Recommendation engines are a way for content owners—such as merchants, marketers, and publishers—to present the most interesting content to each customer at each step in the interaction. Recommendations were popularized a decade ago by Amazon's famous "other people who looked at this bought that" style of recommendation. Today, recommendation solutions are available from a variety of sources, including software-as-a-service providers such as Loomia.

If you are in media, entertainment, ecommerce, online services, or online research and looking for a recommendation solution Loomia should be on your short list.

Loomia's focus is recommendations and personalization for media, entertainment and ecommerce. Loomia's customers use its services not only to generate recommendations, but also as a means to personalize customer interactions and to provide navigation by offering visitors the next items to engage with.

The key strengths and differentiators of Loomia are the breadth of its market; its high-profile client base; the breadth of the recommendation types, goals, and metrics its solutions support; and Loomia's customization.

OVERVIEW OF LOOMIA

Loomia

Loomia was founded in 2005. It is headquartered in San Francisco California; it has fewer than 20 employees and growing, supporting 15 customers and 35 sites.

Loomia's recommendations are delivered as a service. Its clients are primarily media, entertainment, ecommerce, online services and research. Clients include Wall Street Journal, ABC.es, Thomson Reuters, Time, Harvard Business Review, Travelocity, Fancast, DailyCandy.com, BlogTalk Radio, Audible, and Panasonic.

Loomia Products

Loomia has a family of products from which its solution experts select the ideal balance of capabilities for each client. Behavior-based Similar Item Recommendations and Personalized Recommendations represent the base Loomia technology, and all solutions comprise one or both. Other products are added as needed. The range of products includes:

• Behavior-Based Similar Item Recommendations. Recommends similar content or item to anonymous user based on behavior; API or JavaScript version. Behavior-based recommendations are enhanced with textual and metadata information.

• Personalized Recommendations. Behavior-driven recommendations for a known user; API or JavaScript version Behavior-based recommendations are enhanced with textual and metadata information targeted for that specific user.

• Related Items/ Text-Based Recommendations.Recommendations based on the textual or metadata similarity. This is available as an API or JavaScript version

• Social Recommendations. Shows users what their friends recommend on Facebook; other networks are under development.

• Most Popular Recommendations. Recommends most popular item or content

•  Search Recommendations. Employs a visitor's search keywords to create recommendations across the site.

•  E-mail Recommendations. Integrates product and content recommendations into email campaigns.

• Network Recommendations. Recommends items from other sites or catalogues across a client's network of sites, by syndicating the data involved.

• Video Recommendations. Supports discovery and viewing by selecting the most engaging video; can mix different media types in recommendations.

EXAMPLE: XFINITY.COM

Xfinity TV, an iN Demand storefront, is a site for locating video to buy, rent, or watch. Loomia's recommendations identify and present the most relevant content to users during search, shopping, and buying. Loomia customized a blend of contextual, behavioral and social data inputs to generate the video recommendations. See Illustration 1.

SUMMARIZING THE RECOMMENDATIONS EVALUATION FRAMEWORK

Recommendations are hot, solving problems from order size to search to personalization. My guess is that recommendations will be ubiquitous within the next three to four years. I'm always optimistic on these guesses, but since recommendations are widely available as a service, rollout can be very swift. Through customer interviews and research, we have identified the requirements and evaluation criteria. These criteria are set forth in our evaluation framework¹, which we will be using in 2010 to analyze a handful of the leading recommendation solutions, culminating in a detailed comparison.

The framework describes requirements in seven categories: guidance and advice, recommendation structure, managing recommendations, integration, operations, vendor's development and maintenance, and product and company viability. The evaluation requirements are presented in Table B.

LOOMIA PERFORMANCE AGAINST CRITERIA

Bottom Line

If you are in media, entertainment, ecommerce, online services and research, Loomia should be on your short list. Table A summarizes our evaluation of Loomia recommendation solutions in each of our evaluation categories.

Strengths

Loomia's key strength is the breadth of its market and its high-profile customers. A third of its customers are in media and another third are in online services and research, demonstrating Loomia's competence in recommending deep content. The range of recommendation types, goals, and metrics provided by Loomia's solutions exists to support the range of clients' businesses.

Loomia deploys a broad set of recommendation types, blending behavior, social, textual and metadata elements in its algorithms, controlled by rules, filters, map sets and weighting. The resulting recommendations are evaluated and optimized by Loomia's self-learning feedback loop, a step which also ensures recommendations are fresh, not repetitive.

Loomia is very focused on client goals, and will customize algorithms, recommendation strategies, reports, and insert customized functions into the recommendation creation process to meet those goals.

Weaknesses

Loomia meets its clients' needs by studying their goals and then customizing its algorithms, reporting, and integrations as needed to meet clients' goals. This is great for clients, and normal for a company starting out. But these tasks take resources, and it's not a scalable structure. It's time for Loomia to turn more of these tasks over to clients and partners.

Loomia's testing is A/B, rather than multivariate, a weakness against competitors. Since the Loomia account team will set up the testing and will create as many tests as needed, the lack of multivariate testing probably won't be noticed by clients.

Loomia's recommendations can be controlled by rules, but those rules are not granular by customer segment. This is a weakness that, again, the Loomia team can work around to meet client goals.

Like the other solutions we've looked at, Loomia fails to provide a workflow for review and approval, check-in/-out, or comprehensive e-learning, and technical staff are required to create space for recommendations on pages.

Contact Info:
Loomia Media Corporation
www.Loomia.com

This report continues...

**ENDNOTE**

1) “Recommendation Evaluation Framework, Version 1,” January 7, 2010, by Susan E. Aldrich, http://dx.doi.org/10.1571/fw01-07-10cc

**ENDNOTE**