Holiday Shopping at Yahoo! Shopping

I’m hoping to really quickly get out a number of reports on what the comparison engines are doing to prepare for the holiday rush. Here’s the first; an interview with Rob Solomon & Chris Saito from Yahoo! Shopping…

When do holiday sales start coming in? When do you see or expect to see a pick up in traffic due to holiday sales?
Rob: “We start seeing an up-tick with the back to school season. Traffic goes nuts at the end of November. Our busiest day is the 1st or 2nd Monday in December.”

Do you have any specific expectations that you can share in terms of year over year growth in traffic? Or in terms of GMV/GMS growth?
Rob: “We’re seeing 20-30% year over year growth in traffic in general to date. Also, order item size is up year over year.”

Which categories/products do you expect to be breakouts? How about the iPod nano or Leapfrog’s FLY pentop computer?
Rob: “We’re seeing fantastic growth in apparel. Apparel is the breakaway category.
“The iPod nano is getting interest from a search perspective. Digital cameras are always in top.”

Chris: “We’ve seen woodstoves and pellet stoves in the top 50 – where they’ve never been before – because of the rise in heating costs due to Katrina. There are a lot of Halloween costume queries right now as well as the usual suspects like PlayStations and iPods. However, there isn’t a ‘must have’ product this year. There’s no phenomenon around a particular product.”

How do you expect most people to shop – browse or search?
Rob: “People are still doing both, but search is a much bigger driver; people end up on the site through search queries, search engine optimization (SEO), and search engine marketing (SEM). More people start in a search destination and then get into search and browse. Probably 65-70% of people are searching,”

Chris: “Certain categories lend themselves to browsing – like apparel. Browse is driven by category.”

Rob: “We view vertical search as one of the most important things that we’re doing. Browsing is a big part of the shopping experience and it always will be. Once you’re in the shoes category, you see people clicking on women’s shoes, then selecting a style, then size. That’s the type of browsing we’re seeing.”

What do you have planned for the holiday season? What were the top 3 things that you learned last year that might be driving changes on the site this year – people wanting coupons – people wanting fraud protection, people wanting more choices of products, etc.

Rob: “First, people want more and better information. We’ve made it easier to access information about products. There’s a much more detailed product page and all information is on one page. Second, value is a big driver of internet shopping, and we’re aggregating deals, coupons, and rebates. Third, overall usability; everyone has a different way to shop.”

“And obviously [through it all] we’re making search better – that’s part of our DNA and will always be important. We want to make search more comprehensive and relevant.

Chris: “Integration of our service with the rest of the Yahoo! network and being open with our search database; both within Yahoo! and outside of the Yahoo! network. Our APIs are an example; we launched the Yahoo! Shopping Web Service APIs last August and recently added access to user reviews. [Editor’s Note: Find out more about Yahoo Shopping Web Service APIs through my interview with Chris.]

Will there be a special holiday section like there is for Halloween?
Rob: “Yes. We always do that and feature different things like Free Shipping, which is one of the most important features for consumers. We’ll have top 10 lists by categories (toys, electronics, CDs, digital cameras) – lists are a very easy way to aggregate what’s popular.”

Can you talk more about Gift Finder BETA?
Rob: “Part of the technology is from ChoiceStream, but the database, interface, and functionality is ours. We rolled is out for mothers day. We typically elevate the gift recommender during the holiday season.”

What new features will you be implementing/have you implemented which you think will make a difference to consumers this holiday season? Mobile shopping, RSS, etc.?
Chris: “Mobile shopping comparison which lets you search for products while in the store. While looking for digital cameras, you can see what the price is online by performing search queries through your mobile phone. It’s in BETA. We’ve seen a nice up-tick in usage and in far exceeds our projections.

Rob: “Usage on the weekends is spiking.

Chris: “Yes, we definitely see a spike of usage on Saturday and Sunday. We also launched a Send to Phone feature [located at the top right of all product pages]. We just launched a lot of additional information on the product overview page. When you do a search and click on a product, you’ll see price comparison, specifications, a large photo, user reviews, reviews form the web. All of which is big step forward in user experience.”

Rob: “We’ve done a lot of interface things to make ratings and review more accessible, and we’re dedicating more space to user reviews and ratings. We’re definitely driving community features and functionality. RSS improves retention. The goal of [the additional features and community aspects] is to provide better information.”

“From a relevance perspective, we’re refining the search algorithms and there are a lot of behind the scenes projects.

Are there any specific tech improvements that you’re making to the site?
Rob: “Relevancy. More query classification. Deep categorization and attribute extraction. Much finer narrowing ability and allowing people more choice from a narrowing perspective which lends itself to a much better experience.”

Chris: “A lot of it is machine learning, which means it gets better all the time. We’re trying to use machine learning to get smarter. Matching – we’re trying to match merchant offers to specific products. Query classification – breaking down queries into categories. For example, does ‘blender’ belong in Home & Garden or Music [if there’s a band called blender] – that’s an ugly and hard problem to solve.”

You now have Deals/Coupons available, but they aren’t contextual to the product. Can you talk more about this?
Rob: “We’re not announcing the name of the providers, but we’re aggregating deals and coupons. Right now they are merchant contextual. We’re working on product-level contextuality.

Chris: “It’s something we’re working on for the holiday season. For rebates, they are directly contextual to the product.

2 Responses to Holiday Shopping at Yahoo! Shopping

  1. For information, Yahoo! Shopping already uses our technology of attribute extraction for all its European Sites.

    Through Kelkoo (Yahoo! Company), they offer to their visitors, better search services around drill-down and attribute search.

    This technology is based on Artificial Intelligence Engine and works on the principle of Machine-Learning

    See the press release.

  2. Andrew says:

    The “query classification” push makes no sense to me. If a lot of effort is spent trying to turn search into browse, perhaps browse would be better served in the first place. Yahoo has been trying to force search since the launch of Froogle. Search is great for tail products, but for general queries it’s always going to get back into a browse functionality, since that’s how people actually shop.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 298 other followers

%d bloggers like this: