Semantic Publishing of DubLi Auction Ads with Semiomantics II

Synergetic semantic Effects on Louis Vuitton related Auction Ads for DubLi Auctions Part II (Documentation)

Vouliagmeni, November 3, 2008 – by Yorgo Nestoridis - YORGOOpublishing and Semiomantics,  

This article can be downloaded in PDF Format from HERE or Zip PDF: HERE 

This Article describes Test Results about Clustering of Ads, Targeting, semantic conclusions and the semantic interpretation by the Semiomantics EC Dublicator Script. The test has been run on the base of the description of an Auction Item on the DubLi auction Platform. The purpose is to test appropriate targeting and delivery of product ads in a pull marketing environment such as Google. Test set-up has been described in ‘Semantic Publishing of DubLi Auction Ads with Semiomantics I’.

1. Nikon D80 or Nikon D 80 ?

The devil is in the detail as the Germans say. We  started from a product description on DubLi Auctions:

Nikon Digital Camera D 80 Kit incl. DX 18-135      

The D80 Nikon digital mirror reflex camera is an inspiration to any creative photographer.
With its high-performance internal image processing and numerous image adjustment functions,
it takes advantage of the latest technology to develop new creative opportunities.

The D80 is ergonomically designed, easy to use and offers an extraordinary combination
of superior image processing, speed, precision, flexibility, reliability, reaction time
and efficient operation.
The D80 is fully compatible with the Nikon Total Imaging System,
so that your creativity knows no borders.

Success!

Included

Specifications

We notice that DubLi in its Title uses D 80, while in the subsequent text it’s D80. The below results show the difference between targeting D 80 or D80.
Nikon D 80

Nikon D 80

If however we use D80:

 

Nikon D80

Nikon D80

The result is different! Now lets have a look at what DubLi provides us with besides the above Item description to find out why we don’t see DubLi on the search return:
DubLi Headers for Nikon D80

DubLi Headers for Nikon D80

1. The Description starts with D 80 to continue with D80 
2. In the meta keywords you find first 80nikon, then adjustment, then advantage … not really keywords anyone would search for.
On Dublicator you find:
Dublicator Meta tags for Nikon D 80

Dublicator Meta tags for Nikon D 80

See the difference? In the keywords you wont even find the Nikon D 80 or d80, while in the description you automatically find the the full item description. The above has been extracted from a Dublicator script using simply the Y-media feed.
But it’s even more extreme on YORGOOpublishing which beats all test sites on this set of key-words:
YORGOOpublishing headers for Nikon D80

YORGOOpublishing headers for Nikon D80

Now this is amazing: no indications about Nikon D80 and there is absolutely no Meta description in the header of the site….but you find all the synergetic anchors and targets where YORGOOpublishing is involved. What you see here is historic in the sense that there is not one guru who will find an explanation for this and that’s exactly where Semiomantics come in. By the way, you would probably agree with me, that no one would want to copy these headers to promote the Nikon D80 :-). The secret is in Semiomantics’ way to communicate with Google. Unique, efficient and copy-proof, at least for the time being.
Let’s look at something more user targeted, like “[product_name] best buy”, example: here we add “best price”:
Best Price Nikon Digital Camera D80 Kit incl. DX 18-135

Best Price Nikon Digital Camera D80 Kit incl. DX 18-135

Interesting result:
1. The page displays 5 ads and 10 organic search results
2. DubLi related sites occupy 5/15 spots, that’s 33%
3. Of the 5 DubLi related spots 4/5 or 80% are Dublicator, 1 is DubLi itself.
4. Interesting: not 2 or the 5 DubLi related ads are equal, however they all are based on the same DubLi Auction Item Description.
It’s a typical case, where DubLi jumps from 1/15 to 5/15 or from about 7% to 33% due to Semiomantics Publishing by the Dublicator network.
Chances are big that we recover a good part of the traffic under this search result.
Let’s change the add on to “[product name] best buy”:
Best buy Nikon D80

Best buy Nikon D80

Result: 3 advertisers and 10 organic returns
1. Still 5 semiomantics returns but this time 5 of 13 instead of 5/15
2. Note again: the same diversity as above.
3. Each of our site is based on the same original ad with a different semantic interpretation.
Semantic interpretation, function of the semantic framework

The two above samples show that the same add can return different results depending on the semantic framework within which it is published. Each Semiomantics based site is defined based on user’s preferences on multiple parameters; chances that two users use the same customization features within a given topic are about 1:4,000,000,000. At least for as long as not more Semiomantics scripts are in use to promote amongst others DubLi related Auction items. An increase to 10 000 scripts in use would result in a probability of 1: over 100 000, meaning it remains highly improbable that two absolutely identical publications would be published just based on framework settings.

The above demonstrates: it does not really matter what we publish, but how we publish when we use Semiomantics semantics frameworks.

We have tested over 20 expression to distinguish identical posts and they all return top 10 rankings. An other example is: “[product name] best deal”

 

Nikon D80 Best Deal

Nikon D80 Best Deal

Again: 3 spots for the same purpose from 3 different sources.
Note: we have seen the exact 3 publications earlier under different search terms …
Clustering and increased scattering on Louis Vuitton

The below screenshot gives an impressing picture of the clustering and scattering function described in Semantic Publishing of DubLi Auction Ads with Semiomantics I:

 

Louis Vuitton Clustering an Scattering

Louis Vuitton Clustering an Scattering

Result:
1. 70% of the top 10 concern DubLi amongst which a Sold item record from DubLi on rank 8.
 
2. The cluster is interesting here: it comprises auction item descriptions, articles about Items described, articles about the publishing of articles about items on auctions as well as ads in 2 different languages, English and German. (check namely ranks 3 and 4: same source, different language and listed both on English Google.)
 
3. In this example we went for the “gift voucher Louis Vuitton best price” but all the other clustered Louis Vuitton related ads and articles pop up on equal terms and with equal importance.
4. Taking out of the search the terms ‘best price’, we get a completely different result:
 
Louis Vuitton Gift Voucher

Louis Vuitton Gift Voucher

The reason for this difference is the relative importance of the ‘best price’ element in the ad.
The name of a product will always return some high profile search results, however it seems that most people who are ready to buy the product would add some additional keywords to create long-tail entries such as “…best buy”,  ”…best price”, “…discount” and so on. Coupons, vouchers and such like enjoy popularity as well.
By the way: Google gives us sometime guidance about which terms may be hot in related searches as done on Nikon D80:
Best Price Digital Cameras

Best Price Cameras

Please note: the German ad has moved from rank 3 to rank 5, but it’s still there similar to the content of the previous screenshot (clustered ads).
Where are the advertisers?

Serious publishers and advertisers alike spend time and money on assessing the market (target) as well as the competition. One of the question is: where is the center of gravity aimed by competitor’s ads and what are the privileged exact search terms.

Let’s return to the Nikon example:

Remember the controversy about D80 or D 80? It’s a matter of a space and yet, that space makes a whole difference!

 

Here is D 80 …

 

D 80

D 80

… and here D80:
D80

D80

Results:

1. On D 80 there are about 15% more search returns on Google 467 000 compared to 407 000.

2. On D 80 I didn’t see any ads, at least not from here in Greece

3. On D80 I get advertisers

4. INTERESTING: on both search returns we find Nikon USA within the top 3 with the same content, the other top ranked sites change from one search to the other. There is an interesting lesson to be learned here and we will elaborate on this in a later article.

 

We can simplify the D issue by just searching for nikon d80 compared to nikon d 80. Again Google gives us some indication here:

Please note: Google treats ‘d’ as a single expression in the search Nikon D 80:

 

Nikon D 80

Nikon D 80

 

While the search Nikon D80 deals with D80 as a single expression:

 

Nikon D80

Nikon D80

Again: D 80 returns more search results, however D80 seems to be more interesting to advertisers.
Note the Authority Listing for ‘Shopping results for nikon d80′, another topic we will deal with at a later stage.
Finally, what should we target, D 80 or D80?

Clearly both as with Semiomantics Frameworks it will not result in more work. Remember the pair of socks and the finite matches for infinite pairs of socks? Well, it’s along those lines that the issue can be explained. When assessing a market and competitors, the devil is again in the detail. On this issue it is possible that many advertisers get it wrong:

while the exact type designation is D80, many searchers may type D 80 and what counts is what the customer types in. They will use both, depending on the source of their information: if the ads about that camera is featuring D 80, as we noticed in the ad by DubLi, then it’s normal that D 80 will be what they are looking for.

In general one can say that the end user is not functioning like a precision clock. What matters to him or her is not the exact spelling of a product name or type, what matters is, whether he or she finds what he or she is looking for.

And that is may be the best conclusion to all the science and calculations and probabilities and possibilities: what counts for the customer is the result and therefore the publisher must be visible where the customer searches for that result.

In the end the customer is only to a limited degree predictable, some times more others less. There are measurably more probable key-phrases on many items. The publisher needs to take such higher probabilities into account and scatter the adds all around such centers of gravity. And that’s exactly where Semiomantics is pursuing its work and development: to provide publishers with ever better and more predictable publishing results on a target that will never be 100% predictable.

 

YORGOO's Make Money with Home Business – Digest is Digg proof thanks to caching by WP Super Cache