We use cookies. You have options. Cookies help us keep the site running smoothly and inform some of our advertising, but if you’d like to make adjustments, you can visit our Cookie Notice page for more information.
We’d like to use cookies on your device. Cookies help us keep the site running smoothly and inform some of our advertising, but how we use them is entirely up to you. Accept our recommended settings or customise them to your wishes.

Standardizing Comparison Shopping Feeds

About a year ago, I floated the idea of the industry working together to standardize the formats used to send data to and receive data from the shopping comparison engines. After much work from many folks, this dream is starting to become a reality. There will be a press release going on the wires next week from NRF and ARTS on this initiative, as well as an announcement at Shop.org. I don't want to jump the release date, so I'll hold my comments for next Tuesday. I'm pleased this effort is starting to gain some momentum, and am delighted by the many players who have signed on. For background, here's the original post on the project.

Date: Fri, 23 Sep 2005 14:25:36 -0000
To: [email protected]
Subject: Proposal To Standardize Shopping Comparison Feeds


I'd like to bounce an idea off this group: I want to garner support for developing a single XML data feed spec to support all current shopping comparison sites.

Currently, each of the major URL-based or SKU-based online advertising venues (Shopping.com, Froogle, Shopzilla, Yahoo Product Submit, Y! Shopping, Pricegrabber, MSN Shopping, NexTag, CNET/MySimon, etc etc etc) have different formats for accepting data from retailers. And each takes a different approach when reporting cost data back to advertisers. Some don't yet provide advertisers the detailed click and cost data advertisers need to run their campaigns efficiently.

We know the result -- advertising via shopping comparison feeds is currently a pain for many retailers. And many of the feed engines are small, lacking enough traffic to drive meaningful sales. Faced with painful setup and likely small resulting sales, many advertisers skip using many of the feeds.

Outsourcing doesn't solve the problem for most retailers. Search marketing agencies face the same economics. Doing a good job building and optimizing feeds has cost, but again the resulting retailer sales are often small, often too small to support outsourcing.

It is a chicken-and-egg problem: many retailers don't advertise on the feeds because the implementation cost is too high given the available sales; this in turn depresses revenue to the feed engines; this in turn hinders the feed engine's efforts to build traffic to their sites through advertising and content; this in turn reduces the importance of the feeds; which in turn keeps sales low.

One way to create a win-win-win-win situation (for retailers-agencies-engines-consumers, all around) is to take some friction out of the system.


Standardize the format retailers use to send data to the feeds, and standardize the format the engines use to report detailed cost and click data back to retailers.

After the Civil War, different parts of the country used different size railroad tracks, greatly hindering commerce. The current state of SKU-based advertising feels alot like that.

After much debate, in 1886 our nation's railroads settled on standard rails set 4' 9" apart. That's what we need now need for feeds: agreed-upon XML standards for submitting products and getting back detailed cost data.

What would happen if we had such a standard? Retailers would spend more advertising on the dollars on the feed engines, allowing them to expand their audience. Consumers would have an easier time finding products across a broad range of merchants, growing the e-commerce industry. Friction would decrease, leading to more efficient markets. Merchants selling good products at fair prices would see their sales grow.

Like this idea? If so, what are next steps?

We could use this forum to agree on wording of an open letter expressing support for this idea. Folks who wanted to could sign the letter in support of the concept. No real commitment involved, just lending your name in support of the idea. We'd submit this letter to the larger e-retailer community, the trade pubs, the agencies, and the feed engines. Those of us who cared enough to put in some time would convene a small working group. The group would have members representing retailers, agencies, and feed engines. I care about this enough to put in the time, and would volunteer as one of the agency representatives.

The group would combine the current feed submission formats into one common spec for describing products. And the group would develop a common standard for reporting back costs. The hard costs of this project would be modest -- maybe some consulting hours from real XML wizards -- we'd ask the engines to chip in to cover those.

The best is often the enemy of the good. If we went for a perfect solution, we'd get bogged down forever in the complexities of retail; the process would stall; and we'd lose the baby with the bathwater. A good simple 80% solution would move our industry forward tremendously. Think of the RSS spec for content publishing -- RSS isn't perfect, but it is decent and it is simple, and so has become a raging success.

If all went well, there's an outside chance we could have a spec done in a year, in time for Christmas '06. That'd be amazingly cool. Imagine if next year you had the opportunity to increase your advertising reach across all the feed engines, relatively easily.

So, what do you think? Comments? Suggestions? Thrown tomatoes? Volunteers?


Well, we didn't quite make the Christmas '06 deadline, but good progress has been made. More on this next Tuesday.
Join the Discussion