SlideShare une entreprise Scribd logo
1  sur  16
An Oracle White Paper
March 2011



Next-Generation Product Recommendations
Next-Generation Product Recommendations




Introduction ......................................................................................... 1 
The Potential Value of Recommendations Solutions .......................... 2 
   Avoiding the Risk of Overwhelming Shoppers
   with Too Many Choices ................................................................... 2 
Flaws in First-Generation Recommendations Solutions ..................... 3 
   Inadequate Understanding of Shoppers’
   Interests and Preferences ............................................................... 4 
   Inadequate Reach Into and Understanding
   of the Product Catalog .................................................................... 5 
   Lack of Control ................................................................................ 5 
   Limited Reach Across Channels ..................................................... 6 
   Inadequate E-Commerce and Retailing Expertise .......................... 6 
A New Generation of Recommendations Solutions ............................ 6 
   Better Understanding of Visitors’ Interests and Preferences........... 7 
   Greater Reach Into the Catalog ...................................................... 8 
   Greater Reach Out to Customers ................................................... 9 
   More Control to Refine Selections .................................................. 9 
   Relationship with a Well-Established Provider
   with Retail Expertise and Focus .................................................... 11 
Proven Impact ................................................................................... 11 
   The Positive Experience at Tommy Hilfiger .................................. 11 
The Oracle Recommendations On Demand Solution ....................... 13 
Conclusion ........................................................................................ 13 
Next-Generation Product Recommendations




Introduction

Retailers can enhance the value of their e-commerce business by adding personalized product
recommendations solutions to their online stores. These solutions assess a shopper’s interests
and preferences and then recommend merchandise likely to appeal to that shopper. Retailers
using automated recommendations solutions can more effectively convert visitors to buyers,
increase the value of transactions, and generate more repeat purchases from loyal customers.


Despite the potential benefits, relatively few e-commerce sites have adopted recommendations
solutions to date because of several critical flaws in the first-generation of solutions:


•   The inability to understand shoppers’ preferences and present relevant recommendations


•   A lack of control provided to retailers to manage recommendations


•   Limited reach into and understanding of each product catalog


•   An expensive and lengthy implementation process


•   An inadequate understanding of how recommendations fit into the overall e-commerce and
    retail strategy


A new generation of recommendations solutions that resolve these flaws is now available.
Leading retailers who have adopted these improved solutions have realized greater return on
their investment. Visitors who click on personalized recommendations convert up to three
times more often and spend up to 30 percent more than those who buy without clicking on
recommendations. This white paper discusses the flaws of the earlier automated
recommendations solutions and the advantages of the second-generation solutions. The paper
also illustrates how a leading retail brand has implemented an automated recommendations
solution from Oracle and realized a significant positive impact on the business.




                                                                                                           1
Next-Generation Product Recommendations




The Potential Value of Recommendations Solutions
With e-commerce well into its second decade, the keys to sustained online success are becoming well
understood. Online-only retailers, established bricks-and-mortar retailers expanding online, or
manufacturers looking to sell direct to consumers online must achieve certain objectives in order to
maximize revenue from e-commerce platforms:
•    Attract visitors and convert them into buyers. Retailers must draw as many people to their
    various online shopping channels as possible; present them with attractive, relevant offers; and entice
    them to purchase.
•   Increase the size of each transaction. Just as in-store sales associates effectively cross-sell buyers
    before purchase, successful online retailers must also cross-sell and up-sell merchandise to online
    buyers to raise the average value of each transaction.
•    Generate sales from returning customers. Retailers must build and retain a loyal customer base
    in order to sell more online merchandise, while reducing the marketing costs necessary to attract new
    customers.
In order to achieve these objectives, online retailers must accurately and consistently perform one key
task: understand what visitors want to buy and direct them to the appropriate merchandise as quickly
as possible.
Online retailers must accurately and consistently perform one key task: understand what visitors want to buy and direct them
to the appropriate merchandise as quickly as possible.



Avoiding the Risk of Overwhelming Shoppers with Too Many Choices
Retailers must perform this task for both their in-store and online visitors, although doing so is more
challenging in the online world. Freed from shelf-space limitations, online stores typically present a
much broader array of merchandise to shoppers. This wider selection can make it more difficult for
shoppers to quickly find what they’re looking for—too many items can overwhelm shoppers with too
many choices. Instead of finding precisely what they want, shoppers get confused, lost, frustrated, and
often leave before buying. When shopping online, leaving is as easy as a new Google search or a few
clicks of the back button.
Retailers have largely turned to offering recommendations as a way to keep shoppers on their sites.
Early recommendations solutions primarily involved manual product-to-product associations. Online
merchandisers, relying on their knowledge of the catalog, Website analytics, and buying trends,
manually connected different items of merchandise.




                                                                                                                               2
Next-Generation Product Recommendations




Most retailers today understand that manual product recommendations, while providing marginal
increases in conversion rates and order values, are limited in scope and difficult to maintain. In the case
of a large, rapidly changing catalog, manual recommendations can be applied and maintained only for
the best-selling items. The “long tail 1 ” of the catalog is neglected.
Manual recommendations solutions have largely been replaced by automated ones. Driven by
computer algorithms that analyze shopper activity, these automated solutions improve on the early,
manual versions by automating the process of linking different products.
Online shopping sites that offer automated, personalized product recommendations can help shoppers
to more easily discover relevant products as quickly as possible. Unlike other navigational tools like site
navigation and site search, recommendations quickly assess each shopper’s unique interests during each
visit. The solutions provide “personal discovery” by continually directing shoppers toward
merchandise that more aptly suits their current interests and preferences, and steering them away from
irrelevant items that may delay and distract them. Automated recommendations solutions are delivered
as on-demand services to any e-commerce site and other online channels.
Most retailers today understand that manual product recommendations, while providing marginal increases in conversion rates
and order values, are limited in scope and difficult to maintain. In the case of a large, rapidly changing catalog, manual
recommendations can be applied and maintained only for the best-selling items.

Research confirms the potential effectiveness of recommendations solutions. Shoppers respond
favorably to the greater convenience, they appreciate the shopping advice, and they are more satisfied
with their purchases. According to one study, more than three-quarters of online shoppers responded
that recommendations improved their online shopping experience. Despite these improvements,
however, the first-generation automated product recommendations solutions are flawed by their
inaccuracy and inflexibility. As a result, automated recommendations have not been widely adopted.


Flaws in First-Generation Recommendations Solutions
Despite the potential advantages of adding recommendations to e-commerce sites, and the ease with
which they can be implemented, many online retailers still have not adopted these solutions. While
many retailers are aware of the potential value, and more than a dozen automated recommendations
solutions are commercially available, only 27.6 percent of U.S. online retailers are offering personalized
recommendations to their visitors, according to market research from eMarketer. 2
Research indicates that only 27.6 percent of online retailers have currently implemented automated recommendations.




1 A term first coined by Wired magazine Editor-in-Chief Chris Anderson, the long tail refers to the theory
that products in low demand or with low sales volume can collectively make up a huge market whose size
rivals or exceeds that of the bestsellers.
2 eMarketer, “Personalized Product Recommendations: Predicting Shoppers’ Needs,” March 2009.




                                                                                                                                  3
Next-Generation Product Recommendations




The limited adoption of recommendations solutions can be explained in part by the fact that many
online retailers are still relatively new in their experience with e-commerce. These organizations may
still be focused on building and refining the core elements of their online presence: keeping product
information up-to-date, presenting multiple product views, improving the purchase process, and
establishing a consistent brand across all shopping channels. Until now, these higher priorities may
have drawn resources and attention away from recommendations solutions. 3
Perhaps a more important factor that has inhibited adoption of automated recommendations is the
existence of certain critical and widely recognized inadequacies in many of the commercially available
first-generation solutions. Despite the fact that nearly half of the retailers responding to a survey on e-
commerce technologies and practices cited the strong potential for personalization technology like
recommendations, only eight percent indicated that the currently available solutions are effective. 4
Let’s explore some key deficiencies of first-generation solutions.

Inadequate Understanding of Shoppers’ Interests and Preferences
Many first-generation automated recommendations solutions are unable to accurately determine each
shopper’s current interests and preferences. As a result, they are unable to present relevant choices.
Irrelevant recommendations can cause shoppers to lose confidence in the retailer, which, over time,
can have a negative impact on the retailer’s brand.
First-generation solutions have difficulty determining and predicting shoppers’ interests primarily
because they don’t take advantage of all available information about each shopper. Many solutions
consider only historical purchase behavior, relying exclusively on an individual shopper’s past
purchases or the past purchases of similar shoppers. As a result, they often recommend items that are
no longer relevant. This would be the equivalent of an in-store sales associate suggesting that a shopper
who purchased snow tires in January might now be interested in a snow blower, despite the fact that it
is now July and the shopper has moved to southern California.
Other solutions use only real-time clickstream data for each visitor—the number of times and order in
which a visitor clicks on particular items. An online drugstore shopper browsing vitamins, for example,
may only see recommendations for other vitamins, when in fact the shopper’s past two visits (which
did not result in a purchase) focused on hair dryers. By failing to remind the shopper of previous items
that were browsed but not purchased, first-generation solutions miss out on key opportunities to
increase order values.
By relying too heavily on clickstream data, these solutions can also be deceived by temporary “flash
traffic” patterns. For example, a large ad campaign may drive a large amount of Web traffic to a




3   Retail Systems Research (RSR), “Survey of Retail Executives Worldwide”, June 2008.
4   Ibid.




                                                                                                           4
Next-Generation Product Recommendations




specific product page in a short period of time, causing the recommendation engine to think that
product has become extremely popular. Overrecommending that product when many shoppers would
have purchased it anyway is a waste of online merchandising inventory.
By failing to consider the panoply of useful information about each shopper, many first-generation
solutions simply cannot provide relevant recommendations. As the later sections of this white paper
will reveal, second-generation solutions have been built to consider all relevant data points, such as
•   The shopper’s geographic location
•   The time of year, day of week, and hour of day
•   The shopper’s referring site or search engine keyword
•   What other parts of the site have been visited in current or past sessions

Inadequate Reach Into and Understanding of the Product Catalog
First-generation automated recommendations solutions do not take the time to learn about an online
retailer’s unique product catalog and how shoppers tend to browse that catalog. Instead, they
“discover” products in the catalog only when shoppers click on them. This lack of native catalog
awareness and understanding can have significant limitations that ultimately reduce the value that these
solutions can provide.
For one, first-generation solutions without catalog understanding have difficulty recommending
products that are newly added to a retailer’s product catalog. They can only recommend these products
once a certain number of shoppers have discovered them. This can be a significant drawback because
new items, by their nature, require promotion to be discovered by shoppers.
Because of their overreliance on shoppers’ clickstream data to form product relationships, these
solutions also fail to recommend merchandise equally. They tend to overrecommend popular items
because they are clicked on more often. This would be the equivalent of an in-store sales associate only
recommending items on the end-caps and never looking at merchandise on the upper shelves in the
aisles—often the items that are most difficult to find.
Limitations of first-generation automated recommendations solutions include
• Inadequate understanding of shoppers’ interests and preferences
• Inadequate reach into and understanding of the product catalog
• Lack of control
• Limited reach across channels
• Inadequate e-commerce and retailing expertise



Lack of Control
First-generation automated recommendations solutions do not give retailers adequate control to refine
or filter recommendations before they are presented to shoppers. These solutions are essentially “black
boxes” that retailers find unpredictable and difficult to overrule when necessary.
Retailers require the flexibility, for example, to present merchandise inside of the brand or category of
items that the shopper is already considering. They may want to feature only top-selling items on




                                                                                                              5
Next-Generation Product Recommendations




certain sections of the online store or only discounted items on others. They may want to up-sell a
shopper only after he’s added an item to the cart, by presenting only higher-priced alternatives. Or
retailers may wish to make recommendations based on their inventory, avoiding out-of-stock items or
favoring overstocked merchandise.
Earlier automated recommendations solutions limit retailers’ flexibility to make these refinements,
mainly due to their lack of catalog understanding, as discussed previously. They may be able to make
refinements based on what shoppers are clicking on (for example, only show products in the same
brand as current product), but cannot easily filter items based on specific attributes in the catalog such
as margin, inventory level, and price.

Limited Reach Across Channels
Consumers today follow a shopping and buying process that may involve multiple channels, which
they use in no particular order. They may research a product using their smartphone, view the item in a
store, read reviews on a social network site, and seek out promotions in e-mail campaigns before they
ultimately buy the product on the Website.
First-generation automated recommendations solutions, however, generally present recommendations
only to the Website. As a result, they fail to influence shoppers at other key decision-making points in
the purchase process. They are unable to reach shoppers with targeted recommendations via e-mail,
mobile stores or apps, off-site shopping widgets, social networks, or the retail contact center.

Inadequate E-Commerce and Retailing Expertise
Many companies offering the first-generation automated recommendations solutions lack a thorough
understanding of e-commerce and retailing. They do not have established expertise, and their solutions
are not focused exclusively on the needs of e-commerce companies who sell products from an online
catalog. As a result, many of these vendors have difficulty providing effective support and
incorporating the automated recommendations solution into a complete e-commerce strategy.


A New Generation of Recommendations Solutions
Even with their limitations, first-generation recommendations solutions have demonstrated the value
of automated merchandising. Retailers do recognize that well-designed automated merchandising
solutions can be effective in helping them achieve critical e-commerce objectives: converting shoppers
to buyers, increasing revenue per transaction, and generating sales from returning customers.
Achieving these objectives, however, requires a new generation of recommendations solutions that
overcome the flaws of their predecessors.
Fortunately for online retailers, a second generation of automated, personalized recommendations
technology has come to market. These newer solutions resolve the flaws of the early solutions:
•   They present shoppers with more-relevant, predictive recommendations.
•   They reach into and understand each retailer’s unique product catalog.




                                                                                                           6
Next-Generation Product Recommendations




•   They give online merchandisers the ability to refine and control recommendations.
•   They can reach shoppers through multiple channels.
•    They are developed and supported by well-established technology partners that understand
    e-commerce strategy and the retail market.

Better Understanding of Visitors’ Interests and Preferences
Second-generation recommendations solutions take into account a more complete range of
information to assess each visitor’s interests and preferences. Unlike first-generation solutions, they can
not only account for each shopper’s past behavior or the past behavior of others (the “wisdom of the
crowds”) but also make judgments based on each shopper’s current session, as well. For example, they
can incorporate information on geography, time of day, customer status, and how the visitor navigated
through the online store.
With this more accurate understanding of each shopper’s needs, retailers can better predict that
shopper’s next likely click and present more-relevant merchandise. Just as an in-store sales associate
would be unlikely to recommend a snow blower to someone now living in southern California, second-
generation recommendations solutions would prevent this inappropriate recommendation from being
presented. Figure 1 illustrates how second-generation recommendations improve understanding of
visitors’ interests and preferences.




Figure 1. Second-generation recommendations solutions overcome flaws of early solutions by taking into account a
more complete range of information to assess each visitor’s interests and preferences.




                                                                                                                  7
Next-Generation Product Recommendations




Greater Reach Into the Catalog
Second-generation recommendations solutions reach more broadly across the entire merchandise
catalog in order to present selections that best suit each shopper’s preferences. Unlike earlier solutions,
which often were unaware of new or niche items, the new solutions ingest a live feed of the product
catalog regularly, form dynamic relationships among products as they are added, and are aware of new
items and changes to items in the catalog. Second-generation solutions act like a knowledgeable in-
store sales associate, who knows precisely what item the shopper is looking for and where to find it
and can quickly guide the customer to it. Figure 2 illustrates personal merchandise displays.




Figure 2. Using a full range of information about the shopper, retailers can now deliver cross-sells on product detail
pages that are dynamic and personalized based on each shopper’s interests.

Extended reach across the catalog also allows retailers to use recommendations solutions for more
than product cross-sells. Instead, these solutions can be used as automated merchandising tools that
can take over where online merchants leave off. Today’s second-generation solutions can be used to
automate and personalize merchandising across the entire online store. These broad uses of automated
merchandising may include
•    “Just for you” slots on the home page, driven by all previous and current data available about each
    shopper
•    Search engine landing pages that use the incoming keyword search to automatically display
    recommendations and relevant merchandise
•   Null search result pages that use the site search keyword to dynamically display relevant
    merchandise




                                                                                                                     8
Next-Generation Product Recommendations




•   Top-sellers pages showing personalized product selections, but limited to top-selling items
•   Gift guides showing a range of dynamic gift ideas personalized to each shopper’s current interests

Greater Reach Out to Customers
Today’s second-generation automated recommendations solutions can reach out proactively across an
array of shopping channels. Based on the information accumulated about a shopper’s interests and
preferences, retailers can dynamically present attractive and appropriate merchandise via e-mail, mobile
stores or apps, off-site shopping widgets, social networks, or the retail contact center. Figure 3
illustrates this capability through e-mail offers.




Figure 3. Today’s second generation of automated recommendations solutions can reach out proactively to shoppers
and present personalized offers in e-mail marketing campaigns.


More Control to Refine Selections
To overcome the lack of control in first-generation solutions, second-generation automated
recommendations technology now allows retailers to carefully refine and filter recommendations
before they are presented to shoppers. This allows merchants to leverage the benefits of




                                                                                                               9
Next-Generation Product Recommendations




personalization and automation, while at the same time ensuring that solutions adhere to their
merchandising guidelines or strategy.
Recommendation refinements can be based on catalog data, visitor behavior, or a combination of
the two.
•    Refinements based on catalog data allow retailers to recommend merchandise based on particular
    product attributes such as category, brand, price, collection, top-sellers, and new products.
    Merchants can also create custom refinements based on their unique catalogs, mixing “fixed”
    recommendations of specific products or excluding specific products from automated
    recommendations.
•    Second-generation recommendations solutions also allow retailers to make recommendations based
    on common types of visitor behavior. For example, merchants may want to ensure that shoppers
    looking at full-priced items are presented with only full-priced recommendations, whereas budget-
    conscious shoppers see recommendations for discounted items. Recommendations can also be
    refined based on common keywords that shoppers key into a Website search engine. For example,
    all visitors who searched on “TVs” and click on the search engine link for a retailer might see only
    personalized TV recommendations on the home page.
Sophisticated retailers can combine catalog data and visitor behavior to create highly targeted,
automated merchandising displays. For instance, a budget-conscious shopper shopping in a specific
brand may see only on-sale, in-brand recommendations. Figure 4 illustrates the types of control and
refinements that retailers have with second-generation automated recommendations solutions.




Figure 4. Second-generation automated recommendations solutions allow retailers to carefully refine and filter
recommendations before they are presented to shoppers. Recommendation refinements are generally based on
catalog data, visitor behavior, or a combination of the two.




                                                                                                                  10
Next-Generation Product Recommendations




Relationship with a Well-Established Provider with Retail Expertise and Focus
While earlier recommendations solutions were available primarily from technology startups with
limited e-commerce experience, retailers can now work with well-established providers with a deep
understanding of the retail market. These more-established providers take the time and have the
resources to thoroughly understand each retailer’s unique business, focus exclusively on their
e-commerce requirements, and deliver an automated recommendations and merchandising solution
that fits their overall e-commerce strategy.
The advantages of second-generation automated recommendations solutions include
• Relevance. Present shoppers with more-appropriate recommendations based on a complete understanding of their
   preferences.
• Reach. Present recommendations from across the entire catalog of merchandise including new and niche items, and deliver
   them proactively to customers.
• Refinement. Flexibly control the recommendations presented to shoppers.
• Relationship. Work with a well-established recommendations solution vendor with a deep understanding of the retail
   environment and e-commerce strategies.



Proven Impact
Second-generation recommendations solutions can have a significant impact on retailers’ success. The
experience of leading e-commerce merchandisers has demonstrated quantifiable improvements in
lifting online revenue quickly by converting more browsers into buyers, increasing the average revenue
per transaction, and securing more follow-on sales from existing customers.
Data from customers of the Oracle Recommendations On Demand solution, for example, shows that
shoppers who click on recommendations during a session are likely to convert up to three times more
often and spend up to 30 percent more than those who don’t. These dramatic increases in key
performance indicators are direct evidence of the benefit of personalizing online merchandise so that
shoppers can easily find what they want and discover other items that match their unique interests.

The Positive Experience at Tommy Hilfiger
Tommy Hilfiger, one of the world’s most recognized premium lifestyle brands for 18- to 35-year-olds,
implemented an advanced recommendations solution to improve the performance of its online store.
The company’s e-commerce staff recognized the potential value of presenting shoppers with
appropriate recommendations during the shopping process. However, their initial experience with early
recommendations solutions yielded few additional sales.
For one, it was necessary to manually hard-code one-to-one product relationships. This process made
it almost impossible to make changes quickly as new merchandise was added and outdated items were
removed.
“We’ve seen a threefold increase in online shopper conversions when someone interacts with the Oracle Recommendations On
Demand solution. At the same time, our staff is spending less time than ever managing the process. The more our shoppers
use it, the smarter the engine gets and the more value it delivers.”
                                                                       —Tom Davis, Director of E-Commerce, Tommy Hilfiger




                                                                                                                            11
Next-Generation Product Recommendations




Moreover, early solutions ignored much of the data about an individual shopper that could be used to
make relevant recommendations. Recommendations were based on only one factor: items already in
the shopping cart. As a result, the e-commerce site tended to present the same set of merchandise to all
customers. Recommendations could not be effectively tailored to the shopper’s particular interests at
the time of their visit.
Upon deploying the Oracle Recommendations On Demand solution, Tommy Hilfiger saw a significant
increase in sales from its online store. According to Tom Davis, director of e-commerce at Tommy
Hilfiger, “We’ve seen a threefold increase in online shopper conversions when someone interacts with
the Oracle Recommendations On Demand solution. At the same time, our staff is spending less time
than ever managing the process. The more our shoppers use it, the smarter the engine gets and the
more value it delivers.”
Using a full range of information about its online shoppers, Tommy Hilfiger can now construct
personalized merchandise displays, which are dynamically prepared to suit each shopper’s current
interests. These displays are presented while customers are browsing the online store, and they can be
sent to targeted customers via e-mail, as illustrated below.




Figure 5. Using a full range of information about each shopper, Tommy Hilfiger can now construct personalized
merchandise displays, which are dynamically prepared to suit each shopper. Tommy Hilfiger is powering a top-seller
page (shown here) with the Oracle Recommendations On Demand solution, allowing it to personalize what top-sellers
are shown to each shopper.

In addition to better understanding each shopper’s preferences, the solution relieves the Tommy
Hilfiger staff from the difficult task of hard-coding all cross-sell and up-sell options. This allows them
flexibility to react more quickly as they introduce new products, as items go out of stock, or as they
need to make other adjustments to the merchandise mix.




                                                                                                                12
Next-Generation Product Recommendations




The Oracle Recommendations On Demand Solution
Comprised of Oracle Recommendations Single-Channel On Demand and Oracle Recommendations
Multichannel On Demand, the Oracle Recommendations On Demand solution is an automated,
personalized merchandising service that helps online retailers quickly lift revenue by recommending the
most-relevant products from the catalog to each shopper. Balancing predictive relevancy with
merchant control, the solution automatically selects the products that match each shopper’s needs and
each retailer’s business needs, presents them at the right time in the right channel, and demonstrates
the impact on key performance metrics such as page views, conversion rates, and order values.
The Oracle Recommendations On Demand solution is part of Oracle’s family of e-commerce
solutions that lift revenue quickly, easily, and measurably by optimizing every online interaction. The
service can be deployed easily and rapidly to any Website—even those not running on the Oracle ATG
Web Commerce platform.
By using the Oracle Recommendations On Demand solution to automate and personalize online
merchandising, online retailers can realize the following significant benefits:
•    Convert more browsers into buyers by presenting the most-relevant products from the catalog to
    each shopper across channels.
•   Increase order values by automating and personalizing cross-sells and up-sells.
•    Retain more customers and build a core base of loyal, repeat buyers by personalizing the customer
    experience across channels.


Conclusion
Personalized product recommendations solutions can significantly enhance retailers’ e-commerce
business by effectively converting visitors to buyers, increasing the value of transactions, and
generating more repeat purchases from loyal customers. Research and experience with large online
retailers confirms the potential effectiveness of recommendations solutions. However, first-generation
automated recommendations solutions did not adequately understand shoppers’ interests and
preferences. They also had limited reach into the product catalog and across channels and provided
little control for retailers. Oracle is a well-established recommendations solution vendor with a deep
understanding of the retail environment and e-commerce strategies. The Oracle Recommendations On
Demand products—Oracle Recommendations Single-Channel On Demand and Oracle
Recommendations Multichannel On Demand—are the next generation of automated product
recommendations solutions. Together, they capture the full value of your online traffic flows by
providing a better understanding of visitor interests, more-relevant product recommendations across
the entire catalog, and a greater ability for merchandisers to refine and control recommendations,
ultimately allowing your Website to achieve the best possible sales outcome from every online visit.




                                                                                                         13
Next-Generation Product    Copyright © 2009, 2011, Oracle and/or its affiliates. All rights reserved. This document is provided for information purposes only and
Recommendations            the contents hereof are subject to change without notice. This document is not warranted to be error-free, nor subject to any other
March 2011                 warranties or conditions, whether expressed orally or implied in law, including implied warranties and conditions of merchantability or
                           fitness for a particular purpose. We specifically disclaim any liability with respect to this document and no contractual obligations are
Oracle Corporation
                           formed either directly or indirectly by this document. This document may not be reproduced or transmitted in any form or by any
World Headquarters
                           means, electronic or mechanical, for any purpose, without our prior written permission.
500 Oracle Parkway
Redwood Shores, CA 94065
                           Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.
U.S.A.

Worldwide Inquiries:       AMD, Opteron, the AMD logo, and the AMD Opteron logo are trademarks or registered trademarks of Advanced Micro Devices.
Phone: +1.650.506.7000     Intel and Intel Xeon are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license
Fax: +1.650.506.7200       and are trademarks or registered trademarks of SPARC International, Inc. UNIX is a registered trademark licensed through X/Open
                           Company, Ltd. 0311
oracle.com

Contenu connexe

Tendances

How Seamless are You? retail research findings
How Seamless are You? retail research findingsHow Seamless are You? retail research findings
How Seamless are You? retail research findingsBertrand Jonquois
 
iVend Retail Integrated Omni-channel Solutions
iVend Retail Integrated Omni-channel SolutionsiVend Retail Integrated Omni-channel Solutions
iVend Retail Integrated Omni-channel SolutionsSundae Solutions Co., Ltd.
 
Marketing to Consumers Using Social Media
Marketing to Consumers Using Social MediaMarketing to Consumers Using Social Media
Marketing to Consumers Using Social MediaJanet Jaiswal
 
Consumer durables retail efficiency- Images Retail-December 2013
Consumer durables retail efficiency- Images Retail-December 2013Consumer durables retail efficiency- Images Retail-December 2013
Consumer durables retail efficiency- Images Retail-December 2013Shijo Thomas
 
TOTAL RETAIL-CUSTOMER TRENDS THAT ARE CHANGING BUSINESS MODELS
TOTAL RETAIL-CUSTOMER TRENDS THAT ARE CHANGING BUSINESS MODELSTOTAL RETAIL-CUSTOMER TRENDS THAT ARE CHANGING BUSINESS MODELS
TOTAL RETAIL-CUSTOMER TRENDS THAT ARE CHANGING BUSINESS MODELSRetailers Association of India
 
Factors Influencing the Intention to Buy Products Online
Factors Influencing the Intention to Buy Products OnlineFactors Influencing the Intention to Buy Products Online
Factors Influencing the Intention to Buy Products Onlineijtsrd
 
Whitepaper on Customer Experience Management (CEM) perspectives in Retail (De...
Whitepaper on Customer Experience Management (CEM) perspectives in Retail (De...Whitepaper on Customer Experience Management (CEM) perspectives in Retail (De...
Whitepaper on Customer Experience Management (CEM) perspectives in Retail (De...priyadarshinim.r
 
Tutorial on the McKinsey/Harvard "Customer Decision Journey" by John Sing
Tutorial on the McKinsey/Harvard "Customer Decision Journey" by John SingTutorial on the McKinsey/Harvard "Customer Decision Journey" by John Sing
Tutorial on the McKinsey/Harvard "Customer Decision Journey" by John SingJohn Sing
 
Shopper marketing insights for retailers webinar
Shopper marketing insights for retailers webinarShopper marketing insights for retailers webinar
Shopper marketing insights for retailers webinarJanet Jaiswal
 
Co-presented with Mars Petcare on "Disrupting the habitual consumer decision ...
Co-presented with Mars Petcare on "Disrupting the habitual consumer decision ...Co-presented with Mars Petcare on "Disrupting the habitual consumer decision ...
Co-presented with Mars Petcare on "Disrupting the habitual consumer decision ...SKIM
 
Predictive Marketing Engine - Intema Solutions (White Paper)
Predictive Marketing Engine - Intema Solutions (White Paper)Predictive Marketing Engine - Intema Solutions (White Paper)
Predictive Marketing Engine - Intema Solutions (White Paper)Intema Solutions
 
E commerce with data analytics
E commerce with data analyticsE commerce with data analytics
E commerce with data analyticsDaniel Thomas
 
2019 Global Shopper Study White Paper
2019 Global Shopper Study White Paper2019 Global Shopper Study White Paper
2019 Global Shopper Study White PaperDon Talend
 
Progressive Grocer February 2015 - Grocery Loyalty Article
Progressive Grocer February 2015 - Grocery Loyalty ArticleProgressive Grocer February 2015 - Grocery Loyalty Article
Progressive Grocer February 2015 - Grocery Loyalty ArticleGraeme McVie
 
Winners and losers along the digital path to purchase_TNS
Winners and losers along the digital path to purchase_TNSWinners and losers along the digital path to purchase_TNS
Winners and losers along the digital path to purchase_TNSGabriella Bergaglio
 
Retail Analytics: Game Changer for Customer Loyalty
Retail Analytics: Game Changer for Customer LoyaltyRetail Analytics: Game Changer for Customer Loyalty
Retail Analytics: Game Changer for Customer LoyaltyCognizant
 
Think Retail - print's role in the retail market of today
Think Retail - print's role in the retail market of todayThink Retail - print's role in the retail market of today
Think Retail - print's role in the retail market of todayCanon Belgium
 
A Fashion Brand's Guide to Building Consumer Intimacy in the Digital Age
A Fashion Brand's Guide to Building Consumer Intimacy in the Digital AgeA Fashion Brand's Guide to Building Consumer Intimacy in the Digital Age
A Fashion Brand's Guide to Building Consumer Intimacy in the Digital AgeAndy Larsen
 

Tendances (20)

How Seamless are You? retail research findings
How Seamless are You? retail research findingsHow Seamless are You? retail research findings
How Seamless are You? retail research findings
 
iVend Retail Integrated Omni-channel Solutions
iVend Retail Integrated Omni-channel SolutionsiVend Retail Integrated Omni-channel Solutions
iVend Retail Integrated Omni-channel Solutions
 
Marketing to Consumers Using Social Media
Marketing to Consumers Using Social MediaMarketing to Consumers Using Social Media
Marketing to Consumers Using Social Media
 
Consumer durables retail efficiency- Images Retail-December 2013
Consumer durables retail efficiency- Images Retail-December 2013Consumer durables retail efficiency- Images Retail-December 2013
Consumer durables retail efficiency- Images Retail-December 2013
 
TOTAL RETAIL-CUSTOMER TRENDS THAT ARE CHANGING BUSINESS MODELS
TOTAL RETAIL-CUSTOMER TRENDS THAT ARE CHANGING BUSINESS MODELSTOTAL RETAIL-CUSTOMER TRENDS THAT ARE CHANGING BUSINESS MODELS
TOTAL RETAIL-CUSTOMER TRENDS THAT ARE CHANGING BUSINESS MODELS
 
Factors Influencing the Intention to Buy Products Online
Factors Influencing the Intention to Buy Products OnlineFactors Influencing the Intention to Buy Products Online
Factors Influencing the Intention to Buy Products Online
 
Whitepaper on Customer Experience Management (CEM) perspectives in Retail (De...
Whitepaper on Customer Experience Management (CEM) perspectives in Retail (De...Whitepaper on Customer Experience Management (CEM) perspectives in Retail (De...
Whitepaper on Customer Experience Management (CEM) perspectives in Retail (De...
 
A Guide for Omni-Channel Retailers
A Guide for Omni-Channel RetailersA Guide for Omni-Channel Retailers
A Guide for Omni-Channel Retailers
 
Tutorial on the McKinsey/Harvard "Customer Decision Journey" by John Sing
Tutorial on the McKinsey/Harvard "Customer Decision Journey" by John SingTutorial on the McKinsey/Harvard "Customer Decision Journey" by John Sing
Tutorial on the McKinsey/Harvard "Customer Decision Journey" by John Sing
 
Shopper marketing insights for retailers webinar
Shopper marketing insights for retailers webinarShopper marketing insights for retailers webinar
Shopper marketing insights for retailers webinar
 
Co-presented with Mars Petcare on "Disrupting the habitual consumer decision ...
Co-presented with Mars Petcare on "Disrupting the habitual consumer decision ...Co-presented with Mars Petcare on "Disrupting the habitual consumer decision ...
Co-presented with Mars Petcare on "Disrupting the habitual consumer decision ...
 
Marketing Capstone Project Ideas
Marketing Capstone Project IdeasMarketing Capstone Project Ideas
Marketing Capstone Project Ideas
 
Predictive Marketing Engine - Intema Solutions (White Paper)
Predictive Marketing Engine - Intema Solutions (White Paper)Predictive Marketing Engine - Intema Solutions (White Paper)
Predictive Marketing Engine - Intema Solutions (White Paper)
 
E commerce with data analytics
E commerce with data analyticsE commerce with data analytics
E commerce with data analytics
 
2019 Global Shopper Study White Paper
2019 Global Shopper Study White Paper2019 Global Shopper Study White Paper
2019 Global Shopper Study White Paper
 
Progressive Grocer February 2015 - Grocery Loyalty Article
Progressive Grocer February 2015 - Grocery Loyalty ArticleProgressive Grocer February 2015 - Grocery Loyalty Article
Progressive Grocer February 2015 - Grocery Loyalty Article
 
Winners and losers along the digital path to purchase_TNS
Winners and losers along the digital path to purchase_TNSWinners and losers along the digital path to purchase_TNS
Winners and losers along the digital path to purchase_TNS
 
Retail Analytics: Game Changer for Customer Loyalty
Retail Analytics: Game Changer for Customer LoyaltyRetail Analytics: Game Changer for Customer Loyalty
Retail Analytics: Game Changer for Customer Loyalty
 
Think Retail - print's role in the retail market of today
Think Retail - print's role in the retail market of todayThink Retail - print's role in the retail market of today
Think Retail - print's role in the retail market of today
 
A Fashion Brand's Guide to Building Consumer Intimacy in the Digital Age
A Fashion Brand's Guide to Building Consumer Intimacy in the Digital AgeA Fashion Brand's Guide to Building Consumer Intimacy in the Digital Age
A Fashion Brand's Guide to Building Consumer Intimacy in the Digital Age
 

En vedette

Health sectorreformsnadia
Health sectorreformsnadiaHealth sectorreformsnadia
Health sectorreformsnadiaMohamed Rafique
 
Social Commerce by Bounche
Social Commerce by Bounche Social Commerce by Bounche
Social Commerce by Bounche ORORI.com
 
Organizingand Finding Resourceswith Office Share Point Server2007 Refresh
Organizingand Finding Resourceswith Office Share Point Server2007 RefreshOrganizingand Finding Resourceswith Office Share Point Server2007 Refresh
Organizingand Finding Resourceswith Office Share Point Server2007 RefreshLiquidHub
 
Post en eger E bar fliplock
Post en eger E bar fliplockPost en eger E bar fliplock
Post en eger E bar fliplockposteneger
 
Ngsi.ohana presentation.14
Ngsi.ohana presentation.14Ngsi.ohana presentation.14
Ngsi.ohana presentation.14kristenmwatkins
 
10G over Copper by 3M
10G over Copper by 3M10G over Copper by 3M
10G over Copper by 3MSheikh Sabri
 

En vedette (6)

Health sectorreformsnadia
Health sectorreformsnadiaHealth sectorreformsnadia
Health sectorreformsnadia
 
Social Commerce by Bounche
Social Commerce by Bounche Social Commerce by Bounche
Social Commerce by Bounche
 
Organizingand Finding Resourceswith Office Share Point Server2007 Refresh
Organizingand Finding Resourceswith Office Share Point Server2007 RefreshOrganizingand Finding Resourceswith Office Share Point Server2007 Refresh
Organizingand Finding Resourceswith Office Share Point Server2007 Refresh
 
Post en eger E bar fliplock
Post en eger E bar fliplockPost en eger E bar fliplock
Post en eger E bar fliplock
 
Ngsi.ohana presentation.14
Ngsi.ohana presentation.14Ngsi.ohana presentation.14
Ngsi.ohana presentation.14
 
10G over Copper by 3M
10G over Copper by 3M10G over Copper by 3M
10G over Copper by 3M
 

Similaire à Next generation-product-recommendation

Assessment of most selling staples & FMCG products in mop & pop stores close ...
Assessment of most selling staples & FMCG products in mop & pop stores close ...Assessment of most selling staples & FMCG products in mop & pop stores close ...
Assessment of most selling staples & FMCG products in mop & pop stores close ...BHOMA RAM
 
Key insights in ecommerce personalisation - via Shopboostr
Key insights in ecommerce personalisation - via ShopboostrKey insights in ecommerce personalisation - via Shopboostr
Key insights in ecommerce personalisation - via ShopboostrDimitri Haußmann
 
Impact Of Shelf Management On A Supermarket
Impact Of Shelf Management On A SupermarketImpact Of Shelf Management On A Supermarket
Impact Of Shelf Management On A SupermarketStephanie King
 
The Importance of CX for Outdoor Manufacturers
The Importance of CX for Outdoor ManufacturersThe Importance of CX for Outdoor Manufacturers
The Importance of CX for Outdoor ManufacturersElement Three
 
Entrepreneurship & Early Stages Growth Marketing Planning
Entrepreneurship & Early Stages Growth Marketing PlanningEntrepreneurship & Early Stages Growth Marketing Planning
Entrepreneurship & Early Stages Growth Marketing PlanningDan Taylor
 
Clienteling Without Creepiness_Spring 2016
Clienteling Without Creepiness_Spring 2016Clienteling Without Creepiness_Spring 2016
Clienteling Without Creepiness_Spring 2016Dianne Inniss
 
Practical ways to use dynamic recommendations
Practical ways to use dynamic recommendationsPractical ways to use dynamic recommendations
Practical ways to use dynamic recommendationsYesLifecycleMarketing
 
consumer behaviour.pptx
consumer behaviour.pptxconsumer behaviour.pptx
consumer behaviour.pptxvibhore jain
 
Startups & The Retail Opportunity
Startups & The Retail OpportunityStartups & The Retail Opportunity
Startups & The Retail OpportunityRich Bowen
 
Introductory Guide to Clustering
Introductory Guide to ClusteringIntroductory Guide to Clustering
Introductory Guide to ClusteringMohammed Rahman
 
Visual merchandising notes
Visual merchandising notesVisual merchandising notes
Visual merchandising notessuniltalekar1
 
Notes chapter 1
Notes chapter 1Notes chapter 1
Notes chapter 1kamran
 
321 23 hints-concept-review-questions_chapter-1-retailing-role-relevance-trends
321 23 hints-concept-review-questions_chapter-1-retailing-role-relevance-trends321 23 hints-concept-review-questions_chapter-1-retailing-role-relevance-trends
321 23 hints-concept-review-questions_chapter-1-retailing-role-relevance-trendskamran
 
Consumer Behaviour final latest.pptx
Consumer Behaviour final latest.pptxConsumer Behaviour final latest.pptx
Consumer Behaviour final latest.pptxparibhardwaj
 
Product recommendation guide for eCommerce
Product recommendation guide for eCommerceProduct recommendation guide for eCommerce
Product recommendation guide for eCommercePrzemyslaw Mankowski
 
DATA MINING IN RETAIL SECTOR
DATA MINING IN RETAIL SECTORDATA MINING IN RETAIL SECTOR
DATA MINING IN RETAIL SECTORRenuka Chand
 
Elements of production planning for goods and services
Elements of production planning for goods and servicesElements of production planning for goods and services
Elements of production planning for goods and servicesAlexander Decker
 
Avatar shops - how to implement traditional stores experience in online stores
Avatar shops - how to implement traditional stores experience in online storesAvatar shops - how to implement traditional stores experience in online stores
Avatar shops - how to implement traditional stores experience in online storesSasa Bastian Jovanovic
 
Aligning Value with Key Capabilities to Drive Retail Success
Aligning Value with Key Capabilities to Drive Retail SuccessAligning Value with Key Capabilities to Drive Retail Success
Aligning Value with Key Capabilities to Drive Retail SuccessCognizant
 

Similaire à Next generation-product-recommendation (20)

Assessment of most selling staples & FMCG products in mop & pop stores close ...
Assessment of most selling staples & FMCG products in mop & pop stores close ...Assessment of most selling staples & FMCG products in mop & pop stores close ...
Assessment of most selling staples & FMCG products in mop & pop stores close ...
 
Key insights in ecommerce personalisation - via Shopboostr
Key insights in ecommerce personalisation - via ShopboostrKey insights in ecommerce personalisation - via Shopboostr
Key insights in ecommerce personalisation - via Shopboostr
 
Impact Of Shelf Management On A Supermarket
Impact Of Shelf Management On A SupermarketImpact Of Shelf Management On A Supermarket
Impact Of Shelf Management On A Supermarket
 
The Importance of CX for Outdoor Manufacturers
The Importance of CX for Outdoor ManufacturersThe Importance of CX for Outdoor Manufacturers
The Importance of CX for Outdoor Manufacturers
 
Entrepreneurship & Early Stages Growth Marketing Planning
Entrepreneurship & Early Stages Growth Marketing PlanningEntrepreneurship & Early Stages Growth Marketing Planning
Entrepreneurship & Early Stages Growth Marketing Planning
 
Slide_R_A_A2 slide #4
Slide_R_A_A2 slide #4Slide_R_A_A2 slide #4
Slide_R_A_A2 slide #4
 
Clienteling Without Creepiness_Spring 2016
Clienteling Without Creepiness_Spring 2016Clienteling Without Creepiness_Spring 2016
Clienteling Without Creepiness_Spring 2016
 
Practical ways to use dynamic recommendations
Practical ways to use dynamic recommendationsPractical ways to use dynamic recommendations
Practical ways to use dynamic recommendations
 
consumer behaviour.pptx
consumer behaviour.pptxconsumer behaviour.pptx
consumer behaviour.pptx
 
Startups & The Retail Opportunity
Startups & The Retail OpportunityStartups & The Retail Opportunity
Startups & The Retail Opportunity
 
Introductory Guide to Clustering
Introductory Guide to ClusteringIntroductory Guide to Clustering
Introductory Guide to Clustering
 
Visual merchandising notes
Visual merchandising notesVisual merchandising notes
Visual merchandising notes
 
Notes chapter 1
Notes chapter 1Notes chapter 1
Notes chapter 1
 
321 23 hints-concept-review-questions_chapter-1-retailing-role-relevance-trends
321 23 hints-concept-review-questions_chapter-1-retailing-role-relevance-trends321 23 hints-concept-review-questions_chapter-1-retailing-role-relevance-trends
321 23 hints-concept-review-questions_chapter-1-retailing-role-relevance-trends
 
Consumer Behaviour final latest.pptx
Consumer Behaviour final latest.pptxConsumer Behaviour final latest.pptx
Consumer Behaviour final latest.pptx
 
Product recommendation guide for eCommerce
Product recommendation guide for eCommerceProduct recommendation guide for eCommerce
Product recommendation guide for eCommerce
 
DATA MINING IN RETAIL SECTOR
DATA MINING IN RETAIL SECTORDATA MINING IN RETAIL SECTOR
DATA MINING IN RETAIL SECTOR
 
Elements of production planning for goods and services
Elements of production planning for goods and servicesElements of production planning for goods and services
Elements of production planning for goods and services
 
Avatar shops - how to implement traditional stores experience in online stores
Avatar shops - how to implement traditional stores experience in online storesAvatar shops - how to implement traditional stores experience in online stores
Avatar shops - how to implement traditional stores experience in online stores
 
Aligning Value with Key Capabilities to Drive Retail Success
Aligning Value with Key Capabilities to Drive Retail SuccessAligning Value with Key Capabilities to Drive Retail Success
Aligning Value with Key Capabilities to Drive Retail Success
 

Dernier

IIBA® Melbourne - Navigating Business Analysis - Excellence for Career Growth...
IIBA® Melbourne - Navigating Business Analysis - Excellence for Career Growth...IIBA® Melbourne - Navigating Business Analysis - Excellence for Career Growth...
IIBA® Melbourne - Navigating Business Analysis - Excellence for Career Growth...AustraliaChapterIIBA
 
Amazon ppt.pptx Amazon about the company
Amazon ppt.pptx Amazon about the companyAmazon ppt.pptx Amazon about the company
Amazon ppt.pptx Amazon about the companyfashionfound007
 
Developing Coaching Skills: Mine, Yours, Ours
Developing Coaching Skills: Mine, Yours, OursDeveloping Coaching Skills: Mine, Yours, Ours
Developing Coaching Skills: Mine, Yours, OursKaiNexus
 
Ethical stalking by Mark Williams. UpliftLive 2024
Ethical stalking by Mark Williams. UpliftLive 2024Ethical stalking by Mark Williams. UpliftLive 2024
Ethical stalking by Mark Williams. UpliftLive 2024Winbusinessin
 
Team B Mind Map for Organizational Chg..
Team B Mind Map for Organizational Chg..Team B Mind Map for Organizational Chg..
Team B Mind Map for Organizational Chg..dlewis191
 
Fabric RFID Wristbands in Ireland for Events and Festivals
Fabric RFID Wristbands in Ireland for Events and FestivalsFabric RFID Wristbands in Ireland for Events and Festivals
Fabric RFID Wristbands in Ireland for Events and FestivalsWristbands Ireland
 
Live-Streaming in the Music Industry Webinar
Live-Streaming in the Music Industry WebinarLive-Streaming in the Music Industry Webinar
Live-Streaming in the Music Industry WebinarNathanielSchmuck
 
Intellectual Property Licensing Examples
Intellectual Property Licensing ExamplesIntellectual Property Licensing Examples
Intellectual Property Licensing Examplesamberjiles31
 
The End of Business as Usual: Rewire the Way You Work to Succeed in the Consu...
The End of Business as Usual: Rewire the Way You Work to Succeed in the Consu...The End of Business as Usual: Rewire the Way You Work to Succeed in the Consu...
The End of Business as Usual: Rewire the Way You Work to Succeed in the Consu...Brian Solis
 
NewBase 25 March 2024 Energy News issue - 1710 by Khaled Al Awadi_compress...
NewBase  25 March  2024  Energy News issue - 1710 by Khaled Al Awadi_compress...NewBase  25 March  2024  Energy News issue - 1710 by Khaled Al Awadi_compress...
NewBase 25 March 2024 Energy News issue - 1710 by Khaled Al Awadi_compress...Khaled Al Awadi
 
Building Your Personal Brand on LinkedIn - Expert Planet- 2024
 Building Your Personal Brand on LinkedIn - Expert Planet-  2024 Building Your Personal Brand on LinkedIn - Expert Planet-  2024
Building Your Personal Brand on LinkedIn - Expert Planet- 2024Stephan Koning
 
Graham and Doddsville - Issue 1 - Winter 2006 (1).pdf
Graham and Doddsville - Issue 1 - Winter 2006 (1).pdfGraham and Doddsville - Issue 1 - Winter 2006 (1).pdf
Graham and Doddsville - Issue 1 - Winter 2006 (1).pdfAnhNguyen97152
 
PDT 88 - 4 million seed - Seed - Protecto.pdf
PDT 88 - 4 million seed - Seed - Protecto.pdfPDT 88 - 4 million seed - Seed - Protecto.pdf
PDT 88 - 4 million seed - Seed - Protecto.pdfHajeJanKamps
 
Borderless Access - Global Panel book-unlock 2024
Borderless Access - Global Panel book-unlock 2024Borderless Access - Global Panel book-unlock 2024
Borderless Access - Global Panel book-unlock 2024Borderless Access
 
ISONIKE Ltd Accreditation for the Conformity Assessment and Certification of ...
ISONIKE Ltd Accreditation for the Conformity Assessment and Certification of ...ISONIKE Ltd Accreditation for the Conformity Assessment and Certification of ...
ISONIKE Ltd Accreditation for the Conformity Assessment and Certification of ...ISONIKELtd
 
MoneyBridge Pitch Deck - Investor Presentation
MoneyBridge Pitch Deck - Investor PresentationMoneyBridge Pitch Deck - Investor Presentation
MoneyBridge Pitch Deck - Investor Presentationbaron83
 
Cracking the ‘Business Process Outsourcing’ Code Main.pptx
Cracking the ‘Business Process Outsourcing’ Code Main.pptxCracking the ‘Business Process Outsourcing’ Code Main.pptx
Cracking the ‘Business Process Outsourcing’ Code Main.pptxWorkforce Group
 
A flour, rice and Suji company in Jhang.
A flour, rice and Suji company in Jhang.A flour, rice and Suji company in Jhang.
A flour, rice and Suji company in Jhang.mcshagufta46
 
Michael Vidyakin: Introduction to PMO (UA)
Michael Vidyakin: Introduction to PMO (UA)Michael Vidyakin: Introduction to PMO (UA)
Michael Vidyakin: Introduction to PMO (UA)Lviv Startup Club
 

Dernier (20)

IIBA® Melbourne - Navigating Business Analysis - Excellence for Career Growth...
IIBA® Melbourne - Navigating Business Analysis - Excellence for Career Growth...IIBA® Melbourne - Navigating Business Analysis - Excellence for Career Growth...
IIBA® Melbourne - Navigating Business Analysis - Excellence for Career Growth...
 
Amazon ppt.pptx Amazon about the company
Amazon ppt.pptx Amazon about the companyAmazon ppt.pptx Amazon about the company
Amazon ppt.pptx Amazon about the company
 
Developing Coaching Skills: Mine, Yours, Ours
Developing Coaching Skills: Mine, Yours, OursDeveloping Coaching Skills: Mine, Yours, Ours
Developing Coaching Skills: Mine, Yours, Ours
 
Ethical stalking by Mark Williams. UpliftLive 2024
Ethical stalking by Mark Williams. UpliftLive 2024Ethical stalking by Mark Williams. UpliftLive 2024
Ethical stalking by Mark Williams. UpliftLive 2024
 
Team B Mind Map for Organizational Chg..
Team B Mind Map for Organizational Chg..Team B Mind Map for Organizational Chg..
Team B Mind Map for Organizational Chg..
 
Fabric RFID Wristbands in Ireland for Events and Festivals
Fabric RFID Wristbands in Ireland for Events and FestivalsFabric RFID Wristbands in Ireland for Events and Festivals
Fabric RFID Wristbands in Ireland for Events and Festivals
 
Live-Streaming in the Music Industry Webinar
Live-Streaming in the Music Industry WebinarLive-Streaming in the Music Industry Webinar
Live-Streaming in the Music Industry Webinar
 
Intellectual Property Licensing Examples
Intellectual Property Licensing ExamplesIntellectual Property Licensing Examples
Intellectual Property Licensing Examples
 
The End of Business as Usual: Rewire the Way You Work to Succeed in the Consu...
The End of Business as Usual: Rewire the Way You Work to Succeed in the Consu...The End of Business as Usual: Rewire the Way You Work to Succeed in the Consu...
The End of Business as Usual: Rewire the Way You Work to Succeed in the Consu...
 
NewBase 25 March 2024 Energy News issue - 1710 by Khaled Al Awadi_compress...
NewBase  25 March  2024  Energy News issue - 1710 by Khaled Al Awadi_compress...NewBase  25 March  2024  Energy News issue - 1710 by Khaled Al Awadi_compress...
NewBase 25 March 2024 Energy News issue - 1710 by Khaled Al Awadi_compress...
 
Building Your Personal Brand on LinkedIn - Expert Planet- 2024
 Building Your Personal Brand on LinkedIn - Expert Planet-  2024 Building Your Personal Brand on LinkedIn - Expert Planet-  2024
Building Your Personal Brand on LinkedIn - Expert Planet- 2024
 
Graham and Doddsville - Issue 1 - Winter 2006 (1).pdf
Graham and Doddsville - Issue 1 - Winter 2006 (1).pdfGraham and Doddsville - Issue 1 - Winter 2006 (1).pdf
Graham and Doddsville - Issue 1 - Winter 2006 (1).pdf
 
PDT 88 - 4 million seed - Seed - Protecto.pdf
PDT 88 - 4 million seed - Seed - Protecto.pdfPDT 88 - 4 million seed - Seed - Protecto.pdf
PDT 88 - 4 million seed - Seed - Protecto.pdf
 
Borderless Access - Global Panel book-unlock 2024
Borderless Access - Global Panel book-unlock 2024Borderless Access - Global Panel book-unlock 2024
Borderless Access - Global Panel book-unlock 2024
 
ISONIKE Ltd Accreditation for the Conformity Assessment and Certification of ...
ISONIKE Ltd Accreditation for the Conformity Assessment and Certification of ...ISONIKE Ltd Accreditation for the Conformity Assessment and Certification of ...
ISONIKE Ltd Accreditation for the Conformity Assessment and Certification of ...
 
MoneyBridge Pitch Deck - Investor Presentation
MoneyBridge Pitch Deck - Investor PresentationMoneyBridge Pitch Deck - Investor Presentation
MoneyBridge Pitch Deck - Investor Presentation
 
Cracking the ‘Business Process Outsourcing’ Code Main.pptx
Cracking the ‘Business Process Outsourcing’ Code Main.pptxCracking the ‘Business Process Outsourcing’ Code Main.pptx
Cracking the ‘Business Process Outsourcing’ Code Main.pptx
 
A flour, rice and Suji company in Jhang.
A flour, rice and Suji company in Jhang.A flour, rice and Suji company in Jhang.
A flour, rice and Suji company in Jhang.
 
Michael Vidyakin: Introduction to PMO (UA)
Michael Vidyakin: Introduction to PMO (UA)Michael Vidyakin: Introduction to PMO (UA)
Michael Vidyakin: Introduction to PMO (UA)
 
WAM Corporate Presentation Mar 25 2024.pdf
WAM Corporate Presentation Mar 25 2024.pdfWAM Corporate Presentation Mar 25 2024.pdf
WAM Corporate Presentation Mar 25 2024.pdf
 

Next generation-product-recommendation

  • 1. An Oracle White Paper March 2011 Next-Generation Product Recommendations
  • 2. Next-Generation Product Recommendations Introduction ......................................................................................... 1  The Potential Value of Recommendations Solutions .......................... 2  Avoiding the Risk of Overwhelming Shoppers with Too Many Choices ................................................................... 2  Flaws in First-Generation Recommendations Solutions ..................... 3  Inadequate Understanding of Shoppers’ Interests and Preferences ............................................................... 4  Inadequate Reach Into and Understanding of the Product Catalog .................................................................... 5  Lack of Control ................................................................................ 5  Limited Reach Across Channels ..................................................... 6  Inadequate E-Commerce and Retailing Expertise .......................... 6  A New Generation of Recommendations Solutions ............................ 6  Better Understanding of Visitors’ Interests and Preferences........... 7  Greater Reach Into the Catalog ...................................................... 8  Greater Reach Out to Customers ................................................... 9  More Control to Refine Selections .................................................. 9  Relationship with a Well-Established Provider with Retail Expertise and Focus .................................................... 11  Proven Impact ................................................................................... 11  The Positive Experience at Tommy Hilfiger .................................. 11  The Oracle Recommendations On Demand Solution ....................... 13  Conclusion ........................................................................................ 13 
  • 3. Next-Generation Product Recommendations Introduction Retailers can enhance the value of their e-commerce business by adding personalized product recommendations solutions to their online stores. These solutions assess a shopper’s interests and preferences and then recommend merchandise likely to appeal to that shopper. Retailers using automated recommendations solutions can more effectively convert visitors to buyers, increase the value of transactions, and generate more repeat purchases from loyal customers. Despite the potential benefits, relatively few e-commerce sites have adopted recommendations solutions to date because of several critical flaws in the first-generation of solutions: • The inability to understand shoppers’ preferences and present relevant recommendations • A lack of control provided to retailers to manage recommendations • Limited reach into and understanding of each product catalog • An expensive and lengthy implementation process • An inadequate understanding of how recommendations fit into the overall e-commerce and retail strategy A new generation of recommendations solutions that resolve these flaws is now available. Leading retailers who have adopted these improved solutions have realized greater return on their investment. Visitors who click on personalized recommendations convert up to three times more often and spend up to 30 percent more than those who buy without clicking on recommendations. This white paper discusses the flaws of the earlier automated recommendations solutions and the advantages of the second-generation solutions. The paper also illustrates how a leading retail brand has implemented an automated recommendations solution from Oracle and realized a significant positive impact on the business. 1
  • 4. Next-Generation Product Recommendations The Potential Value of Recommendations Solutions With e-commerce well into its second decade, the keys to sustained online success are becoming well understood. Online-only retailers, established bricks-and-mortar retailers expanding online, or manufacturers looking to sell direct to consumers online must achieve certain objectives in order to maximize revenue from e-commerce platforms: • Attract visitors and convert them into buyers. Retailers must draw as many people to their various online shopping channels as possible; present them with attractive, relevant offers; and entice them to purchase. • Increase the size of each transaction. Just as in-store sales associates effectively cross-sell buyers before purchase, successful online retailers must also cross-sell and up-sell merchandise to online buyers to raise the average value of each transaction. • Generate sales from returning customers. Retailers must build and retain a loyal customer base in order to sell more online merchandise, while reducing the marketing costs necessary to attract new customers. In order to achieve these objectives, online retailers must accurately and consistently perform one key task: understand what visitors want to buy and direct them to the appropriate merchandise as quickly as possible. Online retailers must accurately and consistently perform one key task: understand what visitors want to buy and direct them to the appropriate merchandise as quickly as possible. Avoiding the Risk of Overwhelming Shoppers with Too Many Choices Retailers must perform this task for both their in-store and online visitors, although doing so is more challenging in the online world. Freed from shelf-space limitations, online stores typically present a much broader array of merchandise to shoppers. This wider selection can make it more difficult for shoppers to quickly find what they’re looking for—too many items can overwhelm shoppers with too many choices. Instead of finding precisely what they want, shoppers get confused, lost, frustrated, and often leave before buying. When shopping online, leaving is as easy as a new Google search or a few clicks of the back button. Retailers have largely turned to offering recommendations as a way to keep shoppers on their sites. Early recommendations solutions primarily involved manual product-to-product associations. Online merchandisers, relying on their knowledge of the catalog, Website analytics, and buying trends, manually connected different items of merchandise. 2
  • 5. Next-Generation Product Recommendations Most retailers today understand that manual product recommendations, while providing marginal increases in conversion rates and order values, are limited in scope and difficult to maintain. In the case of a large, rapidly changing catalog, manual recommendations can be applied and maintained only for the best-selling items. The “long tail 1 ” of the catalog is neglected. Manual recommendations solutions have largely been replaced by automated ones. Driven by computer algorithms that analyze shopper activity, these automated solutions improve on the early, manual versions by automating the process of linking different products. Online shopping sites that offer automated, personalized product recommendations can help shoppers to more easily discover relevant products as quickly as possible. Unlike other navigational tools like site navigation and site search, recommendations quickly assess each shopper’s unique interests during each visit. The solutions provide “personal discovery” by continually directing shoppers toward merchandise that more aptly suits their current interests and preferences, and steering them away from irrelevant items that may delay and distract them. Automated recommendations solutions are delivered as on-demand services to any e-commerce site and other online channels. Most retailers today understand that manual product recommendations, while providing marginal increases in conversion rates and order values, are limited in scope and difficult to maintain. In the case of a large, rapidly changing catalog, manual recommendations can be applied and maintained only for the best-selling items. Research confirms the potential effectiveness of recommendations solutions. Shoppers respond favorably to the greater convenience, they appreciate the shopping advice, and they are more satisfied with their purchases. According to one study, more than three-quarters of online shoppers responded that recommendations improved their online shopping experience. Despite these improvements, however, the first-generation automated product recommendations solutions are flawed by their inaccuracy and inflexibility. As a result, automated recommendations have not been widely adopted. Flaws in First-Generation Recommendations Solutions Despite the potential advantages of adding recommendations to e-commerce sites, and the ease with which they can be implemented, many online retailers still have not adopted these solutions. While many retailers are aware of the potential value, and more than a dozen automated recommendations solutions are commercially available, only 27.6 percent of U.S. online retailers are offering personalized recommendations to their visitors, according to market research from eMarketer. 2 Research indicates that only 27.6 percent of online retailers have currently implemented automated recommendations. 1 A term first coined by Wired magazine Editor-in-Chief Chris Anderson, the long tail refers to the theory that products in low demand or with low sales volume can collectively make up a huge market whose size rivals or exceeds that of the bestsellers. 2 eMarketer, “Personalized Product Recommendations: Predicting Shoppers’ Needs,” March 2009. 3
  • 6. Next-Generation Product Recommendations The limited adoption of recommendations solutions can be explained in part by the fact that many online retailers are still relatively new in their experience with e-commerce. These organizations may still be focused on building and refining the core elements of their online presence: keeping product information up-to-date, presenting multiple product views, improving the purchase process, and establishing a consistent brand across all shopping channels. Until now, these higher priorities may have drawn resources and attention away from recommendations solutions. 3 Perhaps a more important factor that has inhibited adoption of automated recommendations is the existence of certain critical and widely recognized inadequacies in many of the commercially available first-generation solutions. Despite the fact that nearly half of the retailers responding to a survey on e- commerce technologies and practices cited the strong potential for personalization technology like recommendations, only eight percent indicated that the currently available solutions are effective. 4 Let’s explore some key deficiencies of first-generation solutions. Inadequate Understanding of Shoppers’ Interests and Preferences Many first-generation automated recommendations solutions are unable to accurately determine each shopper’s current interests and preferences. As a result, they are unable to present relevant choices. Irrelevant recommendations can cause shoppers to lose confidence in the retailer, which, over time, can have a negative impact on the retailer’s brand. First-generation solutions have difficulty determining and predicting shoppers’ interests primarily because they don’t take advantage of all available information about each shopper. Many solutions consider only historical purchase behavior, relying exclusively on an individual shopper’s past purchases or the past purchases of similar shoppers. As a result, they often recommend items that are no longer relevant. This would be the equivalent of an in-store sales associate suggesting that a shopper who purchased snow tires in January might now be interested in a snow blower, despite the fact that it is now July and the shopper has moved to southern California. Other solutions use only real-time clickstream data for each visitor—the number of times and order in which a visitor clicks on particular items. An online drugstore shopper browsing vitamins, for example, may only see recommendations for other vitamins, when in fact the shopper’s past two visits (which did not result in a purchase) focused on hair dryers. By failing to remind the shopper of previous items that were browsed but not purchased, first-generation solutions miss out on key opportunities to increase order values. By relying too heavily on clickstream data, these solutions can also be deceived by temporary “flash traffic” patterns. For example, a large ad campaign may drive a large amount of Web traffic to a 3 Retail Systems Research (RSR), “Survey of Retail Executives Worldwide”, June 2008. 4 Ibid. 4
  • 7. Next-Generation Product Recommendations specific product page in a short period of time, causing the recommendation engine to think that product has become extremely popular. Overrecommending that product when many shoppers would have purchased it anyway is a waste of online merchandising inventory. By failing to consider the panoply of useful information about each shopper, many first-generation solutions simply cannot provide relevant recommendations. As the later sections of this white paper will reveal, second-generation solutions have been built to consider all relevant data points, such as • The shopper’s geographic location • The time of year, day of week, and hour of day • The shopper’s referring site or search engine keyword • What other parts of the site have been visited in current or past sessions Inadequate Reach Into and Understanding of the Product Catalog First-generation automated recommendations solutions do not take the time to learn about an online retailer’s unique product catalog and how shoppers tend to browse that catalog. Instead, they “discover” products in the catalog only when shoppers click on them. This lack of native catalog awareness and understanding can have significant limitations that ultimately reduce the value that these solutions can provide. For one, first-generation solutions without catalog understanding have difficulty recommending products that are newly added to a retailer’s product catalog. They can only recommend these products once a certain number of shoppers have discovered them. This can be a significant drawback because new items, by their nature, require promotion to be discovered by shoppers. Because of their overreliance on shoppers’ clickstream data to form product relationships, these solutions also fail to recommend merchandise equally. They tend to overrecommend popular items because they are clicked on more often. This would be the equivalent of an in-store sales associate only recommending items on the end-caps and never looking at merchandise on the upper shelves in the aisles—often the items that are most difficult to find. Limitations of first-generation automated recommendations solutions include • Inadequate understanding of shoppers’ interests and preferences • Inadequate reach into and understanding of the product catalog • Lack of control • Limited reach across channels • Inadequate e-commerce and retailing expertise Lack of Control First-generation automated recommendations solutions do not give retailers adequate control to refine or filter recommendations before they are presented to shoppers. These solutions are essentially “black boxes” that retailers find unpredictable and difficult to overrule when necessary. Retailers require the flexibility, for example, to present merchandise inside of the brand or category of items that the shopper is already considering. They may want to feature only top-selling items on 5
  • 8. Next-Generation Product Recommendations certain sections of the online store or only discounted items on others. They may want to up-sell a shopper only after he’s added an item to the cart, by presenting only higher-priced alternatives. Or retailers may wish to make recommendations based on their inventory, avoiding out-of-stock items or favoring overstocked merchandise. Earlier automated recommendations solutions limit retailers’ flexibility to make these refinements, mainly due to their lack of catalog understanding, as discussed previously. They may be able to make refinements based on what shoppers are clicking on (for example, only show products in the same brand as current product), but cannot easily filter items based on specific attributes in the catalog such as margin, inventory level, and price. Limited Reach Across Channels Consumers today follow a shopping and buying process that may involve multiple channels, which they use in no particular order. They may research a product using their smartphone, view the item in a store, read reviews on a social network site, and seek out promotions in e-mail campaigns before they ultimately buy the product on the Website. First-generation automated recommendations solutions, however, generally present recommendations only to the Website. As a result, they fail to influence shoppers at other key decision-making points in the purchase process. They are unable to reach shoppers with targeted recommendations via e-mail, mobile stores or apps, off-site shopping widgets, social networks, or the retail contact center. Inadequate E-Commerce and Retailing Expertise Many companies offering the first-generation automated recommendations solutions lack a thorough understanding of e-commerce and retailing. They do not have established expertise, and their solutions are not focused exclusively on the needs of e-commerce companies who sell products from an online catalog. As a result, many of these vendors have difficulty providing effective support and incorporating the automated recommendations solution into a complete e-commerce strategy. A New Generation of Recommendations Solutions Even with their limitations, first-generation recommendations solutions have demonstrated the value of automated merchandising. Retailers do recognize that well-designed automated merchandising solutions can be effective in helping them achieve critical e-commerce objectives: converting shoppers to buyers, increasing revenue per transaction, and generating sales from returning customers. Achieving these objectives, however, requires a new generation of recommendations solutions that overcome the flaws of their predecessors. Fortunately for online retailers, a second generation of automated, personalized recommendations technology has come to market. These newer solutions resolve the flaws of the early solutions: • They present shoppers with more-relevant, predictive recommendations. • They reach into and understand each retailer’s unique product catalog. 6
  • 9. Next-Generation Product Recommendations • They give online merchandisers the ability to refine and control recommendations. • They can reach shoppers through multiple channels. • They are developed and supported by well-established technology partners that understand e-commerce strategy and the retail market. Better Understanding of Visitors’ Interests and Preferences Second-generation recommendations solutions take into account a more complete range of information to assess each visitor’s interests and preferences. Unlike first-generation solutions, they can not only account for each shopper’s past behavior or the past behavior of others (the “wisdom of the crowds”) but also make judgments based on each shopper’s current session, as well. For example, they can incorporate information on geography, time of day, customer status, and how the visitor navigated through the online store. With this more accurate understanding of each shopper’s needs, retailers can better predict that shopper’s next likely click and present more-relevant merchandise. Just as an in-store sales associate would be unlikely to recommend a snow blower to someone now living in southern California, second- generation recommendations solutions would prevent this inappropriate recommendation from being presented. Figure 1 illustrates how second-generation recommendations improve understanding of visitors’ interests and preferences. Figure 1. Second-generation recommendations solutions overcome flaws of early solutions by taking into account a more complete range of information to assess each visitor’s interests and preferences. 7
  • 10. Next-Generation Product Recommendations Greater Reach Into the Catalog Second-generation recommendations solutions reach more broadly across the entire merchandise catalog in order to present selections that best suit each shopper’s preferences. Unlike earlier solutions, which often were unaware of new or niche items, the new solutions ingest a live feed of the product catalog regularly, form dynamic relationships among products as they are added, and are aware of new items and changes to items in the catalog. Second-generation solutions act like a knowledgeable in- store sales associate, who knows precisely what item the shopper is looking for and where to find it and can quickly guide the customer to it. Figure 2 illustrates personal merchandise displays. Figure 2. Using a full range of information about the shopper, retailers can now deliver cross-sells on product detail pages that are dynamic and personalized based on each shopper’s interests. Extended reach across the catalog also allows retailers to use recommendations solutions for more than product cross-sells. Instead, these solutions can be used as automated merchandising tools that can take over where online merchants leave off. Today’s second-generation solutions can be used to automate and personalize merchandising across the entire online store. These broad uses of automated merchandising may include • “Just for you” slots on the home page, driven by all previous and current data available about each shopper • Search engine landing pages that use the incoming keyword search to automatically display recommendations and relevant merchandise • Null search result pages that use the site search keyword to dynamically display relevant merchandise 8
  • 11. Next-Generation Product Recommendations • Top-sellers pages showing personalized product selections, but limited to top-selling items • Gift guides showing a range of dynamic gift ideas personalized to each shopper’s current interests Greater Reach Out to Customers Today’s second-generation automated recommendations solutions can reach out proactively across an array of shopping channels. Based on the information accumulated about a shopper’s interests and preferences, retailers can dynamically present attractive and appropriate merchandise via e-mail, mobile stores or apps, off-site shopping widgets, social networks, or the retail contact center. Figure 3 illustrates this capability through e-mail offers. Figure 3. Today’s second generation of automated recommendations solutions can reach out proactively to shoppers and present personalized offers in e-mail marketing campaigns. More Control to Refine Selections To overcome the lack of control in first-generation solutions, second-generation automated recommendations technology now allows retailers to carefully refine and filter recommendations before they are presented to shoppers. This allows merchants to leverage the benefits of 9
  • 12. Next-Generation Product Recommendations personalization and automation, while at the same time ensuring that solutions adhere to their merchandising guidelines or strategy. Recommendation refinements can be based on catalog data, visitor behavior, or a combination of the two. • Refinements based on catalog data allow retailers to recommend merchandise based on particular product attributes such as category, brand, price, collection, top-sellers, and new products. Merchants can also create custom refinements based on their unique catalogs, mixing “fixed” recommendations of specific products or excluding specific products from automated recommendations. • Second-generation recommendations solutions also allow retailers to make recommendations based on common types of visitor behavior. For example, merchants may want to ensure that shoppers looking at full-priced items are presented with only full-priced recommendations, whereas budget- conscious shoppers see recommendations for discounted items. Recommendations can also be refined based on common keywords that shoppers key into a Website search engine. For example, all visitors who searched on “TVs” and click on the search engine link for a retailer might see only personalized TV recommendations on the home page. Sophisticated retailers can combine catalog data and visitor behavior to create highly targeted, automated merchandising displays. For instance, a budget-conscious shopper shopping in a specific brand may see only on-sale, in-brand recommendations. Figure 4 illustrates the types of control and refinements that retailers have with second-generation automated recommendations solutions. Figure 4. Second-generation automated recommendations solutions allow retailers to carefully refine and filter recommendations before they are presented to shoppers. Recommendation refinements are generally based on catalog data, visitor behavior, or a combination of the two. 10
  • 13. Next-Generation Product Recommendations Relationship with a Well-Established Provider with Retail Expertise and Focus While earlier recommendations solutions were available primarily from technology startups with limited e-commerce experience, retailers can now work with well-established providers with a deep understanding of the retail market. These more-established providers take the time and have the resources to thoroughly understand each retailer’s unique business, focus exclusively on their e-commerce requirements, and deliver an automated recommendations and merchandising solution that fits their overall e-commerce strategy. The advantages of second-generation automated recommendations solutions include • Relevance. Present shoppers with more-appropriate recommendations based on a complete understanding of their preferences. • Reach. Present recommendations from across the entire catalog of merchandise including new and niche items, and deliver them proactively to customers. • Refinement. Flexibly control the recommendations presented to shoppers. • Relationship. Work with a well-established recommendations solution vendor with a deep understanding of the retail environment and e-commerce strategies. Proven Impact Second-generation recommendations solutions can have a significant impact on retailers’ success. The experience of leading e-commerce merchandisers has demonstrated quantifiable improvements in lifting online revenue quickly by converting more browsers into buyers, increasing the average revenue per transaction, and securing more follow-on sales from existing customers. Data from customers of the Oracle Recommendations On Demand solution, for example, shows that shoppers who click on recommendations during a session are likely to convert up to three times more often and spend up to 30 percent more than those who don’t. These dramatic increases in key performance indicators are direct evidence of the benefit of personalizing online merchandise so that shoppers can easily find what they want and discover other items that match their unique interests. The Positive Experience at Tommy Hilfiger Tommy Hilfiger, one of the world’s most recognized premium lifestyle brands for 18- to 35-year-olds, implemented an advanced recommendations solution to improve the performance of its online store. The company’s e-commerce staff recognized the potential value of presenting shoppers with appropriate recommendations during the shopping process. However, their initial experience with early recommendations solutions yielded few additional sales. For one, it was necessary to manually hard-code one-to-one product relationships. This process made it almost impossible to make changes quickly as new merchandise was added and outdated items were removed. “We’ve seen a threefold increase in online shopper conversions when someone interacts with the Oracle Recommendations On Demand solution. At the same time, our staff is spending less time than ever managing the process. The more our shoppers use it, the smarter the engine gets and the more value it delivers.” —Tom Davis, Director of E-Commerce, Tommy Hilfiger 11
  • 14. Next-Generation Product Recommendations Moreover, early solutions ignored much of the data about an individual shopper that could be used to make relevant recommendations. Recommendations were based on only one factor: items already in the shopping cart. As a result, the e-commerce site tended to present the same set of merchandise to all customers. Recommendations could not be effectively tailored to the shopper’s particular interests at the time of their visit. Upon deploying the Oracle Recommendations On Demand solution, Tommy Hilfiger saw a significant increase in sales from its online store. According to Tom Davis, director of e-commerce at Tommy Hilfiger, “We’ve seen a threefold increase in online shopper conversions when someone interacts with the Oracle Recommendations On Demand solution. At the same time, our staff is spending less time than ever managing the process. The more our shoppers use it, the smarter the engine gets and the more value it delivers.” Using a full range of information about its online shoppers, Tommy Hilfiger can now construct personalized merchandise displays, which are dynamically prepared to suit each shopper’s current interests. These displays are presented while customers are browsing the online store, and they can be sent to targeted customers via e-mail, as illustrated below. Figure 5. Using a full range of information about each shopper, Tommy Hilfiger can now construct personalized merchandise displays, which are dynamically prepared to suit each shopper. Tommy Hilfiger is powering a top-seller page (shown here) with the Oracle Recommendations On Demand solution, allowing it to personalize what top-sellers are shown to each shopper. In addition to better understanding each shopper’s preferences, the solution relieves the Tommy Hilfiger staff from the difficult task of hard-coding all cross-sell and up-sell options. This allows them flexibility to react more quickly as they introduce new products, as items go out of stock, or as they need to make other adjustments to the merchandise mix. 12
  • 15. Next-Generation Product Recommendations The Oracle Recommendations On Demand Solution Comprised of Oracle Recommendations Single-Channel On Demand and Oracle Recommendations Multichannel On Demand, the Oracle Recommendations On Demand solution is an automated, personalized merchandising service that helps online retailers quickly lift revenue by recommending the most-relevant products from the catalog to each shopper. Balancing predictive relevancy with merchant control, the solution automatically selects the products that match each shopper’s needs and each retailer’s business needs, presents them at the right time in the right channel, and demonstrates the impact on key performance metrics such as page views, conversion rates, and order values. The Oracle Recommendations On Demand solution is part of Oracle’s family of e-commerce solutions that lift revenue quickly, easily, and measurably by optimizing every online interaction. The service can be deployed easily and rapidly to any Website—even those not running on the Oracle ATG Web Commerce platform. By using the Oracle Recommendations On Demand solution to automate and personalize online merchandising, online retailers can realize the following significant benefits: • Convert more browsers into buyers by presenting the most-relevant products from the catalog to each shopper across channels. • Increase order values by automating and personalizing cross-sells and up-sells. • Retain more customers and build a core base of loyal, repeat buyers by personalizing the customer experience across channels. Conclusion Personalized product recommendations solutions can significantly enhance retailers’ e-commerce business by effectively converting visitors to buyers, increasing the value of transactions, and generating more repeat purchases from loyal customers. Research and experience with large online retailers confirms the potential effectiveness of recommendations solutions. However, first-generation automated recommendations solutions did not adequately understand shoppers’ interests and preferences. They also had limited reach into the product catalog and across channels and provided little control for retailers. Oracle is a well-established recommendations solution vendor with a deep understanding of the retail environment and e-commerce strategies. The Oracle Recommendations On Demand products—Oracle Recommendations Single-Channel On Demand and Oracle Recommendations Multichannel On Demand—are the next generation of automated product recommendations solutions. Together, they capture the full value of your online traffic flows by providing a better understanding of visitor interests, more-relevant product recommendations across the entire catalog, and a greater ability for merchandisers to refine and control recommendations, ultimately allowing your Website to achieve the best possible sales outcome from every online visit. 13
  • 16. Next-Generation Product Copyright © 2009, 2011, Oracle and/or its affiliates. All rights reserved. This document is provided for information purposes only and Recommendations the contents hereof are subject to change without notice. This document is not warranted to be error-free, nor subject to any other March 2011 warranties or conditions, whether expressed orally or implied in law, including implied warranties and conditions of merchantability or fitness for a particular purpose. We specifically disclaim any liability with respect to this document and no contractual obligations are Oracle Corporation formed either directly or indirectly by this document. This document may not be reproduced or transmitted in any form or by any World Headquarters means, electronic or mechanical, for any purpose, without our prior written permission. 500 Oracle Parkway Redwood Shores, CA 94065 Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. U.S.A. Worldwide Inquiries: AMD, Opteron, the AMD logo, and the AMD Opteron logo are trademarks or registered trademarks of Advanced Micro Devices. Phone: +1.650.506.7000 Intel and Intel Xeon are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license Fax: +1.650.506.7200 and are trademarks or registered trademarks of SPARC International, Inc. UNIX is a registered trademark licensed through X/Open Company, Ltd. 0311 oracle.com