Comments

1. Nicolas Leroy, May 18, 2009:

   Nice article – hProduct is all about the product, not about the price / delivery cost… In the discussions in the microformat community, all the transaction-related data shall belong to hListing. Strangely, Google hasn’t announced support hListing which is a microformat much older than hProduct.

   So I’m not sure Google will try to go and use microformats to feed Product Search with price data… I believe in the short term, GG is more interested in product data (i.e: reviews) to extend search results…

   But in the long run, I think you are 100% right.

   Nicolas

2. Marc Adelman, May 18, 2009:

   Alan,

   Great post! I wonder what effect hProduct will have on actual SERP rankings.

   Let’s say that some searches for “ipod nano $100″.

   Will Google’s natural search algorithm give any additional weight to this new field and give better placement to a page that has $100 populated in the in the cost field of hProduct?

3. Jay Myers, May 18, 2009:

   I believe that hProduct will actually level the playing field of online commerce. In my experience, large retail organizations (especially those with more traditional brick-and-mortar stores) move incredibly slow and are some of the last to implement changes to their front-end (X)HTML. Smaller organizations with a more nimble technology stack should be able to get these changes implemented in a faster way, and should see considerable future benefit from being early adopters.

   “hProduct doesn’t encompass shipping, tax, bundled pricing, or even UPC. Argh.” True on transactional attributes, like shipping and tax. False on UPC (UPC would fall under the “identifier” attribute). An important thing to note: one microformat will not identify all of the attributes of a typical product detail page. The total solution will use hProduct in combination with other microformats, as building blocks for a complete solution. Transactional details like shipping, availability and condition are properties of a new hListing proposal (see http://microformats.org/wiki/hlisting-brainstorming for more). You should see more about hListing in the coming weeks. For information about combining different microformats with hProduct, I suggest keeping an eye out on the hProduct recipes page: http://microformats.org/wiki/hproduct-recipes. I have begun updating this weekly and would love to see others post their solutions as well.

   In the end, the work done around hProduct is meant to benefit the consumer. The hope is that this microformat will make it easier to discover products and the retailers/ manufacturers that sell them. This gives the end user the benefit of choice — they can easily find the products they want to fit their needs and can choose the retailer or manufacturer they want to purchase from.

   Finally, microformats, RDFa, and related markup methods go beyond SEO and Google; they can serve as the front-end to a dynamic, smarter web of linked data. Don’t implement microformats as a “silver bullet” for SEO. Implement them because your users deserve better, more concrete data regardless of the path they take to discover your products.

4. Jesse, May 18, 2009:

   I like that Google is going to microformats. It could create more refined searches for products, people, reviews, and listings.

   hProduct searches would be a great advantage to eCommerce companies as users could easily search for products without having to fish through google products or multiple websites to find a product.

5. Alan Rimm-Kaufman, May 18, 2009:

   Thanks, Nicolas for pointing out hListing.
   Thanks, Jay, for all the info. Totally agree that semantic markup is far far bigger than an SEO hack — I’m in the camp that these modest little tags will transform the web in a big way.
   Jesse — I agree with you and Jay that providing these data to Google will help (a) consumers and (b) Google. I’m not sure it will help e-Commerce companies across the board. See for example this comment on G webmaster central — incrediblehelp captures the argument saying “Why in the world would I want to short change myself a possible visit by giving Google more information to list on the SERP, when I want that visitor on my site?” Online retailers have long struggled with how selling thru product comparison sites (eg next.com, shopping.com, shopzilla, y! shopping, amazon, etc) degrades their brand, essentially reducing the merchant to a fulfillment house. (Anecdote: in a 2007 NRF survey, consumers named Google the 9th ninth most popular e-commerce site online — old data, but telling.)
   Bottom line is these changes will occur, and online retailers need to be aware they are coming and begin planning for them now. Stores need to ask themselves this hard question: “Once we acquire a new customer from Google, what compelling reasons do we give that customer to begin their next shopping experience by typing our URL into their browser, instead of returning to Google and doing a product-centric search?” Retailers without a strong answer to this question will struggle mightily.

6. George Michie, May 18, 2009:

   I have a long history as a poor prognosticator, but I don’t see this as likely to transform user behavior. The failure of Google Checkout to gain traction, the failure of Google base to take over the world, and the failure of Google plus box as anything other than a novelty suggests to me that the problem isn’t related to Google and a lack of informative tags.

   The problem with these notions is that people apparently don’t want to shop on Google.com. No matter how well the products are tagged someone searching for a “diamond ring” isn’t ready to look at a list of products. They need education on the types of options available to them and the range of styles and price points to think about. The idea that Google will have the right types of navigation options for every type of search seems pretty far flung.

   My rationale boils down to this: look at the web sales generated by CSEs as a percentage of total competitive PPC sales for a variety of retailers. For SKU based commodity retailers the CSE’s can be 1/5 to 1/3 of the overall volume. However, for non-sku retailers (furniture, apparel, bedding, jewelry, etc) the CSEs usually don’t tip 5%.

   We shall see.

7. tom funk, May 19, 2009:

   Thanks for a great post and analysis Alan (and for adding to my things-to-do-list)! This is one of the many things we should be doing to get our product data crawled and parsed and back-at-ya on the search results pages in as many of the buckets of Universal search as possible. So now here’s the new checklist (I would be lying if I said all our clients are already doing all or even most of these!):

   > Submit to Google Product Search (formerly Froogle)
   > Publish new products or specials in RSS format
   > Optimize product images for image search
   > Submit product data feed to the CSEs
   > Enable customer reviews and ratings, and syndicate them

   Speaking of images, I wonder if using the photo property will help rankings for Google Image Search?

8. Martin Hepp, May 26, 2009:

   Dear all:

   Maybe you also want to look into the GoodRelations vocabulary for e-commerce on the Web of Data. It is a fully-fledged standard language for describing all details of businesses, offerings, and products/services as Web meta-data.

   You can use it for simple things as just stating your contact details and opening hours, or for complex things like modeling every single product model with features etc.

   There are mappings from various existing formats either available or underway – e.g. from the various Googgle feed formats.

   For a quick start, please see the following links.

   Best wishes

   Martin Hepp
   mhepp@computer.org
   http://www.unibw.de/ebusiness
   skype mfhepp

   Tool for registering your business:
   http://www.ebusiness-unibw.org/tools/goodrelations-annotator/

   Webcast explaining the Web of Data for E-Commerce:
   http://www.heppnetz.de/projects/goodrelations/webcast/

   Overview article on Semantic Universe:
   http://www.semanticuniverse.com/articles-semantic-web-based-e-commerce-webmasters-get-ready.html

   Project page and resources for developers:
   http://purl.org/goodrelations/

   Upcoming events:
   Full-day tutorial at ESWC 2009: The Web of Data for E-Commerce in One Day: A Hands-on Introduction to the GoodRelations Ontology, RDFa, and Yahoo! SearchMonkey

   http://www.eswc2009.org/program-menu/tutorials/70

   Talk at the Semantic Technology Conference 2009: Semantic Web-based E-Commerce: The GoodRelations Ontology

   http://www.semantic-conference.com/session/1881/

Tags