About two weeks ago I received a call from my friend Brian Smith of SingleFeed and Comparison Engines for some feedback on the addition of new and customized attributes to the Google Base (Froogle) shopping feed.
Brian covered these attributes on his SingleFeed blog but he did not get deep into what these attributes can mean for the the online shopper or how merchants can take full advantage of these attributes to conquer Google listings.
Basically, Google Base lets online merchants upload product description feeds which traditionally include such fields as size, image, weight, color, price… etc., in order to compare and contrast with other merchants when a Google user looks for different items.
And we all also know how the long term plan of Google Base is for Google to serve these Google powered results within Google search, personalized search results, and Google recommendations (think GMail).
In essence, Google Base is letting certain groups of web sites bypass traditional SEO while expanding the art of SEO beyond the web page and into the direct feed.
Here is a list of some of the new attributes from Google Base:
department
made_in
material
style
binding
edition
genre
pages
publisher
film_type
focus_type
resolution
tech_spec_link
zoom
functions
wireless_interface
battery_life
capacity
operating_system
optical_drive
recommended_usage
screen_size
weight
aspect_ratio
display_type
color_output
memory_card_slot
load_type
feature
height
length
width
installation
occasion
heel_height
shoe_width
age_range
platform
rating
When Brian called me he asked “Why do you think Google is not just gathering this information from the merchant web sites?”
My gut reaction is that most online stores use content management systems and shopping cart merchant systems that are entirely boilerplate and do not allow for such customization.
Let’s take my own shopping experiences as an example.
Real Life Shopping Mirrored on Google
I’m 6′5 (195 cm), wear an XL shirt and have a 38 inch sleeve length. 75% of the time, if I try on long sleeve button down shirt, the sleeves end at my forearm, 15% of the time the sleeves run down past my knuckles and 10% of the time they fit just right.
An extra large shirt is supposed to have a 36 inch sleeve. But as you real world shoppers know, such is not always the case.
I know that Banana Republic shirts are usually cut to a longer sleeve but have picked up a shirt off the shelf from time to time and run into problems with the sleeve or the possibility of ripping out the shoulders and back.
So, how can Google let the online shopper have a similar experience to the brick & mortar shopping experience? By pushing online merchants to define every little attribute of their products so the end user knows that this product is perfect for them.
Such attributes for clothing are height, length and width but what Google Base is now doing is letting their merchants define their own custom attributes in feeds! Merchants can add sleeve_length, neck_width and shoulder_width to their shirts! Pleated or flat_front to pants! Shoe_width and length to shoes!
More on custom attributes from Google Base:
To define a custom attribute in a tab-delimited file, you must start the attribute with a “c:”. For example, if you would like to define an attribute called store_department, the attribute in the header column should be called c:store_department. You must use underscores ( _ ) to denote spaces. You may also include information on the value type based on your custom attribute. For example, c:store_department:string lets us know that the custom attribute “c:store_department” has a “string” value type.
Ideally, I’ll be able to go onto Google in a year and search for “cotton poly blend argyle sweater with 37 inch sleeve length” and Google will be able to serve incredibly targeted results from the merchants who took the time to define such attributes.
Ultimate Shopping Experience for Picky Shoppers
Go beyond the clothing sizes, and think clothing shopping experience. Here are some other Google Base attributes:
* color
* department
* made_in
* material
* size
* style
Do your customers prefer ‘Made in the USA’ to ‘Made in Burma’?
Do your customers prefer cotton to spandex?
Do they shop by color? I know some people prefer different colors at different times of the year.
Googlers looking for a white, gold or red dress for New Years’ Eve? If you define your Google Base attributes these picky customers will find you and buy from you.
Beyond Clothing
But Google Base attributes don’t stop at clothes. Think food. Are you lactose intolerant? Do you sell lactose intolerant products?
Why not upload a Google Base feed defining your products as being lactose free, or gluten free, or wheat free?
Ideally, with the new attributes and custom alternatives, Google Base will have no limits, and the merchants which take advantage of this now, will be reaping the rewards down the road when Google integrates more and more Base results into its traditional search engine.
NOTE: Thanks to Barry for posting on Google Base at SER this morning. His posting reminded me of Brian’s call.









Comments
8 responses so far ↓
Ken Savage on Feb 6, 2007 at 11:21 am
It seems to me description, price, product_type, and title would be the most important to have on each product. Are there any content managers or shopping carts that have this feed built into them.
Someone said Zencart a while back but i don’t see anything on searches for it.
Ken Savage on Feb 6, 2007 at 11:27 am
It seems to me description, price, product_type, and title would be the most important to have on each product. Are there any content managers or shopping carts that have this feed built into them.
Someone said Zencart a while back but i don’t see anything in search for it.
Update: I found a few links for a google base script using Zen Cart.
google base script
google base script
Loren Baker, Editor on Feb 6, 2007 at 11:31 am
Ken, most shopping cart technologies do have the basic descriptions and the ability to export via .csv or .xls file.
The ability which Google gives these merchants with the new Google Base attributes is however, so much more powerful.
Think in terms of TickleMe Elmo. What if 100 merchants had Elmo’s description, price, product_type, and title uploaded, but only one has the different colors available?
As for the attributes such as “Department”, imagine a Google 3-D Virtural ‘Second Life’ style department store where users could scroll through the aisles and departments such as cosmetics (with no_animal_testing attribute), men’s, boy’s and others… while being able to purchase from ALL merchants which make up the virtual store via one Google Checkout purchase.
Such a user experience would be as efficient as it is exciting.
Durk Price on Feb 6, 2007 at 3:14 pm
Great article. As someone who is also 6′5″ tall and needing a 2XL with a tall preference I know the frustration of not getting enough information to make an informe buy. And I understand how groundbreaking this effort by Google is to assist in the customer experience. I may be in the guise of forcing the retailer to this point of view, but ultimately it will increae the quality and depth of the online experience.
kid disco on Feb 6, 2007 at 3:39 pm
Damn… you are huge!
CarstenCumbrowski on Feb 7, 2007 at 7:24 am
This will help Google to improve their matrix-like, faceted or dynamic drill-down navigation (however you want to call it)
See Faceted Classification to get an idea of the principle behind that.
This type of navigation becomes more and more popular with merchants. It’s easy to use, flexible and relevant. No pre-defined paths or hierarchy (e.g. static product categories)
See Walmart, eToys, BevMo.com, HomeDepot.com for example who are using products like Endeca Guided Navigation, Dieselpoint Faceted Navigation or Fast ESP Dynamic Drill-down for their faceted navigations.
Implementing those products really forces merchants to clean up their product attributes. Errors become much quicker visible and noticed with that type of navigation. If you use Froogle who had to “guess” the attributes for their faceted navigation, you will know what I mean.
With supporting those attributes, Google has to make less guessing and with increased accuracy of those properties on the merchant side are Froogle’s results also getting more accurate and have less “false positives” or “wrong guesses”.
Also see the notes to “Product Category” at my Merchant Product Datafeeds for Affiliates - 101 article. You can replace “Affiliate” with “Froogle” in most cases since it is technically the same thing.
CarstenCumbrowski on Feb 7, 2007 at 7:32 am
p.s. merchants that implement faceted navigation in combination with full-text search usually experience a growth in conversion in the double digits percentage range. Increase in average ticket (order amount) is also not an unusual effect.
The professional products are currently very expensive and can not be afforded by small and mid-size merchants, but there are tendencies towards main stream adaptation. I saw already “semi-open source solutions” that seem to produce decent results.
I am convinced that it will become mainstream and the future and I believe that it will probably be referred to as a Web 2.0 feature on day as well. :)
cameron on Dec 12, 2007 at 1:54 am
VirtueMart is an open source shopping cart that has a feed capable of delivering the whole store or just “on sale” items.
Leave a Comment