DESCRIPTION
TECHNICAL FIELD
In the field of online shopping of consumer products, a method and system are disclosed to provide recommendations to the advertisement of the consumer products on online marketplaces.
DEFINITIONS
Communication Network: A network of communication devices and stations having wired or wireless interconnection for establishing communication. Communication network includes, but is not limited to, internet, intranet, extranet. Wide Area Network (WAN), wireless WAN (WWAN), Local Area Network (LAN), wireless LAN (WLAN), transducer links such as those using Modulator-Demodulators (modems), telecommunication network, personal area network and Global Navigation Satellite System (GNSS). Telecommunication network includes, but is not limited to, Public Switched Telephone Network (PSTN), Global System for Mobile Communications (GSM), and Code Division Multiple Access network (CDMA). Personal area network includes, but is not limited to, Bluetooth and Infrared, and Global Navigation Satellite System (GNSS). Communication includes, but is not limited to, transmitting and receiving signals. Communication further includes, but is not limited to, transferring information and data such as voice, audio, video, graphics and the like.
Product: A product is a physical commodity manufactured by a person or organization. A product may also be a service or a set of services provided by a person or organization. For example, a product may be software, an electronic item such as camera, cell phone, computers, services like holiday tour packages offered by travel agencies and the like.
Merchant: A person or organization advertising and selling products.
Online Marketplace: A web portal where different products from different merchants are advertised for sale, A comparison shopping engine or CSE is an example of an online marketplace. A CSE may have advertisement pages corresponding to each product. For example, an advertisement page for digital camera product on a CSE may have advertisement of digital cameras manufactured

by different companies such as Canon, Sony, Fuji, Nikon and tine like. Usually, when a product is advertised on an online marketplace, link to product page of the product on the merchant website is included in the advertisement, A click to the link leads to the product page of the product where the individual can buy the product. However, there are online market places where sale can be made and there is no requirement to direct the online shopper to the product page. Example of such online market places is w/ww.ebay.com.
Product information: Product information of a product may include product name, product code, product price, discounts {if any), shipping charges (if any), and the like. Product information may be displayed as part of product advertisement on an advertisement page of an online marketplace. The advertisement page may have product advertisement from different merchants.
Product Data Feed: A product data feed is a file comprising at least some portion of product information. The portion of product information to be included in the product data feed may depend on online marketplace requirements. The product data feed has a plurality of fields for providing product information. The product data feeds are sent to online market places such as comparison shopping engines for the publication of product information.
Merchant Website: A merchant website is a portal maintained by a merchant to publish information related to her products and to sell those products online. A merchant website may sell various products offered by the merchant. Each product may have a separate product page.
Product page: is a web page corresponding to a product on a merchant website. On a product page, product information such as features of the product, its different versions, and product specifications, option to purchase the product online and the like, are present. An online shopper may view the product information and place order for the product through the product page.
Advertisement page: is a web page on an online marketplace having advertisement of one or more products from one or more merchants. Products advertised on an advertisement page may be similar in functionality. For example, an advertisement page of a refrigerator may have refrigerators from merchants such as Abt.com, Bestbuy.com and the like. A particular product may be offered by different merchants for different prices.

BACKGROUND ART
The widespread usage of internet for shopping known as 'online shopping' has facilitated the process of buying and selling products from remote locations, countries and continents. An online shopper typically goes to website of a merchant, views the product information and places order for the product. However, this approach may prevent him from accessing same product from other merchants with additional offerings like discounts. To overcome this problem and to make online shopping easy for online shoppers, online marketplaces have come up which provide easy access fo different products and product information, making purchase of a product convenient, transparent and less time consuming. The online marketplaces are web portals where merchants advertise their products with product information. They provide a platform for online shopper to view comparative product information of similar products from different merchants and chose an appropriate product according to his choice.
Merchants need to keep a regular track of the sales of his products taking place through online market places. This to ensure that the money spent on advertising is not significant vis-a-vis the revenue earned.
Once a merchant has published product information of a product on the online marketplace, there are mechanisms to track sales performance of the product. Sales performance may include number of clicks on the link to the product page, conversion of the clicks into actual product purchase and the like. If the sale of the product is less than expected, there is a need for steps to be taken to increase the sale. The steps may be, for example, modifications in the product information displayed to the user. A combination of promotional and pricing strategies may be applied to increase sales.
Prior art teaches about systems for providing feedback/suggestions to merchants to increase sales. This includes US Publication no. US20080222010A1 titled 'System and Method for Enabling Online Research, Publication, Promotion and Management of User Goods' which describes a system for the management of publication and promotion of products. The system receives product data feed from multiple sources and provides a feedback on the performance of each product data feed to merchants. There also exist several channel management tools in market which provide management of product information and provide some feedback on the performance of products.

The feedback provided by systems in prior art though helpful, is not comprehensive. The key reason being that the recommendations are based on limited information including number of clicks on product page, sales performance, feedbacks from online shoppers and the like.
Further, performance of the online marketplaces is a key factor in determining sales performance of a product. Performance of online marketplaces may vary with products. For example, while a first online marketplace may perform well with respect to electronic items whereas a second online marketplace may perform better with respect to books. Hence, choice of a correct marketplace for advertisement of a product is very important for good sales performance of the product.
Therefore, there is a need for a system which analyzes a wide range of information collated from various sources and come up with comprehensive recommendations for improving the online sales performance of a product. These recommendations are likely to be more useful to the merchant and enable him to make prudent business decisions.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a schematic illustrating an environment for the functioning of Online Marketplace Recommender, in accordance with an embodiment of the invention.
FIG. 2 is a schematic illustrating Online Marketplace Recommender, in accordance with an embodiment of the invention.
FIG. 3 is a flow diagram illustrating Che process of analysis of secondary data set and primary data set of a product advertised on online marketplace, in accordance with an embodiment of the invention.
FIG. 4 is a flow chart illustrating the functioning of Traffic Predictor Module, in accordance with an embodiment of the invention.
FIG. 5 is a flow chart illustrating the process of determining popularity of a product.
FIG. 6 is flow chart illustrating the functioning of Data mash-up module, in accordance with an embodiment of the invention.
DESCRIPTION OF EMBODIMENTS
A method and system for providing recommendations for advertisement of a product on an online marketplace is disclosed. A merchant advertises a product

providing its product information on the online marketplace. The sale of the product may happen on the online market place or the online market place may direct an online shopper to the product page on the merchant website for the sale of the product. The performance of an advertisement may be based on traffic generated to the product page of the product and on the total sales of the product. The invention discloses a system and method to track the performance of an advertisement of a product and to provide recommendations to the merchant regarding improving of sales performance of the product.
1
In the following description numerous specific details are set forth to provide a more thorough description of the present invention. However, it will be apparent to a person ordinarily skilled in the art, that the invention may be practiced without these specific details. Various aspects and features of example embodiments of the invention are described in detail hereinafter.
FIG. 1 is a schematic illustrating an environment for the functioning of Online Marketplace Recommender, in accordance with an embodiment of the invention. The environment comprises communication network 102, online marketplace 104, merchant 106 and online marketplace recommender 108. Communication network 102 connects online marketplace 104 and merchant 106. Online marketplace 104 is a web portal for advertisement of products by different merchants. Products may be advertised on advertisements pages on online market places. As an example, advertisement page for laptops on online marketplace 104 may include advertisements of laptops from merchants such as DeW. HP, Lenovo, Apple, Acer etc. along with related product information. Product information for a product, say laptop, from one merchant may be different from the product information for laptop from another merchant. Examples of online marketplace 104 include a CSE or comparison shopping engine, www.ebay.com etc.
According to some embodiments, online marketplace 104 may categorizes products advertised on it in proper taxonomy. Taxonomy may be hierarchical. For example taxonomy for laptops may be "Computers & software products" -> "Laptops". A user may click on the "laptops" link to visit the advertisement page related to laptop.
An additional 'search' feature for searching a product may be present on online marketplace 104. A user search for a particular category of products using search feature. For example, a result page for search on "Laptops" may have link to

the advertisement page having advertisements of laptops from merchants such as Abt.com, bestbuy.com, etc. along with related product information.
Merchant 106 usually possesses a merchant website where detailed information about his business and products supplied by him are available. Each product from merchant 106 may have a separate product page on the merchant website. An advertisement of a product on an online market place may have a link to direct the online shopper to the corresponding product page on the merchant website. For example, an advertisement of a Dell Inspiron laptop may have link to product page of Inspiron laptop on Abt.com website.
Usually the advertisements on the advertisement pages are in form of a vertical list. Rank of an advertisement on the advertisement page is the position which the advertisement occupies from top. The top most advertisement may be given rank 1 (one). Usually, greater the rank of an advertisement, greater is the chances for a user noticing the advertisement. Therefore merchants typically want higher ranks for their advertisements.
Online market places may provide ratings to merchants. The ratings may be based on past records of the merchants. For example, a merchant with a track record of defect free and fast delivery of product may be given a high rating.
Online marketplace recommender 108 is a system for providing recommendations to merchant 106 regarding advertisements of his product on online marketplace 104. Online marketplace recommender 108 is described in detail in conjunction with FIG. 2.
FIG. 2 is a schematic illustrating Online Marketplace Recommender 108, in accordance with an embodiment of the invention. Online marketplace recommender 108 comprises competitive intelligence module 202, traffic predictor module 204, product popularity module 206, merchant input module 208, merchant database 210 , data mash-up module 212 and recommendation dashboard 214.
Competitive intelligence module 202 determines the effect of market competition on products advertised by merchant 106 on online marketplace 104, According to an embodiment, competitive intelligence module 202 extracts competitor data related to a product from online marketplace 104 where the product is advertised. A perl based crawler may be used to extract the competitor data. Competitor data for a product is related to product information of similar products, advertised by competitors, on the online market place. As an example, a merchant X

may have advertised a product "laptop" on an online market place 104. The advertisement will be on laptop advertisement page. The same advertisement page may advertise "laptops" from competitors like merchant Y and merchant Z. The competitor data comprises product information for products offered by competitors Y and Z, available ranking of different competitors on advertisement page, available rating of merchant 106 on the advertisement page and the like.
Competitive intelligence module 202 may also extract the competitor data from a plurality of advertisement pages of online marketplace 104, where each advertisement page comprises similar products from other competitors. For example, for a product A, there may not be a common advertisement page where all competitors and their products are listed. Competitive intelligence module 202 may search through the entire online market place to extract competitor data for product A.
Using the competitor data, competitive intelligence module 202 prepares a competitiveness intensity metric for the product. Details of formulation of the competitiveness intensity metric are given in conjunction with FIG. 3
As stated before, an advertisement of a product on a particular online market place may have a link to direct the online shopper to the corresponding product page on the merchant website. More the number of online shoppers clicking on the link more is the traffic directed from the particular online market place towards the product page. Further, a particular product may be advertised on several online market places. Clicks on all those advertisements may lead to the same product page.
Traffic predictor module 204 predicts traffic directed to product page of the product. Prediction may be made for each online shopping engine for the amount of traffic that will be directed towards the product page. Traffic predictor module 204 may also give ratings to online marketplaces for traffic flow to the product page on the merchant website. The ratings are with respect to a particular product. An online market place contributing relatively more towards the traffic directed to the product page may be given a higher rating than other online market places. An online market place having a higher rating with respect to product A may have a lower rating with respect to product B. The working of traffic predictor module 204 is explained in detail in conjunction with FIG. 4.

Product popularity module 206 estimates the market pull for a product and foreseen market trends. Product popularity module 206 uses three paths to determine the market pull and the market trends. The first path is to determine the expert reviews for the product that are available on the internet. The second path is to use user generated content related to the product on the internet - it could be user reviews or blogs and other opinion based information on the web. The third path is to determine popularity of a product based on the number of searches for the product on search engines such as Google, Yahoo and the like. Search engines keep track of keywords used in different searches. Product popularity module 206 determines point for search volume of keywords specific to a product listed on an online marketplace, to determine the popularity of the product. Search volume defines the number of times keywords specific to a product have been used in searches over different search engines such as Yahoo, Google and the like. Thus more the point for search volume for a product, more are the number of searches done for the product
According to an embodiment of the invention, data from competitive intelligence module 202, traffic predictor module 204 and product popularity module 206 are used to determine competing merchant information and product performance information which constitutes a secondary data set.
Merchant input module 208 provides an interface for receiving a primary data set related to product strategies. The primary data set comprises following parameters:
1. Price Margins: This includes desired price margins and minimum price margins for a product as specified by merchant 106.
2. Deviations in the price margins depending on product inventory status.
3. Competitive positioning of merchant 106: This includes rank range expected by merchant 106 on online marketplace 104 for a product and preferred number of competitors within the product page of the product on online marketplace 104,
4. Competitive benchmark: This depends on input from competitive intelligence module 202, which helps in determining competitors to benchmark pricing against, depending on competitor ranking and similar information. For example, in an advertisement page of online marketplace 104, there may be 10 (ten) competitors of merchant 106, selling same product. However 7 out of the 10 competitors are in different geographical location as compared to merchant 106. These 7 competitors

are likely to have tittle impact on business of merchant 106. Therefore, merchant 106 may decide to benchmark price of his product against remaining 3 competitors (say competitor A, competitor B and competitor C) which are in the same geographical location as merchant 106. Merchant 106 may also specify conditions like she would like price of her product to be in a 5% band with competitor A while in an 8% band with competitor B and competitor C. The above analysis may also be done for competitors other than those present in online market place 104.
According to an embodiment, the primary data set may be entered in merchant input module 208 by merchant 106. According to another embodiment, the primary data set may be entered by an authorized person other than merchant 106, There may be one or more parameters in the primary data set that may have a default value. For example, default competitive benchmark may be within 10% of top ranked competitor in online marketplace 104, Default value for rank of merchant 106 may be in top 50 percentile of merchants. It should be noted that, the identity of a person entering the primary data set in merchant input module 208 is not a limitation of the invention.
According to an embodiment, the primary data set further comprises intenal data of merchant 106. Internal data of merchant 106 is stored in merchant database 210. The internal data comprises:
1. The pricing and promotions available for a product as supplied by merchant 106.
Pricing strategy for a product may depend on internal factors, including but not
limited to,
a. Cost of the product and the overheads assigned to the product
b. Inventory status of the product
2. Past history of use of oniine marketplace 104 by merchant 106. The past history
comprises:
a. Data regarding number of clicks received on online marketplace 104 to
direct towards product page.
b. Data regarding cost of the clicks received to the product page. This
includes the payment by merchant 106 to online marketplace 104 for the clicks
received by the advertisement.
c. Data regarding conversion to sales i.e. how many clicks to the product
page of the product on the merchant website resulted in sales for the product?

The internal data is consolidated to determine the success rate and the ROAS (return on Ad spend) for the product. ROAS is calculated as revenues from sales through the advertisement minus expenditure on advertisement of the product.
According to an embodiment the internal data may be collected from a merchant log file available with merchant 106. The merchant log file has data regarding number of clicks received to product pages on the merchant website, number of clicks converted into sales, present price of a product, promotion offers and the like.
According to another embodiment of the invention, the internal data may be collected by an embedded script on product page of the merchant website.
According to yet another embodiment of the invention, the internal data may be collected by adding unique identifiers in URLs of product pages present in product information submitted to online marketplace 104 for advertisement. The unique identifiers are referred to as URL tags. The embedded script or the URL tags may be provided by online marketplace recommender 108 for tracking traffic directed to product pages of the merchant website.
Data mash-up module 212 takes data from competitive intelligence module 202, traffic predictive module 204, product popularity module 206, merchant database 210 and merchant input module 208, and processes the data to determine recommendations to merchant 106. The recommendations comprises:
5. Price revision required for the product;
6. Retention or removal of product information of a product;
7. Choice of online marketplace 104 for a product;
8. Promotion strategy for a product; and
9. Trends of product popularity in online marketplace 104 for a product.
The functioning of data mash-up module 212 is described in detail in conjunction with FIG. 6.
Recommendation dashboard 214 displays the recommendaiions to merchant 106. According to an embodiment, recommendations for different products of merchant 106 which are advertised on online marketplace 104 are ranked according to priority and displayed from highest priority ranking downwards with different filters and features. Thus a high rank for a product indicates that immediate steps should be taken to improve the sales performance of that product. Table 1 gives an example of ranking of recommendation for different products.

According to an embodiment, the ranking of recommendations for different products is based on three parameters: ROAS, cost of click and final score of secondary data set. The recommendations are arranged in 1st priority of ascending order of ROAS, 2nd prionty of descending order of cost of click and 3'° priority of descending order of final score of secondary data set. Thus in Table 1, product A having higher cost of click is given higher recommendation rank than product B having lower cost of click, although it's final score for secondary data set is high. Final score of secondary data set is explained in FIG. 6

The numerical values used in Table 1 are for the purpose of illustration of ranking of recommendations, and should not be used for limiting the scope of the invention.
The different filters and features available on recommendation dashboard 214 may be filter for displaying recommendations specific to marketplace, filter for changing order of recommendations, feature for accepting or ignoring a recommendation, feature for displaying recommendations with reason and the like.
For the purpose of explanation of FIG. 3 to FIG. 6, it is assumed that merchant 106 has advertised a product ABC on online marketplace 104.

FIG. 3 is a flow diagram illustrating the process of analysis of the secondary data set and the primary data set of a product advertised on online marketplace 104, in accordance with an embodiment of the invention. The analysis comprises of four parallel steps.
At step 302, competitiveness intensity metric for product ABC is prepared by competitive intelligence module 202. The competitiveness intensity metric comprises three factors: competitive density, price flexibility index and rank flexibility index.
Competitive density determines the number of competitors present in online marketplace 104 selling product similar to that advertised by merchant 106. According to an embodiment, competitive density may range from 1-5. A rate of '1' for a product indicates very high competition i.e. the number of competitors sailing the product offered by merchant 106 is very high in online marketplace 104 and hence the competition is stiff. A rate of '5' for a product indicates monopoly status i.e. merchant 106 is almost the sole supplier of the product and there is almost zero or no competition in online marketplace 104.
Price flexibility index Indicates the tolerance available for price changes of a product given the desired and minimum price margins by merchant 106. According to an embodiment, price flexibility index varies from 0-5. A rating of '0' indicates that price of the product is resulting in a price margin which is close to or even less than the minimum price margins. A rating of '5' indicates that the price of the product is resulting in price margin which is above the desired price margin and advantageously positioned against competitive benchmarks.
Rank flexibility index indicates flexibility of a product towards any change in its rank on an online marketplace given competitive benchmark thresholds of the product. Price of a product is an important factor while deciding rank flexibility index. An example scenario is: a product of merchant 106 has a price of $300 on online marketplace 104. Minimum price for the product as specified by merchant 106 is $200. The product is presently ranked at 10th position on an advertisement page on online marketplace 104. However, merchant 104 desires a position among (op 5 ranked products on the advertisement page. But to achieve the desired position, the price of the product needs to be reduced to $260. Hence the rank flexibility index of the product will be high (4-5), specifying that the rank of the product is flexible to changes and can be increased without incurring much loss in sale of the product. However if there are some constrains specified by merchant 106 and the price

cannot be reduced to $260, then rank flexibility index will be 0, specifying that rank of the product is not flexible to changes and rank cannot be increased without incurring significant loss in sale of the product. Further rank flexibility index of the product will again be '0', if the price needs to be reduced below $200 which is minimum price of the product.
Competitive intelligence module 202 extracts data related to number of competitors, their rankings, prices offered by competitors and other data from product page of product ABC on online marketplace 104. Based on the data, competitive intelligence module 202 gives ratings to product ABC on the three factors: competitive density, price flexibility index and rank flexibility index. The competing merchant information of a product comprises of these three factors.
At step 304, traffic directed to the merchant website of merchant 106 is predicted. The step is discussed in detail in conjunction with FIG, 4.
At step 306, product popularity module 206 estimates the popularity of a product based on user generated content available on internet. Product popularity module 206 tracks a list of websites providing expert reviews and user comments on products advertised on online marketplace 104. Product popularity module 206 further browses for search volume of searches performed for a particular product on search engines such as Google, Yahoo and the like to determine count of searches done for the product. For example, determining search volume may include determining the number of times "powershot A590" keyword has been used for a search on the search engines for a particular period of time. Websites providing expert reviews are periodically crawled and data related to new product reviews and product ratings including positive and negative remarks are captured. In case, expert reviews are not available, then search engine results are checked for new reviews that are credible. For capturing user comments and reviews, different product review websites and blogs are searched for the product related keywords. The result is filtered for credibility and recentness. In case, the volume of result is huge, then further filtering is done on basis of positive and negative correlation keywords. The shortlisted result is evaluated for positive and negative opinion and a weighted rating is prepared taking into account credibility of the websites, quality of the user comments and reviews and the recentness of the user comments and reviews to determine quality score for rating. A weighted rating is prepared for the data captured through the websites, taking into account credibility of the websites, quality

of the expert and user review and the recency of the expert review to determine quality score for rating. Details of selection of websites for expert and user reviews and the process of extracting the reviews are explained in conjunction with FIG. 5
At step 308, the primary data set is analyzed. The primary data set comprises price margins for the product, competitive positioning of the merchant, competitive benchmark and internal data of merchant 106. The analysis of primary data set comprises of ROAS analysis. ROAS analysis comprises determining the ROAS for a product advertised on online marketplace 104. Depending on the ROAS, the product is assigned an ROAS rating. For example, for a product, expenditure on advertisement on online marketplace 104 may be $100 and desired return may be $200. Depending on the returns generated by the product, the following ROAS ratings may be given:

The numerical values used in Table 2 are for the purpose of illustration of logic of ROAS rating assignment, and should not be used for limiting the scope of the invention.
Product ABC is assigned an ROAS rating at step 308.
FIG. 4 is a flow chart illustrating the functioning of traffic predictor module 204, in accordance with an embodiment of the invention.
At step 402, traffic predictor module 204 creates a search query for product ABC. The formulation of the search query comprises identifying keywords related to the product. For example, for a digital camera Canon Coolpix S60, the keywords may be 'digital camera', 'Canon', Coolpix' 'S 60' '10 Megapixel' and the like. According to an embodiment of the invention, traffic predictor module 204 further uses search engine keyword analysis tool (for example: Google's keyword tracker) if available, to get a list of related keywords used in the past searches.

At step 404, the search query comprising the keywords for product ABC is entered in the search engine to get search results.
At step 406, search results for the search query are obtained. The search result may include link to advertisement pages of online marketplace 104 and other online marketplaces where product ABC is advertised. Search results may also include link to product page of product ABC on the merchant website.
According to an embodiment of the invention, first 100 search results are selected and tracked to estimate ranks of different online marketplaces and their landing pages. The number of search results selected may vary and is not a limitation of the invention. For example, a search for Canon Coolpix S60 digital camera may list an advertisement page of online marketplace A as 1^' search result while an advertisement page of online marketplace B as 5"' search result and an advertisement page of online marketplace C as 10"" search result. Typically, the search results are in decreasing order of relevance. The search result ranks of the online marketplaces are taken into account for traffic prediction.
Further an online marketplace 104 may be ranked differently for different products. For example, for a Canon digital camera, online marketplace A may receive topmost ranking in the search results while for a Samsung television online marketplace A may receive 15'*^ rank in the search results.
Based on the ranking of an online marketplace in the search results for a product, the online marketplace may be given a rating from 1-5 for a product, according to an embodiment. For the preceding example, traffic predictor module 204 may rate online marketplace A with a high rating of '5' for excellent traffic directed to product page of Canon digital camera. But for Samsung television, traffic predictor module 204 may rate online marketplace A with a low rating of '2' for \ovj traffic directed to the product page.
Based on logic mentioned in step 406, online marketplace 104 is rated for traffic directed to product page of product ABC in step 408.
FIG. 5 is a flow chart illustrating the process of determining popularity of a product.
At step 502, product popularity module 206 generates search related keywords for a product. Examples of search related keywords are, "Digital camera reviews", "Best Washing Machine", and the like. Search related keywords are chosen such that the search results are related to reviews regarding the product.

At Step 504, the search related keywords are entered in different search engines such as Google search engine, Yahoo search engine and the like, and search results are retrieved. According to an embodiment, first 50 search results may be considered for reviews. However the number of search results considered for review may vary and is not a limitation of the invention.
At step 506, the search results are filtered. Filtering of search results comprises of following steps:
1) Filtering out websites from the search results where no review is available for the product.
2) Filtering out websites and blogs from the search results where reviews are copied from other websites. According to an embodiment, blogs with less than 30 user reviews may be filtered out from the search results.
3) Confirming the credibility and existence of a company hosting a website present in the search results, by performing telephonic checks.
4) Checking traffic intensity to a website in the search results by obtaining traffic data from sources such as Alexa traffic monitor.
The websites obtained after filtering of search results, are checked for frequency of occurrence in the search results. For example, website A might have occurred 10 times in the search results, website B might have occurred 5 times and the like. The websites are then organized in order of frequency of their occurrence. The websites with higher frequency of occurrence are selected for monitoring reviews. For example, out of a list of 20 websites, 8 websites have higher frequency of occurrence as compared with the rest 12. Hence these 8 will be selected for monitoring reviews.
At step 508, reviews are extracted from the websites selected in step 506, According to an embodiment, a perl based crawler may be used to extract reviews about a product from the selected websites. The crawler uses keyword correlation technique to pick correct product and the related reviews from a selected website. Positive keyword correlation is used as selection method where the crawler selects all product pages with positive keyword and rejects the product pages having negative keywords. For example, if the crawler is searching review for "Canon SD 1000 Powershot Digital Camera", positive keywords may be "SD 1000", "Canon", "powershot", "digital Camera" and the negative keywords may be "Accessory", "For

SD1000", "with" and the like. Further, if the crawler encounters "Lens for Canon SD1000", negative correlation Is triggered and the crawler rejects the page.
At step 510, search volume for the product is determined. For determining search volume, keywords related to the product are identified. The process of identifying keywords is same as in the case of traffic predictor module 204. Product popularity module 206 then accesses databases of search engines to find the number of times an identified keyword has been used in search over a time period. According to an embodiment the time pehod is last 30 days. The time period is variable and depends on analysis. It should not be used as limitation of the invention.
According to an embodiment, product popularity module 206 gives 0,0001 points every time a keyword has been searched for in the search engines. For example, if a keyword has been searched for 50 times in the search engines, then total points given will be 50 * 0.0001 = 0.0050, Adding points for all identified keywords, point for search volume is decided for the product. The point for search volume is capped at 1, Hence if point for search volume exceeds 1, it will still be taken as 1.
At step 512, the product is given a product popularity index based on the reviews and point for search volume. According to an embodiment expert and user reviews are given equal weights for determining product popularity index. The expert and user reviews are converted to a scale of 0-1, For example, if a product has an expert review of 6 out of 10, then it is scaled to 0,6, Average of expert review and user reviews is taken and added with point for search volume to give a product popularity rating to the product. Table 3 gives an example of methodology adopted for assigning product popularity index to product ABC based on expert and user



In table 3, final expert review score (A) of 0.5 has been calculated by taking average of expert reviews from Review website 1 to Review website 5. Final expert review score (A) = (0.4+0.4+0.6+0.5+0.6)/5 = 0.5
(n Table 3, final user review score (B) of 0.6 has been calculated by taking average of user reviews from Review website 1 to Review website 7. Final expert review score (A) = {0.6+0.8+0.7+0.6+0.6+0.3+0.4)/7 = 0.57 = 0.6 (rounded off)
The point for search volume (C) has been calculated as 0.8 for the product.
Therefore, product popularity index of product ABC is [(A+B)/2] +C = 0.55+0.8 =1.35. The numerical values used in Table 3 are for the mere purpose of illustration of calculation of product popularity index of a product, and should not be used for limiting the scope of the invention.
According to an embodiment, if a product has universal negative reviews, then its product popularity index is '0'.
According to an embodiment, product popularity index of a product and rating of online marketplace 104 for the product comprises the product performance information. Product popularity index defines popularity and rating of online

marketplace 104 determines level of traffic directed to the product. Hence the two factors determine how well or bad the product is performing in market.
FIG. 6 is a flow chart illustrating the functioning of Data mash-up module 212, in accordance with an embodiment of the Invention.
At step 602, data mash up module 212 collects the secondary data set from competitive intelligence module 202, traffic predictor module 204 and product popularity module 206, and the primary data set from merchant input module 208 and merchant database 210. The secondary data set comprises the competing merchant information and the product performance information which are determined from:
10. The competitive intensity metric of product ABC having the competitive intensity, price flexibility index and rank flexibility index for product ABC.
11. Rating of online marketplace 104 for traffic directed to product page of product ABC on the merchant website of merchant 106.
12. Product popularity index for product ABC.
The primary data set comprises data received from merchant input module 208 and merchant database 210.
At step 604, data mash-up module 212 processes the primary and secondary data set. The primary data set is processed to give product ABC an ROAS rating. If product ABC is producing good returns and its sales are good, then its ROAS rating for online marketplace 104 is high. In case, the returns and sales of product ABC are not good, then its ROAS rating will be low.
For the secondary data set, the different ratings and rankings given to product ABC on the factors (price flexibility index, product popularity index, ranking of online marketplace 104 for product ABC and the like) are multiplied to get a final score of secondary data set for product ABC. For example, product ABC may receive following ratings and rankings:
Competitive intensity: 3/5
Price flexibility index: 4/5
Rank flexibility index: 4/5
Ranking of online marketplace 104 for traffic directed to product ABC: 3/5
Product popularity index: 1.35/2 So, the final score for product ABC with respect to secondary data will be: 3*4*4*3*1.35 = 194.4

If one or more of price flexibility index, rank flexibility index or product popularity index is '0', then the final score will be '0' for product ABC. A final score of '0' suggests that product ABC is either not meeting its price margins or has universal negative reviews.
At step 606, data mash-up module 212 prepares recommendations for product ABC. For preparing recommendations of product ABC, the following logic is followed:
1. If product ABC has already been on online marketplace 104, it is classified between A (unsatisfactory) and E (outstanding) on its ROAS rating, based on past performance that comes from the primary data set.
2. If the ROAS rating of product ABC is 'A' and the final score of secondary data set is '0', then product ABC may be recommended for removal from online marketplace 104.
3. If product ABC has received a ROAS rating of B' or 'C but its final score of secondary data set is '0', then the parameters in the secondary data set which are causing the final score of secondary data set to be '0', are recommended for change to bring the final score of secondary data set to a positive value. For example, if price flexibility index rating of product ABC is '0', then data mash-up module 212 may recommend changing the price margins of product ABC. If the price of product ABC is very high as compared to price offered by competitors, then a recommendation may be made to reduce price of product ABC. If ROAS is "B" and the product popularity index for product ABC is '0' implying universal negative reviews and zero searches made by consumers, then merchant 106 may be recommended to remove product ABC from the market place. If ROAS is "C" and the product popularity index for product ABC is '0' implying universal negative reviews and zero searches made by consumers, then merchant 106 may be recommended to remove product ABC from the market place based on the recentness of last sale. if the last sate has happened within last month, then merchant 106 will be recommended to continue product ABC with lower cost per click bid, else merchant 106 will be recommended to remove product ABC from online marketplace 104,
4. If the final score of secondary data set for product ABC is positive, then for ail ROAS ratings, the price of product ABC is recommended to change incrementally to arrive at optimal competitive intensity index and therefore to achieve better final score of secondary data set for product ABC. The incremental price

change which provides the highest incremental increase to the final score of secondary data set is recommended for changes.
Further analysis may be done with respect to the online market places where the advertisement of the product is not present. Two factors are considered for the analysis: competency level of a product of merchant 106 and competency level of an online marketplace. Competency level of a product determines whether the product is capable of competing with similar products from competitors present on an online marketplace where the product is not advertised. For example, product A of merchant 106 has price of $100. However on an online marketplace Y, similar products from competitors may have an average price of $80. Therefore product A does not meet competency level and hence will not be recommended for advertising on online marketplace Y. However if the average price of similar products from competitors is $ 110. then product A may be considered for advertising on online marketplace Y. A further check on competency level of online marketplace Y may be required before recommending the advertising of product A on online market place Y.
Competency level of an online marketplace with respect to a product is decided by the traffic generated to the product advertised on the online marketplace. A good traffic implies higher chances of sale of the product while poor traffic reduces chances of sale of the product on the online marketplace.
Thus for product ABC and an online marketplace X where product ABC is not advertised, if competency level of product ABC and competency level of online marketplace X with respect to product ABC is met, then product ABC will be recommended for advertisement on online marketplace X.
Thereafter at step 608, the recommendation prepared by data mash-up module 212 are displayed to merchant 106 through recommendation dashboard 214. According to an embodiment, merchant 106 has a login for the recommendation dashboard 214 through which merchant 106 may view recommendations specific to his product advertised on online marketplace 104.
While example embodiments of the invention have been illustrated and described, it will be clear that the invention is not limited to these embodiments only. Numerous modifications, changes, variations, substitutions and equivalents will be apparent to those skilled in the art without departing from the spirit and scope of the invention as described in the claims.

WE CLAIM:
1. A method of providing recommendations related to a product information
published on an online marketplace to a merchant, the method comprising the
steps of:
a) Analyzing the product information for a secondary data set along with a primary data set wherein the secondary data set comprises competing merchant information and product performance information; and
b) Making recommendations on the product information to the merchant based on the analysis.

2. The method of claim 1 wherein the online marketplace is a comparison shopping engine.
3. The method of claim 1, wherein the primary data set comprises price margins for the product, competitive positioning of the merchant, competitive benchmark and internal data of the merchant, the internal data comprising present product price and promotions, number of clicks, cost of clicks, conversion to sales and inventory status of the product.
4. The method of claim 1, wherein the competing merchant information and the product performance information is determined using competitiveness intensity metric, product popularity and traffic prediction for the product.
5. The method of claim 4 wherein the competitiveness intensity metric for the product is prepared using data of competing merchants' prices and ratings within the online market place.
6. The method of claim 4 wherein the product popularity is determined based on
a. analyzing user generated content available on the internet; and
b, determining number of searches performed for the product on a search
engine.
7. The method of claim 4 wherein the traffic prediction for the product determines traffic directed to a merchant website of the merchant from the online marketplace.
8. A method of providing recommendations related to product information published on an online market place to a merchant, the method comprising the steps of:
a) Analyzing the product information for a secondary data set along with a primary data set, the product information being associated with a product, wherein the analysis comprises the steps of-.

i. Preparing a competitiveness intensity metric for the product, wherein
the competitiveness intensity metric is prepared with respect to similar products from other merchants;
ii. Determining product popularity based on:
a. analyzing user generated content available on the internet; and
b. determining number of searches performed for the product on a search
engine;
iii. Analyzing the primary data set associated with the product information;
b) Making recommendations on the product information to the merchant based on the analysis.
9. The method of claim 8, further comprises predicting traffic directed to a merchant website of the merchant from the online marketplace, wherein the traffic is predicted based on ranking of the online marketplace on a search engine result pages.
10. The method of claim 8, wherein the primary data set comprises price margins for the product, competitive positioning of the merchant, competitive benchmark and internal data of the merchant, the internal data comprising present product price and promotions, number of clicks, cost of clicks, conversion to sales and inventory status of the product.
11. The method of claim 8, wherein the recommendations comprise of:

a) Price revision of the product;
b) retention or removal of the product information;
c) choice of the Online market place for the product; and
d) Promotion strategy for the product;
12. A system for providing recommendations related to product information published
on an online market place to a merchant, the product information being
associated with a product supplied by the merchant and a website associated
with the merchant, the system comprising:
a) a competitive intelligence module, the competitive intelligence module configured to determine the effect of market competition on the product;
b) a traffic predictor module, the traffic predictor module configured to predict traffic directed to the website from the online market place, the prediction being based on ranking of the online market place on a search engine result page;

c) a product popularity module, the product popularity module configured to determine popularity of the product based on
a. analysis of user generated content available on the internet; and
b. number of searches performed for the product on a search engine;
c)a merchant input module, the merchant input module configured to
receive inputs regarding the product;
d)a merchant database, the merchant database comprising internal data of the merchant; and
e)a data mash-up module, the data mash-up module configured to;
i, collect a primary data set and a secondary data set. wherein the primary data set comprises data from the merchant input module and the internal data of the merchant, and the secondary data set comprises data from the competitive intelligence module, the traffic predictor module and the product popularity module; and
ii. process the secondary data set and the primary data set to make a list of recommendations for the product;
13.The system of claim 12 further comprises a recommendation dashboard, the recommendation dashboard configured to display the recommendations.
14. The system of claim 12, wherein the internal data comprising present product price and promotions, number of clicks, cost of clicks, conversion to sales and inventory status of the product.