google (6)

A recent Yeoman study of over 1,500 items sold on Amazon from a leading US manufacturer found a whopping 70% contained errors or did not follow Amazon’s best practice recommendations for providing product details. This study further showed the same level of inaccuracy on Google and Bing product searches. This is a maddening problem that plagues manufacturers in every industry. Once a product is “out there,” resellers, partners, distributors and reviewers end up shaping and revising the product details that your customers are going to use to make their purchase decision.

Products sold by Amazon definitely had better data quality, but less than 4% could be considered 'optimized' with the proven best practices for any item:

  • Descriptive title at least 50 characters long (including brand, purpose, color, and set/quantity info)
  • Multiple images of product
  • Bullet/summary list of product features
  • Complete description that includes product benefits, usage instructions, what's in the box, and etc.
  • MSRP reference price
  • Proper brand name and manufacturer info
  • Mfg part number, ISBN, or UPC
  • Proper categorization within Amazon browser tree


This isn't necessarily Amazon's fault. Amazon relies heavily on third party merchants as part of their product information set. Manufacturers who don’t understand how product details evolve online end up with a mish-mash of ‘ugly baby’ product details that hurt their sales - both offline and online. 

To address or avoid data quality problems, make sure you have these three key components in place:

  1. Baseline data components in an accessible format. It's amazing how many manufacturers and publishers claim they have great data only to find out its on a spreadsheet somewhere in engineering or marketing.  If you're data isn't in an easily accessible format that partners can use, you can't expect them to get it right.  There are 64 "standard" fields that every manufacturer could pull together and offer up to their partners.  Do you know what they are?
  2. A monitoring program for your online product set. If you don't have a program that monitors your product details on the major channels (including eBay, Amazon, and the general web) you can't take any action.
  3. A product detail action plan. Once you've begun monitoring, make a plan to address bad data as you find it. This includes:
    1. Registering with Amazon brand registry (key for disputes)
    2. Addressing any trademark or misrepresentations on Google or other sites
    3. Providing easy access to your product details (either via website or product feeds)


Yeoman specializes in working with manufacturers and publishers to help you understand your online sales and distribution channels.  Contact us today for a review and assessment of your current data quality online.  We've helped scrub, clean, and optimize over 500,000 items in the last year alone! 

Learn more...

Read more…

What if, as a manufacturer, publisher or IT company, you could predict the future? You’d probably start off by predicting how much inventory you need to carry, followed closely by predicting the next products your customers are going to want to buy, so you can have them ready and waiting the next time they hit “search.”


Magic? Not really. Impossible? No. Mining the data you already have access to about your customers and web visitors can give you amazing insight into their future behavior and preferences – but only if you capture it and use it.


Case in point: Netflix cashes in on their own crystal ball

When Netflix went looking to produce its own TV show to compete with cable, they – like any other business – wanted to be as sure as possible it would be a big hit. But unlike a lot of other businesses, Netflix captures and mines enormously detailed data on their customers. That’s how they predicted that their new series “House of Cards” would be a huge success before they even taped one episode.


Specifically, Netflix data crunchers used 30 million data tags per day (tags generated by 60 million worldwide viewers) to predict the success of House of Cards. They basically took long tail search (the type most often used for automated recommendations like those on and used it to predict what their next piece of original content should be. And it worked.


House of Cards is directed by David Fincher, a director many Netflix customers love, stars Kevin Spacey, an actor many Netflix customers love, and is based on a popular British series – and lots of Americans love to watch TV shows that were hits in Britain first (Downton Abbey anyone?).  This diagram from the article in the New York Times sums it up nicely:

The verdict: data mining works to predict the future

According to Netflix, within weeks of release, House of Cards was the most streamed piece of content in the United States and 40 other countries.


As a fairly typical Netflix viewer, I have to admit that I love political dramas, British shows that are remade for an American audience, and Kevin Spacey. I just never would have put it all together myself, until House of Cards came out, and now I can’t wait for Season 2. And, FYI, we were about to cancel our Netflix membership.


So what are YOU doing with your customer data? “Nothing” is no longer an acceptable answer. Get in touch to find out what we can do for your predictive abilities.

Read more…

Google has confirmed it - a Buy Button is “imminent.”  The button is expected to be rolled out on mobile devices, and will enable people who click on product ads in search results to buy those products without navigating to a third-party site.  The button, following similar moves by Facebook and Twitter, are a significant departure for the search giant, which has built its business based on ads that link to other websites.

"The rationale is to reduce friction for customers'" said Omid Kordestani, Google’s Chief Business Officer, "making it simpler to complete online purchases."  Trust us, there's another reason - Google is facing significant competition from Amazon and others when it comes to people searching for products and has been steadily moving to be more "direct."  Recent examples include:

Their skill sets in selling direct have been honed by almost a decade of direct sales for its mail apps and cloud services.  They have even tried the payer route with Google Wallet.  In fact, the "Google Graveyard" is a robust list of products and services that were tested and then scraped including Google Reader, iGoogle, Google Commerce Search, Google Wave, etc.

Don't expect them to quit on the Buy Button any time soon.  Don't think of this as a test, but more of an 'evolution' in online commerce.  For a manufacturer, this presents several key questions:

  1. How real is the Buy Button threat for your market segment?  Mobile purchases vary greatly by market segment.  The most likely segments that will be impacted are any purchases that are under $50 (a proven price point for mobile) OR items that have a 'fast pickup' option
  2. Are your retail and distribution partners properly aligned with Google?  If you have B2C end customers you should already be verifying if your products are available via Google Express
  3. If they are not aligned, should you take a more direct position with Google?
  4. Have you looked beyond Google and assessed all the 'digital' players that continue to make inroads?  This includes the major players like Amazon and Wal-mart, as well as the up and comers like Hayneedle, Jet, and Rakuten.
  5. Who is responsible in your organization for following up on this?  If you're thinking "Google is handled by marketing," you're dead wrong.  This is a sales and distribution challenge you need to address. Let marketing market.

If you don't have solid answers to the questions above, its' time for a conversation.  Pick your option - phone, fax, Live Chat, LinkedIn, or Email.  We're ready to when you are.

Read more…

Google recently announced that their Google Shopping feeds will no longer offer up product details in search results for free. The original program, started about 5 years ago, let manufacturers, retailers, and distributors load specific product information into Google's search system. Product detail results would show up within the main search itself or within the specialty shopping tab

The original system (called Froogle) started off gathering information by crawling existing ecommerce sites, but Google eventually decided that allowing vendors to upload their own data would make it easier for everyone. Microsoft soon followed with their own version for MSN/Bing.

Neither service charged for product listings, allowing them to crush pay-per-click price comparison sites like shopzilla, pricegrabber, nextag, bizrate, etc. But now Google has decided to switch to the very same pay-per-click model it was competing against only a short while ago.

Anti-monopoly forces will likely cry foul; the #1 search vendor introduces a 'free' product listing service that directly competes with paid price engines, builds up market share and acceptance, then adopts its competition's pricing model four years later?

They may be able to prove their point. Yeoman did an analysis of the traffic of five of the top price engine sites and saw a marked decrease in activity from 2009-2011.  What is interesting is that these sites' traffic seems to be ticking up given the recent changes (Source Compete Data 2009-2011). What's even more interesting is that many of these sites modified their models and began using Google's shopping API to pull in relevant data.

How does this impact you?

For a manufacturer there are several items that should be addressed:

  1. Review the current role of Google Shopping in your Data Quality, Merchandising, and Demand Generation programs. Make no mistake, every organization should be actively managing their product feeds. Google remains the #1 site in the world in terms of visitors and holds 70+% of the market share for search. Even if your products are sold 100% by a downstream distribution system, this is where many customer get a  'first look' at your products.
  2. If you are already managing the feeds - continue the process and review whether or not the marketing units want to add a paid campaign to integrate your sites. If you are the original manufacturer of any item, we highly recommend creating a budget for this. Not only will it pick up product queries your downstream channel misses, it will give you the ability to directly control the quality of your products on the general web (both Google and Microsoft consolidate data via UPC or manufacturer part number and do enforce copyright rules when pushed).
  3. If you do not already have feeds - Audit which of your partners do have feeds and make sure your preferred partners are aware of the change.
  4. Revisit independent price engines - These sites will likely see some level of resurgence in traffic if retailers start to test out different pricing options. The good news is that the major players will likely have to be below Google pricing, giving existing users a break. Major players include: Shopzilla, Nextag, Pronto, PriceGrabber, TheFind, PriceRunner (UK), Smarter, and Bizrate.

The challenge for all will be to make sure this change is addressed at multiple levels including your overall marketing/branding strategy, online data quality programs and your online sales and distribution model.

How can we help? 

Yeoman has a passion for online sales and distribution and optimizing the components that make a manufacturer successful.  Each organization has a unique marketing and sales strategy, but there are core components that everyone needs in order to be successful. We live those components, whether it's price engine analysis, data quality programs, or outsourced web operations. Give us a call or chat now to learn more.

Official Google post:!/2012/05/building-better-shopping-experience.html

Background anti-trust stories / complaints:

US Hearings (2011):

Brazilian Charges (2012):

European Investigation (2012):

Read more…

Get in Touch

Phone: 800-667-6098
Fax: 603-546-7274
Email Us