A recommendation systems goal is to help customers find what they want easily and help them discover products they like but were not looking for. Customer’s satisfaction due to recommendations is becoming more and more important. Understanding how the recommender works and why a product has been recommended builds trust and customer loyalty.

Two types of explanations based on two personalization degrees can be given to build credibility:

1. Explanations Based On How The Recommendation Engine Works:

Sometimes customers are not conscious of that fact that the site they are visiting has a recommender system. By eliminating the typical black box recommendation process and adding transparency (see screenshot below), customers can better understand how a recommender system works and also use the system to improve the relevancy of their recommendations.This benefits both the customer in better recommendations and the retailer in increased sales.

In addition, if the recommender encourages customer feedback to correct the recommendations trust will be increased. These options include messages like: “more like this” or “don’t show me this again”.

2. Explanations Based On Customer’s (Stated or Learned) Preferences:

This type of explanation does not tell the customer why the recommendation are given and instead shows how. Such as “You Viewed” or “You Bought”

These explanations are based on the customer’s profile which includes: their purchases, visits, favorite colors, brands, sizes and tag cloud. The explanations based on the customer’s profile are clearly more personalized than the first type shown above, and may be more effective when helping customer’s make a decision.

In some cases explained recommendations might not be convincing enough. Encouraging customer’s input in ways such as “I own this” or “not interested” helps customers fix the recommendation to obtain better results, or to adapt them to their current preferences.

Retailers Also Benefit:

Some studies have shown how the user can even be manipulated to give a valuation or specific rating for a product. Although giving not very accurate recommendations, a sufficiently persuasive explanation can influence users’ decisions. A good explanation could make the retailer obtain benefit greatly as it can raise awareness and make the user to purchase products more effectively than without a personalized explanation.

Another advantage to the owner of the site, whether retailer or social media site, is that explanations can be fun for the customer. If the customer sees that they can interact with the system, it will engage the customer and increase their satisfaction with the shopping experience. Likewise, this higher interaction with the recommender will produce a greater amount of feedback, which helps the recommender and owner know their customers better.

Have you found value through explanations? Did it lead to a purchase? What are your thoughts? Comment or tweet at us @strandsrecs.

Based on the video below, which was shot on opening day at a new H&M store, it looks like customers are bouncing back and storming the stores literally!

via:ScrapeUp

In a recent Bloomberg story, H&M’s CEO Karl-Johan Persson said the economy will be “a little better” this year and confirmed that Europe’s second-largest clothing retailer plans to open 240 stores in 2010. “2010 will be a challenging year for the economy and somewhat better than 2009,” The Swedish retailer, which sells ladies’ denim jackets for 14.95 euros ($20.32), continues to benefit from customers “trading down” during tough times, the executive said. Sales in January rose 11 percent.

Signs from other retailers such as Walmart, Kohl’s and Costco are positive, hinting at increases in sales in single digits in Q1. It looks like cost conscious customers are ready to open their pocket books and luxury retailers are treading water but not for much longer. We’ll see what the rest of 2010 brings. Until then watch out on your next H&M visit you might need to practice your rush and tackle techniques to get that shirt or pair of jeans you’re looking for!

What do you think will be the outlook for 2010 in retail? Will customers continue to search for stellar deals and continue to hold off on luxury goods? Or is 2010 the start of a rebound where customers will start opening up their pocket books again? We’d love t to hear your opinion on Twitter: @strandsrecs.

We are excited to announce the Strands Recommender Miva Merchant Module. We have spoke with multiple Miva Merchants over the past month and have taken your feedback to create an easy to install module for all merchants.

This is an exciting time, as we are bringing recommendation technology that has been used by multichannel merchants directly to you at an affordable price of $49.99 per month for 50,000 recommendations.

Benefits

• Helps Convert Organic Search Traffic
• Increases Average Order Value & Sales
• Boosts Repeat Visitors and Customer Loyalty
• Decreases Time Spent Manually Merchandising Cross/Up-Sells

Start Your 30 Day Free Trial Now: Download The Strands Recommender Module via our website here.

What You Get With The Module

Personalized Product Recommendations:

In your Home, Product, Shopping Cart and Order Confirmation Pages.

Example of Product Page Personalized Recommendations

Easy To Use Dashboard:

Quick & Easy Install Wizard:

Our automatic wizard will ensure you are up and running in under an hour.

Boost Your Sales By Up To 15%.

Most merchants see a sales lift in the first 2-3 weeks. On average sales increases range from 5-15%. For a merchant making $80,000 a month that means an increase in sales of $2 – 10K per month for a cost effective price of $49.99 per month for 50,000 recommendations. We also have larger plans for bigger merchants.

Start Your 30 Day Free Trial Now: Download The Strands Recommender Module via our website here.

We are excited to announce the Strands Recommender Miva Merchant Module. Our module allows merchants of any size to add personalized product recommendations to their home page, detail page, shopping cart and order confirmation. Stop by our booth to learn more or visit our site to get the module.

WIN A NETBOOK AT THE CONFERENCE:

Visit our booth at the Miva Merchant Conference, drop off your business card and you might win this Gateway LT2104u Netbook at the end of the conference. The Netbook giveaway is only good for Miva Merchants.

In case you can’t find our booth here’s what it looks like

Follow Us On Twitter during the conference for industry tips, updates and news.

You’re probably wondering how can I make my site more “social”, help keep visitors longer and buy more?

Wet Seal Helps You Build & Share Outfits With Others:

Facebook Now Helps You Find Friends Old And New:

Three Ways You Can Increase Engagement & Community:

1.) Update Content Frequently: In order to keep your customers engaged it’s important to keep products and content fresh. Meaning, new content should be released daily or multiple times per day. In our consumer culture with Twitter and new forms of media making real-time content easier, releasing new content helps keep your customers interested and also helps spur new conversations.

2.) Encourage User Generated Content (UGC):
Twitter, YouTube and other networks provide platforms for customers to create their own relevant content around your products, which is beneficial in two ways. It helps generate new content for your site with no cost to you and also it is a great way for your customers to interact with your brand and products boosting their brand loyalty. The only suggestion when using UGC is that you filter the content before using it. Every customer has a different world view, so filtering the content before it is published will save you some headaches.

User Generated Content on Folica.com

3.) Utilize Social Engagement Tools:
The content on your site is only as good as the quality of the community that interacts with it. A few tools that can help encourage interaction on your site include Rating & Reviews and Recommendations. Ratings & Reviews providers such as RateVoice, PowerReviews and Bazaarvoice all have affordable and creative solutions that allow customers to provide feedback about your products or services which spur conversations and keep customers on your site.

RateVoice Ratings & Conversations on Heels.com

Personalized product and content recommendations are a second engagement tool that helps bring your products/content and customers together into a community. This type of community recommendations utilize user-to-user collaborative filtering which is based on this idea of creating recommendations based on other users with similar profiles to yours. These profile’s consist of products and content that your customers have viewed, purchased and indicated they like explicitly “by telling you” or implicitly “by being learned by a recommender, i.e. you wear small and like levi’s”. The beauty of recommendation engines is that they anonymously learn from each customer, build a profile and deliver recommendations personalized to each customer automatically.

Recommendations can help your customers:

1. Find other people with similar tastes

2. Find similar things to those I like the most

3. Discover new things through other people.

All three of the above customer engagement hints should give your site a leg up on the competition. What other things have you found that help connect your customers? Tweet at us @strandsrecs.

As most of you may or may not know: When delivering recommendations to customers different methods can be used.

On one hand we have personal recommendations, that is, those recommendations based on the user’s profile. Which are built from the user’s interaction history on the site and explicitly mentioned personal preferences. On the other hand we have collaborative filtering (CF) methods, which make recommendations based on the behavior of other users whose preferences are similar to those of the target user. However, these methods do not consider how the customers’ purchase behavior may vary over time.

Recommending for the Mainstream:

The mainstream audience of trendsetters and their followers mimic similar behaviors

Let’s use a customer who follows the fashion trends or certain patterns driven by the trendsetters. The behavior shown over successive visits to a site will be very similar to other people. The recommendations based on past behavior of these customers can work, but the recommender is not going to be able to react in time to a trend change if we rely only on this data. Only when one begins to search for items according to the latest fashion trends, will they start receiving new recommendations tailored to their preferences. The collaborative filtering, in a way, could anticipate that moment, because there’s always someone who has already begun to follow the new trend, and has already made purchases or have visited product pages that the target user will want soon. The store owner has an important role here too. A good recommendation engine is aware of emerging trends amongst products, and automatically adjusts business rules so that each person receives recommendations based on the new lines of products without having to wait for the users profile to “react” to the trends.

Outliers Can Put A Wrench In The Recommendation Equation:

What would you recommend for her?

However, recommender systems can sometimes fail for people whose tastes fall out of the mainstream, or have a very narrow or too wide range of tastes. Changing tastes and preferences are usually not sudden but morph over time. The recommendations for these users based on their profile and their interactions are inherently harder to create. In these cases current data should be given greater weight than historical data, focusing on the changing preferences of these users who follow their own way.

Collaborative filtering is more complicated. Because there is little correlation between these outsiders and the rest of the community and there is fewer users of that type lacking matches between them. This situation makes it difficult for the recommender to predict which will be the customers next preference. But these kinds of customers are not a hopeless case. The CF could work if the recommendation engine would only consider a few customers that may be similar to our target, and they are usually his/her friends. If the recommendation engine can automatically select smaller groups of people to learn from and to trust, they will receive more accurate recommendations that are tailored to their tastes and their possible evolution. This is what we call Not Anonymous Collaborative Filtering (NACF).

What’s Best? Both Methods or User Choice?

We usually receive different types of recommendations at the same site, depending on the location, based on different data sets or based on a mix of them. Some of them are going to work better for you, than others. The ideal case for both situations previously shown is that the user could choose what sort of recommendation they want, based on what other users bought, only showing related items, similar or complementary, not showing more one type of recommendation or only what my friends recommend me. This way, the customer can choose what to use and thus enhances both the learning of the recommender according to your changing preferences, and the results obtained by the user and the store.

Is this then a failure of the recommender? Should they satisfy everyone equally and there is a need to refine the algorithms for finding the right recommendations for everyone? or do we have to admit that it doesn’t always work? Would it improve if the recommendations based on smaller groups of users, and consistent with the customer? Or is the point that the customer’s should choose what kind of recommendation they wish to receive? Post a comment below or tweet at us on Twitter @strandsrecs.

As many of you know now Apple just launched the long awaited iPad tablet device about an hour ago. Watching the blog updates from the keynote event it looks like Apple’s new device has extreme potential for retailers.

Here’s the stats:
9.7 inch display
Touch screen
1.5 pounds, .5 inches thick
1 ghz apple A4 chip, 16 to 64 GB flash storage
WiFi 802.11n
Bluetooth 2.1 +EDR
Speaker, Microphone, 30-pin connector
Accelerometer & Compass
10 Hours of active battery life, 1 month of battery life on stand by
Earth-friendly!

Here’s possible potential:

In-Store:

Store representatives could enter a few details about you into the iPad or swipe your frequent loyalty card and help you find products you will like. A recommendation engine would be a perfect addition to this process searching profiles both online and in store to recommend relevant products. This application would also help the store rep know the customers, favorite, sizes, brands, etc making the shopping experience for the customer seamless.

Shopping & Sharing On The Go:

Mobile commerce is slowly picking up, think of the possibilities that a larger screen provides. Not only can customers shop from retailers websites on their iPad’s while on the couch and lounging around. Sharing features such as Facebook and others will add a social shopping component that will allow shoppers alike to take their favorite new clothing item, device, etc and show their friends.

Shopping Apps:

With 10 inches of surface area the iPad provides a great deal more space to create innovative apps for shopping. iPhone apps are the latest craze among retailers, many apps that have been released are basic apps that display the retailers product or catalog and allow a purchase. The new SDK and larger screen provide amazing potential for very interactive shopping apps.

Consumer adoption of the iPad will be a large determinant of the potential of new shopping, but if iPhone and iPod are any signs of Apple’s ability to jump from early adopters to the mainstream I think the iPad will be a big hit. What are your thoughts? Do you think the iPad will make an impact on the way we shop today? Comment below or tweet at Strands Recommender on twitter @strandsrecs

In most instances, sites do not incentivize users enough or provide a true purpose to how the time and information provided by the customer will help. The result is typical users just move on.

Why is this you may ask? generally it seems that the recommendations and their mechanisms are a secret, but an open secret. The user is not conscious of the fact that the site they are visiting has a recommendation system and does not bother to interact with it. If the website adds feedback mechanisms, where they are asked to make a review, or votes by stars, hands, and an endless list, the user loses interest, without realizing that they are loosing a good opportunity to get something back.

One site that breaks the mold is Amazon. The question is why do people bother to vote in Amazon and no where else? Because they know how it works and know the results they will receive. They know that the recommendations help them make educated purchases.

The figure below shows sites where users felt recommendations were a beneficial element, and almost required part of the user experience. It is interesting that customers know that recommendations are useful in e-commerce, but they don´t know about the benefits of recommendations in web searches, entertainment and media, or even social networks.

By:UX Expert Afshan Kirmani
Published: March 23, 2009

Every day we want more things fitted to ourselves, customized to suit our needs, but we are generally lazy as humans and don´t feel like doing anything to get it. Why don’t sites follow the Amazon model and openly say or infer that they need to know your tastes?

Implicit behavior on a site is very useful, and surprisingly it can be tracked and understood. Some recommendation systems such as Strands Recommender are able to learn a customer’s brand preferences, colors, sizes and context automatically as they interact with the site. The goal is making a customer’s life much easier and the experience seamless.  Explicit feedback such as ratings, reviews, etc are very powerful and  can direct you toward what you want faster. But users need to know how to use explicit feedback to obtain better result from it. A combination of implicit or learned feedback, when paired with implicit feedback can significantly improve the speed of their experience and ease of finding what they want.

Customer’s general stubbornness is hard to beat because their time is valuable, but with good motivation and explanations, even the most stubborn will provide feedback.

What do you think do websites do a good enough job of explaining why their feedback mechanisms are important? Or are we as consumer’s just averse to providing feedback without some incentive?  Or both? Send us your comments below and tweet at us @strandsrecs.

You hear the buzzwords every day in the office, on the street and in the latest hip hop and country songs: Blog This, Social Media That, Retweet It and Follow Me. What does it all mean and is Twitter applicable to your growing E-Commerce business?

The short answer is YES.

Why Twitter?

1.) Provides transparency inside your company:
When people buy from you online or in your store, they believe in the story you are trying to telling them, that is why they are buying from you in the first place. Most likely they will never see who is running your company, the employee’s that make the magic happen, how your product was made and so on. Customers want to know that there are real human beings behind the curtain, not robots. Twitter is a perfect mechanism to open a dialog with your customer base and show them your true personality.

2.) A feedback mechanism for your customers:
A customer’s worst nightmare is to dial the dreaded 1-800 number and speak with someone who has no connection to the product, company or their native language.Twitter provides a perfect mechanism for your customers to submit feedback and for you to solicit feedback on products, customer service

3.) An open forum for customer conversations:
Whether your business is big or small it’s hard to stay on top of customer chatter about your products. Not only is Twitter great to solicit feedback from your customer’s, it’s also a great tool to listen to what your customers are saying about your product. Listen in on the conversation and you might find a new product, feature or customer you may have missed.

4.) A new targeted marketing channel:
Twitter is a perfect place to build a new community of followers that you can direct special offers to. When you build your following, reward them with special offers only for Twitterers. The beauty of Twitter is that your followers can forward specials to their friends growing your customer base.

5.) Improves your Google searches:
Twitter is now crawl able by Google. If someone talks about your company or products on Twitter, when your customer’s do a Google search, tweets will now come up in the search results.

How to get followers:

1.) Email newsletter: Easily add a “Follow Us On Twitter” link in your emails. Another good way to get a higher click through rate is to add “For Special Offers”, or if you want to take it further give away a gift card to the 500th follower, etc.

2.) Your website: Adding a banner to your home page, or even a form for a customer’s Twitter name during sign up/registration is a great way to acquire information.

3.)Other social media: Add a Twitter link on your Facebook, Blog or Youtube page.

Tweeting Tips:

1.) The more the merrier: If you do decide to tweet, which I highly recommend you do, make sure you tweet often. Twitter is just like a blog, if you only tweet once a month, customers will loose interest. Tweeting something valuable daily is suggested.

2.) Add value: Make your tweets valuable. While it’s cool to know Shaq is taking a shower, your customers want to hear things that are meaningful and either pull them closer to your brand or start a dialog with them.

3.) Pics & videos: Link to pictures and videos. Twitpic and twitvid are great tools that allow you to link to pictures and videos from your Twitter account. Both video and images provide an even higher level of transparency with your customers.

4.) Multiple accounts: As your company grows, you may want to create multiple Twitter accounts, e.g. one for your customer service, another for possibly the CEO, CMO, etc. If each person has an interesting story to tell and generally they are inaccessible to the general public, a Twitter profile will help increase the transparency into your company. Tony Hsieh’s the CEO of Zappos, most people will never meet him in real life, but he always responds to his tweets.

A Few Noteworthy Twitter Pages:

Metroparkusa: www.twitter.com/metroparkusa
Heels: www.twitter.com/highheels
CycleGear: www.twitter.com/CycleGearsales

A Few Noteworthy Twitterers:

Express: http://twitter.com/ExpressLisaG
BestBuy: http://twitter.com/BestBuyCMO
Zappos: http://twitter.com/Zappos

Embrace social media and let it work for you. Feel free to comment below and add in your feedback on how successful Twitter has been for your E-Commerce site. Tweet at us @strandsrecs on Twitter too. Stay tuned for more tips to help your small business grow in 2010.

Not only is Natal a very cool concept, it also helps to solve a few problems many online shoppers experience as well as encourages community based shopping.

Sizing & Fit Problems Solved

There are a few issues with shopping online if you haven’t bought from a store before. Fit and look. Natal solves both of these problems by mapping your body and matching it with the appropriate size online. In addition to ensuring a perfect fit, Natal helps you pick the perfect look by allowing you to try on an outfit and see how it looks in a 360 view on your own body.

Recommended By A Friend

Another cool concept in the video is the ability for your friends to suggest clothes for certain occasions as well as build outfits. This provides the opportunity for retailers to upload their catalogs and allow shoppers to pick and chose outfits and share them. Also with Xbox360′s recent addition of Facebook to the console’s offering I can see some pretty interesting opportunities here for community based recommendations.

With a user base of over 39M users, the Xbox360 console is morphing into a family entertainment device that holds some great promise for some amazing shopping experiences. What do you think about Natal? Would you use it to shop? What would you like to see in a shopping experience on a console? Comment below or tweet at us @strandsrecs.

Here’s a sneak peek at what Natal will look like when its released later this year:

The US News 2009 Retail Chain Winners & Losers are in, drum roll please… topping the list of winners was Aeropostale with a sales increase of 10% and coming in dead last,  Abercrombie & Fitch with a sales decrease of 19% for the year. Others in the winners category were Ross Stores, Costco and surprisingly Saks and Nordstrom’s. Losers sharing the bench with A&F included American Apparel, Hot Topic and Wet Seal.

Many of the retailers on the list who are losers in 2009 were cleaning house a few years ago, the question arises: Are some retailers not focusing enough on the customer experience?

Aeropostale: has a much better grasp of their customer’s experience on their site. It’s very easy to navigate, the marketing messaging to the user is clear and they have a generic merchandising approach that generally works for each item viewed. One lacking functionality on the site is ratings & reviews and also any form of personalization.

A&F: I have to admit that I worked at A&F for a year and value their store experience and atmosphere they create, but their online experience leaves much to be desired. The site is cluttered with branded marketing messages, their detail page, search and checkout pages are very basic, lacking reviews, previously viewed, any form of merchandising and also personalization.

While this is only a basic sample of the retailers who were involved in the report, it becomes evident that a lack of focus on the customer experience on your website can lead to diminishing sales and unhappy customers. Stay tuned for our weekly blog postings on ways to get your website in top shape for 2010, starting next week!

What are your thoughts? Are retailers losing touch with the customer experience? Who are some other retailers who need to step up their game? Post your comments here or tweet at us @strandsrecs.

After looking at numerous recommendation providers, Metropark chose Strands due to our quick implementation, easy to use merchandising interface, detailed reporting and also our reasonable pricing.

Currently Metropark is utilizing Strands Recommender on it’s product detail under “May We Suggest” and will be adding email recommendations and shopping cart up-sells soon:

Metropark is one of the many retailers who have taken advantage of our Standard Pricing plans, which start at $249/month and provide a similar feature set to our premium customers at an affordable price. It’s a great way to improve your merchandising, sales and learn more about your customers buying behaviors. As your revenue grows, you can always upgrade to our premium service.

To continue their growth, Kedin was looking for new forms of differentiation and personalization was key to their strategy. Kedin was impressed with Strands ability to drive increased engagement and customer loyalty on similar sites such as 11870.com and Blip.tv. With these goals in mind, Kedin was able to implement Strands in two functional areas of their site:

1.) Personalized Events Page: Once a user creates a profile, Strand’s automatically begins learning about their tastes and preferences and makes relevant suggestions.

2.) Related Events: Strands learns from all visitor’s behavior on the site and suggests related events in their area.

If you’re visiting Spain or already live there and are looking to bring out your inner socialite visit Kedin.es. Also visit our website for more information on how recommendations can dial up your social media site.

The results aren’t quite in yet, but ShopperTrack will release their Black Friday report Saturday morning and we’ll share our ideas soon after. For the economy’s sake, we hope the weekend numbers are high.

Here are a few links to our Strands Recommender clients and the great holiday promotions they are running:

• Free Gift Wrapping at TravelSmith
• 5 days of deals at SkyMall
• Free 3 day shipping on all orders over $200 at Footwear etc
• Free shipping on orders over $49 at Luggage Point

Winner Outstanding Website, WebAward 2009

Brian Walker, @BKWalker, of Forrester Research, recently shared the B2B eConference Trends and Solutions Market Overview. The presentation highlights the growth drivers and range of models in B2B ecommerce plus the evolving B2B site. Check out the full details at the above link. Since many of our clients are preparing for 2010, we thought we’d share Brian’s 5 Key takeaways (with our Strands spin on it).

1. Design your site for the customer.

Consider the end user who needs your product, information or service. Identify their values and search path on your site. Think of Amazon or Apple and follow their website schema when designing your own.

2. Don’t design your site for the sales manager.

Allow your site to sell the end product that relates to the end user. The sales manager will understand the value in your product when seeing the value it serves to their customers. Think of it like this: they are focused on sales, so show them (indirectly) that you are too.

3. Employ familiar UI principles from online retail.

Keep your website structure, simplicity and visibility in mind, but also position your site as a one-stop-shop. Allow someone to easily navigate and identify all the possible things they can purchase. Artistic navigation schemes will decrease time spent on your site and increase the bounce rate.

4. Focus on marketing and merchandising tactics.

There is no better way to cross-sell and up-sell than with Strands Recommender. By showing what other items people have purchased and what items they might be interested in, you will increase your average cart total by 20%. Shamelessly, we’d like you to find more information about the Strands Recommender by clicking here or below, or to the right…

Strands Recommender

5. Don’t let internal business processes, systems, and sacred cows drive the online experience.

What processes can you streamline to make it easier for a customer? It doesn’t matter how it’s always been done, what matters is how it’s currently being done in your customers opinion. Innovation and modernization is key.

For further discussion of key trends or assistance with recommendations, please feel free to contact the Strands Recommender Team.

The number of online services providing users with real-time recommendations has increased exponentially over the last few years. Recommender Systems that were originally only accessible to a limited number of high-tech companies are now widely available through a growing number of both technical choices and vendors. The acceptance, however, of automatically delivered recommendations by users depends on numerous factors that go far beyond the algorithms that constitute the major focus of researchers.

Over the past 10 years, Strands has been creating and operating recommender systems in a multitude of domains, ranging from digital music to fitness plans and personal finance management, and in a multitude of business settings ranging from lightweight integrations to highly-coupled integrations within secure bank environments.

As summarized by Neal Lathia of MobBlog, below are the top 10 lessons we’ve learned developing, deploying, and operating real-world recommender systems:

• Lesson 1. Make sure a recommender is really needed! Do you have lots of recommendable items? Many diverse customers?… also think Return-on-Invesment… a more sophisticated recommender may not deliver a better ROI.
• Lesson 2. Make sure the recommendations make strategic sense. Is the best recommendation for the customer also the best for the business? What is the difference between a good and useful recommendation? Good recommendations vs useful recs; obvious recommendations may not be useful; risky recs may deliver better long-term value.
• Lesson 3. Choose the right partner! Select the right rec vendor vs hire some #recsys09 students. If you are a big company the best thing you can do is organize a contest.
• Lesson 4. Forget about cold-start problems (!) …. just be creative. The internet has all the data that you need (somewhere…).
• Lesson 5. Get the right balance between data and algorithms. 70% of the success of a recommendation system is in the data, the other 30% in the algorithm.
• Lesson 6. Finding correlated items is easy but deciding what, how, and when to present them to the user is hard… or don’t just recommend for the sake of it. Remember, user attention is a scarce and valuable resource. Use it wisely! … dont make a recommendations to a customer who is just about to pay for items at the checkout! User interface should get at least 50% of your attention.
• Lesson 7 Don’t waste time computing nearest neighbours (use social connections)… just mine the social graph. Might miss useful connections?
• Lesson 8 Don’t wait to scale!
• Lesson 9: Choose the right feedback mechanism. Stars vs thumbs …. the YouTube problem. More research on implicit and other feedback mechanisms is needed. The perfect rating system is no rating system! … focus on the interface.
• Lesson 10 Measure Everything! … business control and analytics is a big opportunity here.

Keynote Takeaway

Think about application context; Focus on interface as much as algs; be creative with startup data. … the UI needs to get the lion’s share of the effort (50%) compared to algorithms (5%), knowledge (20%), analytics (25%).

Strands Recommender welcomes it’s newest client, CostumesInc.com, in time to offer recommendations for Halloween.

CostumesInc.comquickly deployed recommendations powered by Strands Recommender throughout their site just in time to offer recommendations during their busiest time. The recommendations are offered on the product and shopping cart pages.

Additionally, Costumes Inc. uses Strands recommendations to power each of its tabs for accessories, wigs and hats, makeup and related items shown on individual product pages. Click here to get your Halloween Costume!

We are pleased to invite you to participate in this first workshop on Recommender-based Industrial Applications, to be held on Sunday, October 25th, 2009 in New York City, USA, in conjunction with The 3rd ACM Conference on Recommender Systems (New York City, October 22-25, 2009).

Complete workshop schedule and details can be found here.

Recommender systems help users deal with large amounts of information by assisting them in their decision-making processes.  Recommender systems have been applied very successfully in a variety of domains, and one of their most exciting characteristics is that they combine interest from industry while posing very interesting research challenges, having significant impact on end user experience, and on business.

Motivation and Topics

Recommender-based industrial applications have hit the mainstream: they are being used today by millions of users in a variety of domains (e.g., film, music, book purchases, travel, advertising, etc.), having significant impact on end users and contributing to the generation of millions of dollars of revenue in some industries.

In addition, the explosion of social network websites, on-line user-generated content platforms, and tremendous growth in computational power of mobile devices are generating incredibly large amounts of user data, and an increasing desire of users to find what best matches their personal interests.  This is creating new technical challenges as well as new business opportunities.  However, in spite of significant progress in the research community, there are still important gaps in making accurate and timely recommendations in industrial applications.

This workshop aims at bringing the gap between academic researchers and industry practitioners on the area of Recommender Systems.  We are interested both in research work that faces real industry problems, and in industry cases that create research challenges.

We aim to discuss problems the industry is facing in real-world applications when applying recommendation and personalization techniques.  The goal is to bring together top-level researchers from academia and recommendation practitioners.  Ideally the workshop should serve to establish new collaborations and projects involving academic researchers and companies.

Topics of interest include industrial applications and research challenges related but not limited to:

• case studies of recommender system implementations and deployments
• personalization vs. content-based vs. collaborative-based applications
• evaluation and user studies of recommender systems
• recommendations in the long tail
• user preferences and recommender systems
• social and cultural impact of recommender systems
• security and privacy in recommender systems
• sparsity and lack of data in recommender systems
• context-aware and ubiquitousness in recommender systems
• user interfaces and usability issues in recommender systems
• scalability in large recommender systems
• new business models related to recommendation and personalization
• search vs. recommendation

Workshop Chairs

• Dr. Alejandro Jaimes, Telefonica Research
• Dr. Marc Torrens, Strands

Please address questions to recsys [at] strands.com.

Online Merchants Implement Strands Recommendations
to Enhance Customer Experience and Increase Sales

SHOP.ORG SUMMIT – LAS VEGAS, Nev., September 23, 2009 – Strands Recommender a leading provider of personalization and behavioral recommendations announced today that TravelSmith, a Cornerstone Brands company, Body Glove and MotorcycleSuperstore.com (#223 on Internet Retailer’s Top 500) have selected Strands to employ a new level customer experience throughout their websites.  The sites implementation of Strands Recommender will provide automated cross-sell and up-sell opportunities for visitors of the site based on real-time behavioral data, historical data, and merchandising.

“We chose Strands as a partner because we were confident in their ability to help us cross-merchandise our products appropriately, while driving our online conversion rates. The ultimate result was an enhanced user experience and lift in sales”, said Jeff Miller, Director of Interactive Marketing and E-commerce, Body Glove.

Strands Recommender increases conversion rates and average order values by continuously learning about the behavior across all sections of a website.  This flexibility allows for the most relevant and timely recommendations for each site visitor.  “We are happy to become the partner of choice for a growing number of Internet Retailer Top 500 merchants and look forward to rolling out large-scale infrastructure enhancements and innovative features in the coming weeks,” said Jesus A. Pindado, VP and General Manager, Strands Recommender.

About Strands

Strands is the leading provider of social recommendation technologies that deliver personalized shopping experiences and targeted product recommendations for online shoppers. The Strands Social Recommender™ learns consumer tastes and preferences as they evolve over time and is typically used with email marketing solutions, search engines, shopping carts, and e-commerce platforms.  Strands is headquartered in Corvallis, Oregon and has offices in San Francisco and Madrid, Spain.

To learn more about Strands products and services, visit us at http://recommender.strands.com or call 415.398.4333.

The 45-min webinar discusses the best practices for optimizing implementation with the Strands Recommender and shares the experience of customer Nimli, an online marketplace for eco-brands as told by CEO/Founder, Rajat Arora. The key points summarize Strands best practices for:

1. Uploading your product catalog
2. Tracking user behavior
3. Installing recommendations for multiple use cases

These guidelines were followed by a walkthrough of Nimli’s current implementation on the product detail and shopping cart pages and ongoing improvements to them. In the next phase, Nimli will use personalized Strands recommendations in the purchase confirmation and abandoned shopping cart emails. Seventeen percent of the revenue is attributable to Strands recommendations, along with a CTR of 40-50% on the recommendations. Rajat also shares his key takeaways and best practices such as treating recommendations as a continuous process for improvement and leveraging the customization features. Learn more today by creating an account at recommender.strands.com.

tunesBag uses Strandsâ„¢ flexible algorithm configuration to influence how the recommendations are calculated. Based on the current track that the user has selected, other tracks with the same tags and artists are given high priority resulting in similar music recommendations. Through Strandsâ„¢, users can also discover new artists that share similar styles with the tunes in their music libraries. While users can get recommendations from day one, the quality of recommendations improves with continued use as the user behavior starts to influence the results. Strands Social Recommenderâ„¢ builds on listening histories and evolves with the current behavior ensuring that recommendations are relevant and offer consistent content discovery. Unlike recommendation features in other music clients, Strands Social Recommenderâ„¢ is able to truly personalize recommendations by filtering out tracks that are no longer relevant, ie, user gave track low ratings, already owns track or added track to favorites.

We’ve been busy at Strands Business Solutions, working with some really great partners.

We have some insightful and interesting content about our technology, as well as some case studies up on our new site, and you can check them out here:

The secret sauce: What Strands technology is really all about
http://recommender.strands.com/tech_social.sbs

The Strands recommender API for developers
http:/recommended.strands.com/dev/

Skymall: Personalized Recommendations drive double-digit sales lift
http://recommender.strands.com/ecommerce_case.sbs

IHeartMovies: Personalized recommendations result in double digit CTR (Click through rate)-lift
http://recommender.strands.com/social_case.sbs

We’re also out and about meeting some great people at conferences this month and next, and would like to introduce ourselves!

Here’s where we’ve been and where we’ll be:

eTail West 2009 in Phoenix, AZ
http://www.wbresearch.com/etailusawest/

eTail has been around for 10 years and it is the premier event to discuss e-commerce small and large.

Jane Kwak and Scott Doan were there representing the team and would love to follow up with you.   Email us at sales-sbs@strands.com

South by Southwest: Interactive (SxSWi) – March 13th -17th
http://www.sxsw.com/interactive
Austin, TX

We’ll be making a splash in Austin this year at the Superbowl of internet technology conferences.   If you’re planning on going, be sure to make it to our panel session:

“You may also be interested in…”
How recommendation affects our digital lives.  An interactive discussion about recommendation technology, implementation and analytics.  How can it help your site or app?  Come find out.   Sponsored by Strands Business Solutions.

Drew Olanoff of Strands will be moderating, and the panelists include Mike Hudack from Blip.tv, Trevor Legwinski from Strands Business Solutions, and Alex Hillman, founder of IndyHall.

The session is on Monday the 16th at 5pm, come out and join the discussion!

11870.com, the European equivalent to Yelp, is used for people to save, share and keep track of the places and services they like around the world through reviews, pictures and videos.

From now on, 11870.com will offer its users suggestions of places and services based on users interests and behavior. 11870.com will use Strands Social Recommender, which delivers advanced social recommendation and personalization technology, as an easy-to-use Strands hosted service.

The recommender works in several areas:

Personalized recommendations: when user log in, they will find a list of places and services tailored just for them.

Recommendations within a specific location: each city or location includes a “recommendations” feature, where users will find personalized recommendations of places and services within that region.

Recommendations inside categories: when you visit a category such as “hotels in London” or “Spanish restaurants in San Francisco”, the recommender suggests each user a number of services that might fit with his or her interests.

Similar services: when you go to a product page, the system suggests a list of similar services, broadening the list of options for you to chose from. This is available both for registered and non-registered users. For example, if you chose an Italian restaurant, the recommender will suggest similar Italian restaurants.

Strands is especially happy to be involved in this project, as 11870.com is one of the most successful European start-ups. Watch them carefully!

You can read more information on the 11870.com blog.

Thanks to everybody who attended Recked, the informal event for engineers interested in recommender systems. It was awesome. And thanks to Robert Gaal for putting the event together.

You can find the three presentations (Strands, Wakoopa and Reccoon) below.

Launching in early November, Strands Social Recommenderâ„¢ will be powering personalized product recommendations for SkyMall’s extensive online store of over 7,500 gifts, gadgets and home innovations—from the unusual to the practical—just in time for the holiday season.

Strands Social Recommender will provide an enhanced shopping experience to SkyMall customers, providing highly relevant recommendations to help them discover products based on their taste and intent.  Analyzing thousands of customer touchpoints, Strands Social Recommender “learns” what SkyMall shoppers are looking for, and then dynamically recommends tailored suggestions, while they browse.

We are thrilled to be working with SkyMall!

Business owners and charities shop at DollarDays to find deals on small business-sized orders for more than 30,000 consumer products–from toys and household décor to apparel, electronics and seasonal merchandise.

Strands will also offer a free 30-day trial of Strands Social Recommenderâ„¢ to DollarDays’ 1.1 million retail customers/members.  Starting October 1st, members will be able to trial the same recommendation and personalization technology powering www.dollardays.com without any setup fees and implementation costs.

DollarDays adds 900 new members each day that include a variety of retailers from convenience stores, medium-sized brick & mortar retailers, shopping mall kiosk companies, discount stores, seasonal shops to large established companies.

Strands Social Recommender™ is a SaaS plug-and-play recommendation engine that can be easily and quickly installed by eCommerce and Media Web sites. The Strands Social Recommender can be installed through the Strands API or by adding personalization widgets to a retailer’s site, and can be up and running in as little as 30 minutes.

We are thrilled to be working with Dollar Days and its network of members!

Developed by BBVA and Strands, BBVA Tú cuentas (You count, in Spanish) is an innovative personal finance tool that allows online users to manage their personal finances better, offering friendly analysis tools to better understand spending and savings. A 100% customizable interface, personalized suggestions, and versions for the iPhone, Blackberry and Nokia phones are some of the unique characteristics of BBVA Tú cuentas, being offered to BBVA customers.

BBVA has 4.1 million online customers performing over 1.3 billion annual online transactions. Strands Social Recommender, a key component in the solution, will help BBVA better understand its customers’ needs and behavior, offering them personalized suggestions of financial and non-financial products.

BBVA Tú cuentas is launched today in private beta with the following features:

- Analysis of personal financial information. BBVA Tú cuentas displays a graphical representation of the customers’ financial health, break-down of their income and expenses, as well as their evolution through time. Charts help customers better understand their finances and see whether they have done better or worse than planned over time.

- Categorization of transactions. BBVA Tú cuentas organizes all transactions into a series of categories, and customers can also use personal tags if they prefer them. Having data well organized can help customers make better financial decisions.

- Comparison tools. Customers can tag themselves and BBVA Tú cuentas will compare them to similar customers. Offering meaningful comparisons with similar customers will help people better understand their personal finances and put them into context.

- Budget Goals. BBVA customers can set up their budget and keep their spending within the limits of what they can realistically afford. Additionally, they can choose to be notified if they happen to go over their set budget.

- Alerts. Customers can configure their own financial alerts and BBVA Tú cuentas will notify clients whenever they cross predefined limits. To round things off, customers can also combine these Alerts with their Budget Goals.

- Just for me. The collective knowledge of BBVA user base allows for BBVA Tú cuentas to offer an automatic selection of personalized suggestions for customers to make better decisions. (1) Practical financial and consumer tips to inform customers before they make money choices; (2) short list of most liked products matching each customer spending patterns; (3) financial products and services that may better address customer needs; (4) relevant facts and insightful aggregate stats based on community behavior.

- Aggregation of accounts. BBVA Tú cuentas allows BBVA clients to keep all of their financial data conveniently in one place. In addition to aggregating their BBVA accounts, customers can automatically add any other bank and utility accounts they may have.

Other key characteristics of the Strands personal finance solution are:

1.- A powerful widget platform enables maximum customization with minimum set-up time. Each of our 20 widgets is a little application that performs a specific feature. Clients can personalize each tab with the widgets they choose, to view the content they need.

2.- Powered by Strands Social Recommender, BBVA Tú cuentas provides users with the best personalized suggestions possible, helping them find new ways to save money and invest.

3.- It comes with tailored versions for Blackberry, iPhone and Nokia (S60) browsers, so that the service can be used on mobile devices just as easily as it can be used on computers.

BBVA is a financial services group with more than $783 billion in total assets, 42 million customers in 40 countries and a market capitalization of approximately $85 billion.

We are thrilled to be working with them in this amazing solution!

The Strands Social Recommender has been proven to increase online sales by up to 35%, by offering site visitors personalized recommendations of new products and content that they might like, in real time.

The Strands Social Recommender can be easily installed through the Strands API or by adding personalization widgets to a retailer’s site, and can be up and running in as little as 30 minutes. More than 100 online businesses have piloted the Strands Social Recommender, with many customers seeing a significant increase in sales.

Key features of the Strands Social Recommender include:

• Cross-product flexibility: Recommendations can be generated across multiple product lines and data models, enabling sites to scale across diverse product, content and service categories.

• Personalized emails: Online businesses can opt-in to offer personalized product and content recommendations via multiple marketing channels, including email.

• Discovery widgets: Businesses can present content, merchandise, and services to users via fully customizable widgets

• Reporting: Business Intelligence reports and visualization tools complement standard Web site analytics to measure the performance of each recommendation. This includes conversion rates, click through rates, and session lift.

The Strands Social Recommender delivers relevant product suggestions to first-time users, as well as offering filtered, personalized results over time to returning customers. These personalized recommendations are based on an individual’s taste profile, but don’t involve asking the user to rate content or build a profile on each site. No personal data is ever exchanged and each user’s interaction remains anonymous and secure.

As we noted above, we will be demonstrating the Strands Social Recommender at the Internet Retailer 2008 Conference and Exhibition in Chicago, today through June 12 at booth # 1216, let us know if you are going to be there.