the left and right of the keyword, you will see search terms used the most often in combination with your keyword.

A “Juiced” website screenshot of the Keyword Tree below:

-

One thing that stands out above are the queries related to pricing.  That is a question easily answered by contacting us on our online form.

A great tool to check out! Next post, I’ll explore the Referral Flow tool to answer the question “What pages are people viewing on your site and where are they coming from?”.

Is that the search query you were looking for?

Google Instant allows you to get to the right content faster because you don’t have to finish typing  in your full search term.  You get instant feedback during your search. Would that cause some search visitors to hit enter and conclude a query before there is a more exact match to the keyword they are looking for?

In our analytics, we noticed that during the course of a query some hits were being picked up as only partial keyword searches. Apparently, some search visitors are believing they have concluded a search query when they have actually not.

For example, let’s say I want to search for Christmas toys.

I start my query and get to the point below where I have typed in “christmas to” and not finished typing completely to “christmas toys”.

On the screenshot above, you notice that there are not any sponsored results.

But if I complete typing, this is what happens…

Now above, sponsored ads fill up the top and right sections of the screen.

What this means is that a window of opportunity exists (albeit perhaps a small one!)  during the course of a search query for top ad positioning by bidding on a exact match for [christmas to] . Suppose that after bidding on the close match, the real possibility exists for searchers catching sight of my ad, stopping the query before completion,  and clicking on the ad. The brief moment of top ad position is  likely far cheaper than bidding on “christmas toys” and might be worth the effort for the traffic you receive.

People say that giving is better than receiving, but that is certainly not the case with AdWords.

I wonder how modifed broad matches variants behave?

There are certainly different approaches to take with your pay-per-click advertising.

One of the best ways to learn ppc advertising technique is to copy what the big multinational corporations are doing online with their massive marketing budgets.

Proctor & Gamble is a favorite of mine.

Lets take a look at what P&G and some of its competitors are doing in the baby diaper category.

There are no pictures to display. No jingles here. Combined with calls to action, there are appeals that are proven to work such as:

• Save money
• Comfort
• Freedom from worry

Easy to copy ppc techniques that are simple and effective. Get yours today!

Many online businesses  get assailed by web marketing companies. Mark Dirks has written a series of blog posts about this bombardment called Web Marketing Pollution.

Below are some further questions you can ask and online tools to help you determine if an online company is on the up and up.

• Does the company have a legitimate address?

No, a post office box won’t do. Check the address in Google Maps Street View to see if an actual physical location exists.

• Does the company have a published phone number listed that they can be reached at?

Check the phone number using a reverse lookup at 411.com. Check  to see if the phone number is unpublished or not.  An unpublished number can raise a red flag.

• When was the website created? How long in business has the company been around?

Check the company’s domain at www.whois.domaintools.com.  At whois.domaintools.com, you can see the creation date of a website and get lots of great info.  If dates are very recent or don’t match with what is posted, a red flag is raised.

• Does the website have a past history that is in conflict with today’s information?

See if a previous version of the website exists at the Internet Archive WayBack Machine. Check past information on the  website for anything that may have changed

• Does the company list a portfolio for a source of references?

Check the websites referred to and see if information can be verified by a third party.

The online tools I have listed above should help you determine whether or not an online business is legitimate.

For some tips, let’s take a look at some of Google’s  paid ads and the ppc techniques Google is using. I took screenshots of a total of six ads.  The first one is for Google Docs.

Some points to take away from each ad:

-> Ad states a benefit to the customer “..from Any Computer.”

-> Call to action “Try now!”

-> Ad states a benefit to the customer “Reach Millions..”

-> Low price of $5 is emphasized.

-> First letter of every word is capitalized.

-> Exclamation mark used in last line.

-> Ad states a benefit to the customer “Boost Your Business”

-> Call to action “Use Right Now”.

->First letter of every word is capitalized.

-> First letter of every word capitalized.

-> Low cost and quickness is emphasized.

-> Ad states a benefit “Get Qualified Visitors..”

-> Call to action “Place ..Today!”

->Ad states a benefit of  ”Gain Customers in Minutes”

-> Call to action ” Sign Up”

So, in summary the main tips we can garner from Google are:
1) State a benefit to the customer.
2) Include a call to action.
3) Emphasize low cost, ease of use, and speed.
4) Start every word in your ad text with a capital letter.
5) Repeat one word from the headline into the text.
6) Try some ads using an exclamation point.

Mayday!  Mayday! We are sinking!

Matt Cutts did a Google Webmaster Central  YouTube video post on the recent Google algorithm Mayday update.

In his reply to the update on rankings for long-tail searches, Matt stressed the impact on the improved “quality” of search queries.  He also stressed that the “change” was “algorithmic” and not temporary.  The change is separate from the upcoming caffeine update, and that the caffeine update is  still on track.

What exactly is the main thrust of the Mayday algorithmic change? We do know that long tail rankings went down which may have something to do with a dampening of PageRank for internal on-page site maps. Doing a survey across websites with PageRanks of 4 or less, I’ve noticed a pattern of site map links in the footer not passing PageRank. Although the pages are indexed, there is not the expected ranking boost associated with having a site map link off the home page.  It may be very likely that additional stop words have been added to the algorithm to not allow anchor text to pass PageRank for a site that has a low score for trust.

I’m sure the algorithm change is not that straight forward but  I do believe that site trust plays a major factor.  A new combination of factors are now in play that affect long-tail search queries.