Want to Build a Holiday Shopping Mashup? First Choose the Best Shopping Search API

This guest post comes from Marc Mezzacca, founder of NextGen Shopping, a company dedicated to creating innovative shopping mashups. Marc's latest venture is a social-media based coupon code website called CouponFollow that utilizes the Twitter API.

The holiday shopping season is upon us, and that means finding gifts, and at this point… quick! For consumers, it translates into searching and researching, and lots of it in a relatively small amount of time. It also means shoppers are looking for innovative websites or applications to aid them in finding gifts, low prices, and sweet deals -- all while making the shopping experience more enjoyable.

Developers can leverage existing Shopping APIs to quickly create innovative shopping mashups, seamlessly add shopping search to their existing website or application, or even create an entirely new website or mobile application.  In this post I'll tell you all about the shopping search APIs available today, and quickly breakdown their strengths and weaknesses.  If you haven’t already, be sure to read my previous post entitled Breaking Down E-Commerce APIs. What is the best Search API for Shopping? It depends on your business goals. Are you looking to allow users to compare prices?  Are you looking to generate revenue from referrals, or are you happy generating revenue from other methods such as advertisements?  Do you provide a loyalty program and require detailed reporting?  These are the things to consider when choosing any Shopping API as a publisher.

Search from Price Comparison APIs

Shopping.com, ShopZilla and PriceGrabber all provide very similar, production-ready price comparison APIs.  They have large amounts of data spanning tens of thousands of merchants, and millions of products.  While each Price Comparison API is very similar, the differences between the APIs come in terms of the number of merchants, number of products indexed, search algorithm, report tracking, and of course referral commission fees.  You’ll be able to generate revenues directly from these as they each have an individual revenue share partner / affiliate programs.  Note: These all work on a CPC basis.

Search from Affiliate Network APIs

Commission Junction and LinkShare both offer product search pieces included in their suite of APIs.  Utilizing one of these APIs is a great way to create a simple mashups for a select number of stores, but remember that ultimately you are limited to only merchants within that specific affiliate network.  Also, their search algorithm and response times are not really anything to write home about.  Affiliate Window released version 3 of their ProductServe API earlier this year, and it seems to be a bit more robust than the others, although I haven't tried it yet myself.  Again there is still the limitation of the results coming only from merchants within the Affiliate Window network.  Several other affiliate networks also offer product search APIs, but generally the smaller the networks, the less data you will find. Wouldn't it be nice if someone could put all of those networks’ feeds together in one universal search? Enter SkimlinksSkimlinks provides publishers access to over affiliate networks 27 networks and over 15,000 affiliate programs instantly. They handle the commission tracking and the merchant relationships. And they've been working on a shopping product search API called EveryFeed. EveryFeed does exactly what its name says, attempts to aggregate products from merchants feeds across every affiliate network in their system.   This sounds perfect right?  Unfortunately, not quite yet.  After testing this for a few days I was disappointed with the results.  A majority of the results were coming from Amazon, and it seems like they don't nearly have enough merchants' feeds coming into the system yet.  I expect this to get better over time, especially with the acquisition of Atma Links earlier this year. Ahh, so close, but back to the drawing board for now. Up to this point a problem shared between both price comparison APIs and affiliate APIs is that the products returned are only from merchants with relationships to the API provider.  For example, Shopping.com is only going to list results from merchants where it is has relationships and is generating revenue from those Merchants. What about Google? If generating revenue directly from referred product sales (or leads) isn't important in your business model, you can look elsewhere, specifically to Google. Google’s Search for Shopping API (formerly Google Base API) allows for searching a very broad range of e-commerce websites.  This is because on the business-side, Google allows for any retailer to submit their product data, for free.  With the Google Affiliate Network, it seems like a perfect fit for getting developers to use their Shopping API, and combining affiliate tracking when applicable, but there isn't a clear vision coming from Google.  I've been following the Google Shopping Group since the beginning of the year, and haven't seen much progress.  By limiting developers to 2,500 API calls a day, clearly they don't want people creating white-label shopping sites using their API, and it actually makes it relatively unusable in any real-world application.  One example is the web application called Lockboxer which uses the Google Shopping API to allow you to check prices on stuff you own or want. Finally, a major drawback of affiliate product search APIs versus shopping comparison APIs is what I'd call “productization”.  That is essentially the grouping together of offers from different merchants for the same product based on a universal product identifier such as a UPC, EAN, GTIN, or combination of Brand Name and MPN (Manufacturer Part Number).  Shopping comparison engines offer this capability, while the affiliate search APIs do not, with the exception of Google's shopping search which has enough flexibility in its API to allow for grouping product results by the GTIN.

Search Directly from E-Commerce Retailer APIs

Amazon, Best Buy, Zappos, and several other popular retailers offer APIs to their product catalog directly.  The upside here is that you can dig a little bit deeper into their database, and pull some extended information out, all while getting a bit fresher information. Plus these retailers (mentioned above) all include pre-formatted affiliate links (how you track your referrals) built in for you, so you don’t have to hack-together your affiliate links. For example the Zappos Peer Pressure Mashup shows high resolution product images that are available through Zappos.com, and allows users to query results by recent sales and by state.   If you have some time to kill and are a shoe or fashion fanatic, this mashup might be worth looking at.  It’s almost like a screen saver for the impulse shopper. Best Buy’s Product API is also robust and returns up to 70 attributes and includes pricing, availability, specifications, descriptions and images for nearly a million current and historical products, even the ability check product availability by local store.  Getting the data directly from Best Buy is definitely closer to "real-time" then from a 3rd party API.  As an example, local shopping engine Milo uses it to pull real-time local product inventory (pictured below). Best Buy takes this a step further with its Commerce API that it launched last year (currently invite-only).  The Commerce API allows end-users to complete the entire transaction process without leaving the application.  This is a big step into seamless transactions for end-users, and heavenly for publishers who can potentially lose sales due to things like affiliate cookie overwrites or expirations. The general drawback with any direct retailer API is that you are limited to only what is in that merchants' product catalog.  Note that Amazon’s product catalog is enormous, and it includes products offered by their Marketplace shops.

Concluding Thoughts

While getting setup and implementing an API can be done in just minutes, becoming reliant on an external data provider as a business model is a serious investment.  Taking the time to research your API provider is a critical setup in determining the future of your business.  Is the API stable?  Is the response time fast enough for your application?  Do they have an SLA?  These are things to consider as a consumer of any API. There isn't a perfect Shopping Search API solution yet, but a variety of options do still exist.  It comes down to your preferences, business goals and the needs of your application users.  Do your homework, and weigh all the benefits of your API provider before making a choice.  But hurry, the holiday shopping season is quickly fleeting!

Be sure to read the next API article: "Mobile Engagement" Platform Adds Revenue Analytics, Engagement Quality