WO 2013/052081
`
`PCT/US2012/000426
`
`[0085]
`
`[0086]
`
`[0087]
`
`[0088]
`[0089]
`
`[0090]
`
`[0091]
`
`[0092]
`
`[0093]
`
`[0094]
`[0095]
`
`[0096]
`
`[0097]
`
`[0098]
`
`[0099]
`
`[0100]
`
`[0101]
`
`[0102]
`
`[0103]
`
`[0104]
`
`[0105]
`
`[0106]
`
`[0107]
`
`[0108]
`
`[0109]
`
`[0110]
`
`[0111]
`
`[0112]
`
`[0113]
`
`20
`
`25
`
`Toys
`
`Video games/music/video/books
`
`Furniture
`
`Household appliances
`
`Jewelry
`
`Mattresses
`
`Musical instruments
`
`Pet products
`
`Powertools/lawn/garden
`
`Hardware
`
`Sporting goods
`
`Lighting
`
`Investments
`
`Healthcare
`
`Pharmaceuticals/Vitamins
`
`Equipmentrental
`
`Telco hardware andservices
`
`Cable/satellite services
`
`Apparel
`
`Children’s accessories
`
`Baby items
`
`Insurance
`
`Personal products
`
`Fine Wine and Champagne
`
`Beer
`
`Alcohol
`
`Gourmetchocolate
`
`Gourmetcoffee
`
`The system comprises hardware, software, databases and servers, processors and
`
`30
`
`communication networks both wired and wireless that provides a robust, extremely fast and
`
`highly reliable experience for target customers including consumers or Shoppers, Retailer
`
`10
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`marketing or staff personnel and brand Marketers/staffpersonnel and researchers and other
`professionals. An embodiment comprises four inter-linked technology platformstargeting
`Shoppers, Retailers and brand Marketers. For shoppers the system will enable the preparation of
`a shoppinglist or the evaluation of purchase choice(s) using a web portal or smart phone
`
`application. The system includes a multi-level analytic system that elicits from the shopper how
`
`they intend to use a product and what specific performancevariables are most important to the
`Shopperto establish purchase interestor utility for a given product offering. The system
`recommendsa sorted/prioritized list of items that best meet the shopper's requirements (Highest
`
`purchaseinterest or utility), the “acceptable set”, calculate the overall dead-net price
`
`10
`
`automatically and recommenda specific retail location for the shopper to achieve the lowest
`
`overall price or cost inclusive of available promotions and couponsfor the items selected.
`
`[0114]
`
`The application in essence automates the way a shopper normally researchers and
`
`then makes purchase decisionsthereby streamlining the purchase process. In addition, the
`
`application functions as a Personal Sales Agent, or PSA, helping guide the Shopperin self-
`
`15
`
`categorizing their intended product usageprofile so that the system can establish appropriate
`
`preferencesto aid in the down-selection of many feature rich and technical products into the
`
`“best”list of products that meet the Shopper’s intended use. The system enables bothretailers
`
`and brand marketers, herein “marketers”to deliver marketing incentives of various kinds
`
`including additional price-related incentives, directly to the Shopper during the preparation and
`
`20
`
`execution of their purchase process. These incentives are included into the net purchaseprice
`
`and could be targeted by Retailers and Marketers based on a broad number of Shopper
`
`characteristics, preferences, behaviors, purchase history and the like. The System further
`
`provides post and pre purchase analytics to help Marketers and Retails evaluate their
`
`promotional efforts, better understand the Shopperand profitably grow their respective
`
`25
`
`businesses or increase Shoppersatisfaction and loyalty.
`
`[0115]
`
`This system can influence Shoppers after they have indicated whatthey intend to
`
`purchase — but before the actual purchase has been made ~ in a streamlined highly functional
`
`application.
`FIGS. 1-10 summarize the general method. FIG. 1 is a block diagram of an
`[0116]
`exemplary system. In some embodiments,at least one database server provides access to the
`
`30
`
`system databases, includingretailer pricing/product database server 100, SKU item image
`
`11
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`database server 200, SKU/Item level retailer weekly circulars database server 300, SKU/Item
`
`level coupon offers database server 400 and comprehensive SKU Features andtechnical
`specification database server 500. Data may also be received from other external data sources
`501, such as manual data entry, automated web capture, web crawlers, secret shoppers or the
`like. The retailer pricing/product database server 100 includes: by-store, pricing zone, and
`individual item data for SKU #, Pricing, Brand/Mfg., List Price, Promoted Price, Description,
`and optionally other data. A plurality of retailers 1 ..., N access the databases 100 and 300,
`
`and a plurality of product manufactures 100, 200, ... , NOO access databases 200, 400 and 500.
`
`The data transmissions between the various servers and various non-transitory computer readable
`storage medium may bevia internet, wireless or other electronic or physical means.
`
`10
`
`FIG. 2 shows an example ofthe retailers in communication with retailer
`[0117]
`incentives database server 600, other external data sources 501 (shown in FIG.1), such as
`manualdata entry, automated web capture, or the like. retailer location based offers database
`
`server 700 , retailer advertising database server 800, andretailer logo and tag line database server
`
`15
`
`900 (which maybe hosted in the same computeror a different computer from the servers of FIG.
`
`1. The server 600 contains conditional Retailer Promotional offers (Rls) insertion rules, to offer
`
`values, items for etc. offered to Shopper by Retailer within Basket Cost Minimization Algorithm
`
`impacting the Initial Lowest Cost (ILC)list or basket. Server 700 includes Marketing incentives
`
`presented to Shoppers whentheyenter or select a specific retailer for item purchase — similar to
`
`20
`
`in-store “wall of values” or in-store circular offers. Server 800 includes video, andstill ads for
`
`display within Shopper smartphoneandinternet portal applications. Server 900 includes Retailer
`
`logo andtag line files for insertion into Shopper smartphoneapplication and Internet portal
`
`applications. Logo and tag lines are also available to Retailer and Marketerportal applications.
`[0118]
`FIG. 3 is a block diagram of an example of the product manufacturerinterface.
`
`25
`
`The product manufacturers 100, 200, NOO communication with marketing incentives database
`
`server 1000, marketer coupon offers database server 1100, marketing advertising database server
`
`1200, and marketer companyand brand logo andtag line database server 1300. Data mayalso
`
`be received from other external data sources 101, such as manualdata entry, screen scraping, or
`
`the like. Server 1000 provides conditional Promotionaloffers , insertion rules, offer values,
`
`30
`
`items or the like offered to Shopper by Marketer within Value maximization/Price Minimization
`
`function impacting the Final Lowest Cost (FLC)for item (s). Server 1100 provides coupon
`
`12
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`incentives delivered online orother off-line efforts for use in calculating Initial Lowest Cost
`
`(ILC) list or basket. Couponincentives are delivered within Shopper smart phoneandinternet
`
`portal for use in calculating Initial Lowest Cost (ILC) for item (s)or basket . Server 1200
`Includes Marketer video,andstill ads for display within Shopper smart phone andinternetportal
`applications. Server 1300 includes Marketer company logo, brandlogo, tag line files for
`insertion into Shopper smart phone application and Internetportal applications. The Logo and
`tag lines also available to Retailer and Marketer portal applications.
`
`FIG. 4 show an example of the shopper preference databases 1400, shopper
`[0119]
`information database 1500 and shopperpredictive purchase interest model (PPIM) database
`
`10
`
`1600. Database 1400 includes information such as Location/HomeStore; Other Favorite Stores
`
`— Store substitutes; Decision rules - How Shopper wants price minimization algorithm to
`
`perform relative to substitutes (e.g., lowest comparable price, price differential to choose brand
`
`substitutes, lowest price override etc.); Preferred brand(s) by category; Suitable substitutes by
`
`category (other brand, private label etc.).and Shopper ID code. In one non-limiting example, the
`
`15
`
`database 1500 includes name,address, log-in name and ID, demographic information,
`
`psychographic and behavioral information, item purchase history; basket/ring purchase history;
`
`promotion responsehistory; preferences information; pricing differential response history; store
`
`preference history. In other embodiments, other data and/or combinationsof data are included.
`
`Database 1600 includes Shopper ID code; Shopperbasic category needs & usage (BCNU)
`
`20
`
`promptclassification question responses (e.g., for determining the shopper's presets); Shopper
`
`Predictive Purchase Interest Model (PPIM)results from feature set prompt and feature driver
`weights; Shopper PPIM self-weighting inputs; Refined PPIM feature driver weights based on
`actual purchase.
`In some embodiments, the system uses learning for refining the PPIM feature
`driver weights.
`|
`[0120]
`In one embodiment the shopper preferences of the shopper database of product
`
`25
`
`substitutes (SdBP)are initially set based on presets that are designed to representdifferent
`
`product preferences relating to various shopperaffinity, demographic or psychographics.
`
`In one
`
`preset embodimentpreferencesare set for large families favoring larger sizes. In one preset
`
`embodimentpreferencesare initially set to include a greater representation of premium and so-
`
`30
`
`called super premium products. In one preset embodimentinitial preference presets are set to
`
`include those products more appropriate to smaller households. Other preset embodiment can be
`
`13
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`6e
`
`“scratch cookers,”
`
`o>
`developedfor various groups including older shoppers, “foodies”,
`savings focused”etc.
`.
`{0121]
`FIG. 5 shows an exampleof additional databases accessed by the system as
`secondary sources. Productlevel reviews database includes a database of product reviews from
`Shoppers and a Database ofproduct reviews from industry experts and other expert reviewers.
`Retailer reviews database 1800 includes Databaseof Retailer reviews by Shoppers, and Database
`of other Retailer information (# of complaints, # of units sold etc.). These databases may receive
`data from public sourcesavailable on the Internet, and/or proprietary sources, such as the
`registration information collected by the system, and information exchangeswith partner systems,
`websites, loyalty programs, manually input data, screen scraping, andthelike.
`
`39°66
`
`“aggressive
`
`10
`
`[0122]
`
`FIGs. 6A and 6Bare flow charts of an example of how Buyer value/preference
`
`information can be determined to streamline and automate purchase selection and decision
`
`process in complex goods. Referring first to FIG. 6A,at step 601, the shopper picks a product
`
`category for purchase. At step 602, the shopper is prompted to answerbybasic category needs
`
`15
`
`and usage (BCNU)questions. Theresults are stored in the shopper predictive purchase interest
`
`model (PPIM)database 604, At step 603, the BCNU algorithm developsthe initial shopper
`
`segmentation and feature importance hypothesis. At step 605, the shopper is prompted with
`
`various feature-price pairs that cover choice set and is asked to indicate purchase interest At step
`
`606, Shopper inputs information form the Predictive Purchase Interest Model (PPIM) andsets
`
`20
`
`baseline variable weights. These results are stored in the data base 1600. At step 607, the
`
`shopperis prompted with questions seeking to refine PPIM by probing inconsistencies between
`
`BCNUhypothesis and PPIM view (if any). Referring now to FIG.6B,at step 608 the PPIM is
`applied to all product/ feature options to calculate “fit” or purchase interest score. At step 609,
`the shopperis presented with product choices with the highest PPIM score and, in some
`
`25
`
`embodiments, additional subjective information.(e.g., reviews).
`
`. At step 610, the Shopper uses Purchaseinterest sliders to refine recommended
`[0123]
`product set. At step 611, the shopperselects a product. At step 612, the system incorporates RIs
`
`and presents Shopper with Retailer Recommendation and MIs. At step 613, Shopper makes
`
`purchaseelectronically or at Retail location. The results of the actual purchase are compared to
`
`30
`
`the PPIM 1600.
`
`14
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`[0124]
`
`FIG. 7 is a block diagram of an example of a system 701. A shopper application
`
`702 receives inputs from the user (shopper) and displays results. A retailer web portal campaign
`
`managementdisplay and input system providesthe interface to retailers. A marketer web portal
`
`campaign managementdisplay and input system 704 provides the marketinterface.
`
`Analytics/research web portal display and input system 705 providestools for the retailers and
`
`marketers. The databases used by the system include the product database including
`
`products/SKU-level information 100, 200, 300, 400, 500. The retailers provide Retailer and
`
`Marketer Incentives, Ads and Other Information 600, 700, 800, 900, 1000, 1100, 1200, 1300.
`The analytic cost minimization module 701 receivesall of the above inputs and determines
`
`10
`
`Shopper Characteristics 1400, basic Category Needs & Usage (BCNU) 1500 Predictive
`
`Purchase Interest Model (PPIM) 1600 and Product Reviews 1700.
`
`[0125]
`
`FIG. 8 is a flow chart for the method of Making a List and Identifying what and
`
`where to buy item spending the minimum amount within defined Shopperpreferences. At step
`
`800, the shopperlogs on to the website or starts a streamlined version of the shopper app on the
`
`15
`
`mobile device. At step 802, the shopperselects a category ofinterest (e.g., cameras). At step
`
`804, the shopper responds to Questions to inform BCNUandPredictive Purchase Interest Model
`(PPIM). At step 806, the PPIM recommendsa productlist. At step 808 and 810, the system
`
`issues one or more non-on-list (NOL) couponoffers for consideration. At step 812, the shopper
`
`selects “Done” endinglist creation and initiating cost minimization algorithm. At step 814, the
`Minimization Algorithm Calculates DNPfor each product recommended by PPIMatall alternate
`e-Retail/Retail locations. At step 816, the minimization function Identifies “Initial Lowest Cost”
`
`(ILC) item and basket option and Retail location providing. At step 818, a determinationis
`
`made whetherthere are any retailer incentives (RIs). If so, step 820 is performed. If not step 822
`is performed. At step 820 the Minimization Algorithm incorporates RI’s and recalculates basket
`cost, now the “Subsequent Lowest Cost” basket and Retail Location providing. At step 822,if
`
`20
`
`25
`
`. there are no RIs, the ILC becomes the SLC. At step 824, SLC is displayed to Shopper along
`
`with MIs. At step 826, a determination is made whetherthere are any marketing incentives
`
`(MIs). If so, step 830 is performed. If not step 828 is performed. At step 830, the Minimization
`
`Algorithm recalculates basket or item cost including all MI’s — final SLC becomes“Final Lowest
`
`30
`
`Cost (FLC) item(s) or list. At step 828 the SLC becomes the FLC. At step 832, the Lowest cost
`
`for item and Retail store (FLC) presented to Shopper for shoppingtrip/e-purchase.
`
`15
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`[0126]
`
`FIG. 9A is a flow chart showing Retail store, and FIG. 9B showsthe
`
`corresponding flow for an eRetail store, The shopper uses FLClist to shop for item(s) in
`
`streamlined fashion. In FIG. 9A,atstep 902, the shopper takes the shoppinglist to the retail
`store on the smart phone or other mobile device. At step 904, as Shopperenters store, Shopperis
`presented with a splash page of “In-Store Incentives” (ISIs) or Ad. At step 906, a determination
`is made whether ISI is accepted. If so, step 908 is performed, and the selected Items(if any)
`addedto list and basket total. At step 910, the shopper picks up itemsonthe list. At step 912, the
`
`shopperchecksout, (optionally scanning his/her smart phone or other mobile device) to record
`
`and validate purchases, redeem e-coupons andcreate valid redemption record for coupon
`fulfillment In some embodiments, couponsand other promotional offers are validated and
`fulfilled by the displaying of a scanner bar, shopper number, frequent shoppercard, radio signals,
`
`WIFIsignal, blue tooth signal, near field communication, LED modulatedlight interacting with
`the scanner system or other electromagnetic, sound, visual means or other methods.
`[0127]
`In FIG.9B, at step 914, the Shopper goes to eRetailer and enters tracking code.
`At step 916, the Shopper is presented with a splash page of additional incentives, if any (ISIs) or
`
`10
`
`15
`
`Ad. At step 920, the selected Items (if any) added to list and basket total. At step 922, the
`
`shopper proceeds to checkout. At step 924, shopper checksout, entering tracking code to receive
`
`any additional incentives, ecoupons andrecords the purchase.
`
`[0128]
`
`In some embodiments, whenthe userinitiates checkout, the mobile device
`
`20
`
`determines whetherall of the items on that list have been checked off. FIG. 22 shows a warning —
`
`issued if there is one or more item onthelist that has not been checkedoff, and askingif the
`
`shopper wants to return to the list and continue shopping.
`
`[0129]
`
`In some embodiments, when the useris finally ready to check out, the mobile
`
`device displays a bar code and numberrepresenting the customer's loyalty card numberto be
`
`25
`
`scannedor read in by the cashier, as shown in FIG. 23. The Shopper may use device-to-device
`
`communication methods other than barcode as enabled by system including: Bluetooth or
`cellular (eg CDMA)wireless transmission, near field communication, LED, modulated light,
`audio (sound wave) or other means.
`
`[0130]
`
`The first componentplatform, herein the “Shopper App,” is web portal and
`
`30
`
`Smartphone application for consumersor herein “Shoppers”that: '
`
`16
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`[0131]
`Allows Shoppersto enter a specific product category of interest (eg. “Cameras”)
`forming the initial starting point for determining how a Shopper intendsto use theitem and what
`is important to the Shopperfor establishing which specific products to research.
`
`[0132]
`Enables the prompting of the Shopper via web or smartphoneinterface to answer
`a series of questionsrelating to the category and how the Shopperintendsto use the product —
`herein called “Basic Category Needs & Usage” (BCNU)questions. BCNU questions may form
`
`an initial segmentation hypothesis for how Shopperwill use product and what products are
`
`relevant to the Shopper.
`[0133]
`Optionally presents the shopper with a series of product feature/price
`combinations to understand how Shopperassigns value or purchase interest to different products.
`
`Answers to prompts will be used to create a Predictive Purchase Interest Model (PPIM)that can
`
`be used to evaluate any combination of product feature and price and will provide a estimate of
`whether new product meets Shopper needs (and how well). PPIM categorizes the shopperinto a
`buyer group or market segmentand link this segmentto a set of products
`[0134}
`Optionally prompts the shopper with questions to further develop the PPIM based
`
`on and found discrepancies/anomalies between the predicted market segments and products and
`
`10
`
`15
`
`the BCNUresponses.
`
`[0135]
`
`Uses the shopper's responses to the questions to refine the PPIM modeloractual
`
`behavior.
`
`20
`
`[0136]
`
`Presents the shopper with a list of 1-10,000 product/feature and price options.
`
`Product options are ranked by the over “Fit” or purchase interest as determined within the PPIM
`
`model. The productlist includes links to peer and professional reviews, review key-word search
`
`database and other subjective or related information(reliability data).
`
`{0137}
`
`In some embodiments, using a web inerface, presents “slider” controls on key
`
`25
`
`predicative product features for Shopper to make adjustments to values, impacting
`
`recommendationlist; Recalculates value and re-presents when sliders were changed. Showing
`
`original rank next to items. Final slider settings are recorded in system for post purchase
`
`validation.
`
`[0138]
`
`Calculates for the shopper a lowest cost for the item orlist of items so as to
`
`30
`
`minimize the individual item andtotal basket cost within shopper defined or presetsorinitial
`
`setting preference parameters taking into account all available or “live” price discount
`
`17
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`mechanismsofany kind stored in system database. Thislist initial recommendationis herein
`
`called the “Initial Lowest Cost”list (ILC).
`
`[0139]
`
`Allowsfor Retailer Incentives (RIs) to alter the recommendedretail store
`
`outcome presented whenthe ILC is calculated for the shopperby the inclusion of a post-ILC
`
`incentive that further reduces the basket cost and shifts the cost minimization “win” from one
`
`Retailer to another as presented to the shopperin a final list/recommendation. Thislist is herein
`
`called the Subsequent Lowest Cost List (SLC).
`
`If noretailer incentives are active the SLC and
`
`ILC are identical and the shopper is presented with the ILC. Otherwise, if a RI changes the
`
`minimization outcome, the shopper sees the SLClist only.
`
`10
`
`[0140]
`
`Displays recommendlist of items for the shopper that includes a “check box” icon
`
`or other similar graphic device or icon for different choices.
`
`[0141]
`
`Allows the shopper to choose a recommendedproduct option and then displays
`
`additional Manufacturer Incentives (MIs) that may be available. As used herein, MI's are any
`
`form of promotional offer that is displayed to the shopper along with the final lowest cost (FLC)
`
`15
`
`list. MI's can be planned andinserted into the shopper app by at least one web-based campaign
`
`management system. MI's can be inserted by marketers, retailers, ad agencies or interested third
`
`parties.
`
`20
`
`Displays MIs that may be accepted by shopperand addition of other items or
`[0142]
`more units or larger size or the like, which result in: the replacementof item in shopping basket
`that was immediately adjacentto icon, recalculation of basket cost and inclusion of additional
`incentives relevant to item and basket costing and item cost.
`.
`[0143]
`After MIs are reviewed and selected a new final recommendation and basket cost
`
`will be calculated with a total cost for the basket and savings versus the néxtretail option
`
`displayed. Onceal] MIs are considered or no moreare selected the list will be a final list. This
`
`25
`
`list is herein called the Final Lowest Costlist (FLC), In some embodiments, the Shopper can
`
`then select an option or trigger that orders that basket list for more rapid product selection and
`
`purchase at the recommended store. In some embodiments,the list is ordered by item category.
`
`In some embodiments,the list is organized by location in the store. In some embodiments,the
`
`list is organized by store department. In some embodiments,thelist is organized by shelf
`
`30
`
`position.
`
`18
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`Once the shopper has completed shopping, and items can be purchasedeither at a
`[0144]
`physical retail outlet or at an eRetailer. The system will allow the shopper to use their
`smartphoneapplication to present a scan-able bar code, blue-tooth signal or near field , Led
`
`signal, number communication (NFC) enabled signal to record the purchases andlink electronic
`coupons, and shoppercard incentives to the shoppers' identity, to reduce basket cost
`appropriately and provide purchaserecord for validation. Alternatively, the system will present
`the shopper with a purchase promotional code that will verify shopperidentity, trigger
`incorporation of various shopper discounts and incentives (RIs and MIs) and record purchase in
`shopper information database and PPIM validation system.
`
`10
`
`[0145]
`
`The shopper App captures servers and databases, consumer demographic
`
`information, purchase information, preferences and decision rules and other informationrelating
`to purchase behavior.
`
`[0146]
`
`The shopper App and supporting sytems is compatible with iPhone (Apple OS),
`
`Android (Android OS) , Blackberrydevices (RIMOS), Microsoft OS andother suitable
`
`15
`
`smartphone or mobile device operating systems, tablets and other mobile devices..
`
`[0147]
`
`Allows the shopper to share a shopping list and final purchase price with social
`
`
`
`networking peers.
`
`Technical requirements
`
`
`
`Databaseof product specific
`Webor Smartphone interface prompts the
`1. Shopper inputs
`
`
`information including
`Shopperwith a series of category specific
`
`preference
`
`specifications, features,
`
`questions to determine what the key areas
`
`information
`
`
`performance and pricing.
`of need or benefit are for a given Shopper
`
`
`
`and whatthe relative importanceis for the
`
`Mechanism for Shoppers to
`various performancedrivers within the
`
`input preferences in a Web
`product category. These questions form
`
`
`
`or Smartphone format
`
`the “Category Needs & Usage (BCNU)
`
`
`througha series of
`questions and help initial Shopper
`
`
`intelligent prompts.
`Segmentation and product
`
`
`
`recommendations. BCNU prompts include
`
`
`Algorithms to tease out
`questions on intended usage,self-
`
`
`
`benefit/need prioritization,
`categorization as “professional” or
`
`
`
`“amateur” and other relevant, category
`feature driver importance
`
`
`
`and weight in purchase
`specific criteria. Eg. for computers; laptop
`
`
`
`
`decision. Algorithms to
`or desktop, student, professional, screen
`
`
`develop a predictive
`size, business or gaming, word processing
`
`
`purchase intent modelfor
`or graphics intensive etc.
`
`
`
`
`19
`
`

`

`WO 2013/052081
`
`PCT/US2012/000426
`
`
`
`various product
`
`
`feature/price option by key
`The Shopper responsesare then used in an
`
`Shopper segmentsand for
`algorithm to form an initial BCNU model
`
`
`
`individual Shopper.
`that classifies the shopper into a shopper
`
`
`
`
`segmentlinked to specific product
`performance needs.
`
`The shopperis then prompted with a series
`
`of product feature-price options that
`provide the analytical input to a Predictive
`
`PurchaseInterest Model (PPIM) that are
`used to evaluate all category product
`offerings and establish a “Fit” or relative
`
`purchaseinterest value representing how
`
`well a specific offering meets the Shopper's
`
`needs.
`
`
`
`
`The shopperis then prompted with a few
`questions to resolve any discrepanciesin
`responsesand predictions between the
`BCNU and PPIM inputs.
`
`
`
`
`
`in some embodiments, using a web
`
`interface, the PPIM then presents from 1-
`10,000 product options ranked on howwell
`
`they meet Shopper needs(or Fit) and Price.
`
`Items are presented with links to peer and
`
`professional reviews and other functional
`characteristics
`
`
` In some embodiments, the system allows
`
`the shopper to control key purchase
`
`
`interest drivers by a series of “slider”
`controls post the PPIM first listing. PPIM
`recalculates “fit” and redisplays a new list
`
`after each slider change. The original fit
`
`recommendations are maintained as
`
`
`“Recommendation Rank”on display.
`
`
`
`
`The PSAevaluates a set of acceptable
`2. Shopper
`Databaseof product specific
`selects from an|purchase options based on preferences
`information including
`
`“acceptable
`determined in step 1.
`specifications, features, and
`
`
`
`
`20
`
`