Our results suggest that BD-enabled CRM initiatives could require several . responding to market opportunities or threats'': assets, both tangible and .. " Customer Relationship Management" in Title, Abstract, and . Used with other meanings, for instance needs as a verb and not International Journal of Hospitality. ABSTRACT. With the increasing competition that businesses especially in hotels industry face today in order to survive, firms hotels, but also embrace all CRM practices that make the hotels more competitive to ever changing customer's. described using tangible nouns such as object or device. services are an abstract concept for consumers, meaning that they are highly subjective. It would, for example, include the atmospherics, our relationships with staff and other customers, during the early days of hospitality management literature much store was.
Built as an evolution of previous projects , the novelty of this approach is the inclusion of an enhanced recommender system built upon artificial immune systems, the inclusion of geo-representation enabled by PostGIS and the enhancement of opinion mining by means of feature-based opinion mining through ontologies  and opinion summarization . Figure 1 depicts a three-layer scheme that represents POST-VIA architecture along with its main interactions and information flows.
The architecture follows best practices in distributed systems design in order to assure dependability  and proper service levels . The three main actors of the system are: The first actor deals with the system and communicates specific recommendations and tips to tourists along with overall supervision of the system and communication with POIs. Some of these points are under direct supervision of DMOs, like museums, for instance, while others, like restaurants are normally lead by POI owners or administrators.
The second actor is the POI administrative, which is responsible of communicating news, offers, availabilities and general information about the POI. The last actor is the tourist that interacts with the system during pre-visit, during-visit and post-visit. In a nutshell, he or she browses information, receives recommendations, includes comments to the system or simply visits POIs.
International Review of Management and Marketing
All this information is later used in the development of specific recommendations. The architecture and components of the system are structured into three layers which provides the possibility to apply different replication settings at the operating system  or the web service  levels.
The purpose and functionality of the layers is as follows: This layer provides a web interface and a configuration interface, given that is the component aimed to interact with tourists but also POIs administrators and DMOs.
Two kinds of interfaces are available for the system. Firstly, traditional web interfaces. These interfaces are coded using HTML5 in order to easily adapt to all kinds of devices including mobile.
CUSTOMER RELATIONSHIP MANAGEMENT
Secondly, a native mobile interface is also available for tourists. Regarding functional subsystems within the Web interface, there is a subsystem aimed at tourists including aspects like travel album, travel experience and social network. A second subsystem provides POIs management that can be used to maintain basic POIs information but also to access POIs metrics and other elaborated scorecard-like metrics.
Finally, there is a subsystem aimed to show and maintain DMO information including managed POIs, specific DMO information, scorecard information about the use and rating of its management services, but also an interface to classic CRM functions. This business logic layer consists of four components as depicted in Figure 1. These components are explained in what follows: This component enables the valuation of the travel experience by the users before, during and after the travel. It also includes the possibility of creating a multimedia album associated to the trip including relevant information as tickets, brochures, weather conditions, geolocation-enabled itineraries but also opinions, check-ins and other personal information.
This component includes all traditional operational CRM features like, for instance, sales, marketing and service management. Traditional CRM approaches have been enriched by the use of ontologies.
A view of the main features of the component is explained in detail in . The use of semantic infrastructures allows querying the information using SPARQL queries via a specific interface designed for the system that receives the information required and is in charge of generating the appropriate SPARQL query to access the semantic model.
Regarding the storage authors adopted Jena API as main framework to load the ontology model and store the data. Artificial Immune Recommender system. Artificial immune systems is a field of study inspired by the human immune system that emerged in the s as a new computational research area inspired by theoretical immunology and observed immune functions, principles and models .
The use of these kind of systems have been reported by literature  with promising results compared to other equivalent solutions. In this case, given that the system will be used in a mobile setup with spatial features, it is needed to add spatial information to antigens and antibodies to be able to perform spatial queries that requires the use of areas or distances.
Regarding the model, POIs and visits include spatial information in order to be able to include geographic data in the recommender engine. The general design of the algorithm is adapted from the works of Cabanas-Abascal et al. However, in order to meet the specific characteristics present in POST-VIA, specific adjustments were made in the algorithm parameters.
These parameters are depicted in Table 1. Algorithmics are, as already reported, similar and their parameters are adjusted from the ones adopted by Cabanas-Abascal et al. However, the scope is broader in our case. POIs are not just hotels, like in the previous effort; they include hotels, bars, museums, sights This makes the algorithm different in the information processed and the kind of recommendation provided.
With regard to the recommendation algorithm for a POI, formula is also adapted from the work of Cabanas-Abascal et al. Set of antibodies that take part in the recommendation.
Artificial immune system parameter configuration d. Opinion mining engine is based on the use of ontologies. Given that, apart from star-based ratings, users provide opinions on POIs that can be connected or not with their ratings that, in many times are absentit is needed to design and deploy an engine to extract information from this data.
Natural language processing NLP module. This module is aimed to obtain the morphological and syntactical structure of sentences. Semantic-based feature identification module.
The module is aimed to perform a double task. Firstly, and using a domain ontology, features included in the opinions later transformed in the corpus are extracted from it.
Secondly, a score that represents the importance of a given feature is calculated. This latter task is achieved by grouping features in accordance with their semantic distance and then attached to a main concept of the ontology. In order to identify positive, negative and neutral values of verbs, nouns and adjectives, SentiWordNet 3. Final polarity is calculated by means of an intelligent average of the positive and negative senses of the set of words in the SentiWordNet database.
This module is designed to provide effective feature sentiment classification. Features are represented by means of their Euclidean vectors and polarity is calculated taking into account positive, negative and neutral directions coded in the vector. This opinion mining engine is used to enrich POI ratings and feed the recommendation mechanism with their enriched features. This layer stores knowledge of tourists, POIs and DMOs, with the relations among them in terms of visits, ratings and comments.
This layer also stores relations about the different concepts which make up the platform. Some of the features are also coded using spatial databases PostGIS and some others are simply coded in a relational database. Validation In this section, authors describe the validation of the tool.
Design Once POST-VIA had been developed and tested, the second step was to test the validity of the tool in a real scenario to measure the accuracy of its recommendations. Given that the overall aim is to test the accuracy of the recommender system, a traditional approach to test these systems was adopted. The evaluation was conducted in two steps. In the first step, the system is used by a set of users during a trip to a city different from the one they live in.
A set of data was also collected from users regarding their preferences by means of a questionnaire. The aim of this first step was to provide a set of test cases in order to execute the second step.
To do so, a set of users were asked to fill in the questionnaires and use the system for a weekend in a short weekend trip to Salamanca, Spain Asking the system for at least seven recommendations for each user regarding restaurants, cafes and bars.
In the second step, and, by means of a GPS manipulation program, a set of recommendations for restaurants and bars will be performed in the area of Madrid city centre Spain. At the same time, a set of experts also provided a ranked set of recommendation for these users in order to compare their results with the ones provided by POST-VIA Thus, two kinds of recommendations are produced in the validation phase. Each of them recommends three different POIs for every situation given. A situation entails a geographical point in the map where the user is supposed to be situated, a given time and a track of previous actions by this user during this day minimum three.
Three different situations are provided by researchers as a testbed. POIs were imported including bars and restaurants in Madrid city centre and are aimed to be within walking distance. To wrap-up, every situation and user will have six different recommendations, three provided by the experts and remaining three that are provided by the system. These measures must be complemented in order to get a full view of the recommender system from an evaluation perspective.
Thus, authors decided also to perform a coverage test. The coverage of a system is a measure of the domain of items in the system over which the system can form predictions or make recommendations i. Sample The sample is divided into two different subsets: In the first set, 14 individuals took part in the experiment 8 men and 6 women. The average age of users was Subjects were selected from those who answered positively to a personal invitation sent by the authors to personal contacts aimed to take a trip to Salamanca and to participate in the study.
This first set of participants performed the task in a time frame of 3 months. Secondly, to provide recommendations, a group of four experts was invited to participate in the study.
They all serve as professional tourism guides in the city of Madrid. All of them were invited to participate by researchers who contacted them using their personal networks. Results and discussion Once the information from users was gathered by the system and the questionnaire was sent to experts, along with a briefing on the actions performed by the users, experts were gathered to submit their recommendations.
This classification was carried out by the four experts using the Delphi method. Each expert issued a pair of recommendations for every user and situation.
Finally, agreement between experts was achieved by applying the Delphi method, producing an output of three recommendations per situation and user. This is a moderate agreement, however, due to the high number of POIs and the important number of recommendations, this could be acceptable.
This set is composed by three different recommendations per situation three different situations and user twelve usersreaching the number of recommendations made by the system and an equal number performed by experts.
For each situation, data for the 12 experts is presented. In sum, 36 situations have been tested. In general, there is a remarkable similarity in the recommendations comparing experts and POST-VIA for all situation and expert pairs.
However, to evaluate the quality of recommendations from the two sources in a scientific way, precision, recall and F1 were used as measures. These measures are widely used in the literature to evaluate the quality of recommender systems.
The recall-metric indicates the effectiveness of a method for locating interesting topics. The precision-metric represents the extent to which the items recommended by the system are interesting to users compared to the ones performed by experts. F1 is calculated as the sum of the weights of precision and recall. In the first analysis performed, the recommendations by the experts were considered correct. So, if the product recommended by POST-VIA coincides with that recommended by the experts the first recommendation on every set then it was considered a correct recommendation.
Firstly, precision score determines the number of correct recommendations offered by the system with respect to the overall recommendations offered by POST-VIA Recall score represents the relation between the number of correct recommendations with respect to the overall correct products in this first analysis the products selected by the expert.
In the case studied here, precision, and recall and, as a consequence, F1 values are exactly the same because the number of correct recommendations is equal to the number of recommendations offered by the system. In this first analysis, these values reach 0. The reason behind this low accuracy could root on the big amount of POIs available for a recommendation and the relative similarity among them.
The first argument that supports this low accuracy is the high number of POIs available for recommendation. The sample included a set of POIs. This is a considerable amount of data to choose from without previous information on these specific POIs and their interaction with users. Given that POST-VIA is aimed to work in a setting in which recommendations will be made on the basis of previous visits to a specific place, authors are confident in increasing the accuracy in these situations.
However, authors believe that the most important argument supporting the modest accuracy is the composition of the sample. In a nutshell, it is important to note that the differences between POIs in the same category, price and area are just a matter of taste for experts and tourists. Given that Madrid is a city with a huge offer in the culinary arena and that, for instance in the traditional restaurant category, there are many similar restaurants with similar prices and placed in the same area, it is logical to have recommendations in all of them in similar situations.
The Speed Dating Session with our 38 sponsors will be held on Thursday from If you have not completed or printed your Speed Dating Dance Card, please let a staff member know so we may process one for you. Please refer to Page 10 for a map of tables. Evaluations Staff Staff members wear identifying badges and are here to help, so please feel free to call on them for questions or assistance.
Please contact us at What does that mean, exactly? We help companies understand their best customers, and how to keep them coming back. Because when you know and treat customers better than your competitors, you get a leg up. In fact, lots of good things happen when you put customer science at the heart of your strategy: These new models allow predictions of human behavior in domains ranging from financial decisions to health behaviors to consumer consumption.
In practical application these new models give us insight into how to best use social networks to more efficiently and effectively incentivize customer behavior. Audience members will learn about these new models, and how they can be used to improve both business and personal lives. Connect with consumers across all channels on their journey from awareness to long-term loyalty.
Customers are looking for convenience, value, and style… all while online shopping continues to grow. To be heard above the clutter of marketing messages and increasing competition, retailers need to assign a purpose to their brand. The future of brick and mortar lies in the rethinking of many core beliefs. Attend this session to hear from Jeffrey, co-founder, to hear how a collaboration between four close, Jack Kerouac-reading friends from Wharton inspired two concepts that are taking the retail industry by storm.
Glean tips and steps you can take today to drive successful customer engagement with your brand. Aimia leverages your brand, strategy and technology to create trusted relationships and retail connections.
Day 1 continued Shedd Aquarium 7: Guests will be given private access to the marine mammals, corals, fish and other sea creatures. Contact us today at or visit us at brierley.
Sponsored by Grand Ballroom - Second Floor 8: In this provocative keynote, Joseph will outline how retention can become the new acquisition for retailers today and in doing so, literally transform the way they go to market and establish a critical competitive edge and advantage.
Sponsored by Grand Ballroom — Second Floor Do it Best member-owners, most of whom operate hardware stores, home centers, or lumber yards, are able to pick and choose from a menu of services offered by Do it Best Corp.
The Do it Best footprint includes 3, stores in all 50 states and 53 countries. He will share experiences of doing so in a world in which each store can choose whether to participate or not, has its own unique logo and brand, and can vary the reward threshold and level based on its particular needs.
Fast forward to and Hallmark has evolved and recently launched a brand-level all-channel loyalty program, Hallmark Card Rewards. This branding, and the new store formats to go along with it, had set the stage for a successful Research indicated that one of the key barriers to driving repeat usage for Vitamin World was convenience. A key omni channel objective was to drive repeat online business, enabling customers to get the same product they had found in store online to replenish their supply.
The enhanced website served to provide a seamless online experience, while the CLM program tested four new approaches. Together, these initiatives have driven true synergy between In-store and online. Repeat purchase rates have increased dramatically, and responders are twice as likely to shop online. Sponsored by Continental Ballroom - First Floor How can we help you? Attend this session to meet up to eight CRMC Sponsors in less than two hours, and find out how they can help you achieve your goals.
If you did not sign up for Speed Dating during the registration process, or need help with your Dance Card, please stop by the Registration Desk. We hope you take part in what promises to be the best networking opportunity of the conference! Day 2 continued Grand Ballroom — Second Floor 2: In a market dominated by big box retailers, how does Orchard use their strong historical and local roots to build presence in the community and create customer loyalty among the most desirable and profitable customer segments?
Roxanne will explain how any retailer can localize their customer loyalty through community outreach and old fashioned service. Sponsored by who has what in their shopping carts, and what you should do about it.
Following an overview of past programs, they will take you through the process step by step. Learn how analysis of past programs led to the decision to make changes supporting customer centricity. Hear from Gianluca and Nicola why Luxottica decided to make their insourcing decision, and how they insourced their data management and analytics in record time. In addition, hear why Luxottica is building an expanded customer engagement hub with new channels; why and how they are adding new data sources; and what new programs will be developed in the future.
Attend this session to hear about the complex and intense journey of the Luxottica team, and learn some of their keys to success, as well as pitfalls to avoid. Return on Relationship In a connection economy, relationships are the new currency. If you want to continue to reach your target customers in this hyper-competitive age of social media, your marketing needs to focus on building relationships. Of course, with a value proposition like this, who needs marketing?
Retail, especially in the NYC area, is brutally competitive, especially in the off-price world. Valentino Vettori will share the story of how Century 21 is transforming into a data-driven, brand inspired and customer relevant marketing organization that is increasingly an asset to the merchants running the retail side of the business. The investments C21 is making in customer marketing — and the data environment to support it -- are part of a compelling story that illustrates the importance of committing to integrated marketing, a strong brand, a renewed customer focus and the requisite supporting resources.
Valentino will share these elements that include not only an enhanced loyalty program and an actionable customer segmentation strategy but also a complete brand repositioning, a branded mobile experience and the internal resources that pull it all together. Sponsored by Grand Ballroom - Second Floor They want to have a personal connection with your brand wherever they are - whether they buy online and pick up in the store or they research in-store and then buy online. They want to receive offers directly from your website and your brand messaging on their smartphone and via email.
Steve will highlight how OfficeMax has been able to leverage the power of personalized communications to bring some clarity into these blurred channels with the following strategies: Join us for a six-person panel representing six segments of shoppers across the loyalty spectrum. You will learn which brands are inspiring true brand loyalty, and how they are doing it; which customers are most likely to be loyal: Attitudes and Drivers With the global economy recovering from a fragile state, there has been a need for brands to not only acquire new customers, but also convert their everyday customers into true brand loyalists.
Discovering the secret to turn a customer into a true brand loyalist is the key to a successful loyalty program. To do so, you need more than attractive rewards, coupons and discounts.
But what makes a consumer loyal to your brand? What are the driving forces behind their purchasing behaviors? Epsilon partnered with Wylei Research to uncover the driving forces behind consumer loyalty in a Customer Loyalty Study conducted at the end of the fourth quarter in As a retail marketer, you know you need to be more personally relevant in your direct mail and email programs to meet changed consumer needs and improve your response rates and sales impact.
You are not alone! And we have the proof. Last year, we looked at preference centers, and the year before that, mobile readiness. This year it is relevant content… And you just might be on our list! In this session, we will share case study examples and practical guidance for how to embrace and apply true relevance and personalization to your existing programs, without breaking the bank or killing your team.
And, new for CRMC The Relevance IQ audit of 50 retailers and consumer brands in which we have assessed the degree to which these marketers successfully or not apply the concepts of relevance and personalization. Are you on the Relevance IQ list?
How did you score? By re-engaging a dormant customer base through personalized, multi-channel programs they turned dormant consumers into active shoppers. You will find that they have changed substantially from just a few years ago.
Attend this workshop to learn how to apply these factors to your own business. Follow us on Twitter: Key aspects include integration of mobile and social media into websites, enabling online commerce and building the marketing, e-mail, operational and customer service functions necessary to support omni-channel sales.
He has been responsible for enhancing omni-channel business models by helping integrate functionality with enhanced usability, marketing, operations and merchandising. Steve also has extensive international marketing and database marketing experience, having helped a number of companies build their international marketing and customer relationship management programs. John Bartold Vice President Loyalty Solutions — Epsilon John specializes in developing marketing initiatives to build relationships and alter customer behavior for increased profitability and reduced churn.
Craig is responsible for developing the strategic frameworks and plans designed to drive the creation of high impact marketing for both internal and external business partners.
Craig joined Hallmark in and has held positions in product development and several areas of marketing. Craig holds a B. He and his wife, Ann, reside in the Kansas City area.
They have 3 sons and 1 daughter. Michele Fitzpatrick Executive Vice President, Strategy and Marketing - The Agency Inside Harte Hanks Michele brings more than 20 years of marketing experience in strategically designing, managing and evaluating multichannel relationship marketing strategies for leading consumer brands.
Focusing much of her high-energy industry passion into boosting relevance and customer-centricity, Michele has a history of developing strategic roadmaps that bring the personality and value of retail brands to life.
By leveraging and combining the power of analytics, market research, and multiple digital and traditional channels, she creates engaging customer experiences that drive businesses forward and build audience retention.
Michele is a public champion of brand engagement strategies and regularly participates in industry associations and events. Michele holds a B.
CUSTOMER RELATIONSHIP MANAGEMENT
Michael has worked in marketing technology services and consumer marketing for his entire career, and has written many columns for Chief Marketer, DM News and other publications on marketing strategy and technology. InJoseph founded and ran strategic consultancy, crayon, until it was acquired by Powered in January of and again by Dachis Group in December of Inhe launched his first foray into video in the form of JaffeJuiceTV in an effort to prove once and for all that he does not have a face for radio.
Founded in as a co-op of farmers in the San Jose area, Orchard Supply Hardware now has 73 stores in California and Oregon focused on being the destination for all products related to paint, home repair and the backyard.
In addition to her tenure with Orchard, Roxanne has a history of leading similar marketing efforts for highly recognizable brands across a variety of industries. He is focused on leveraging customer data and marketing automation to improve email relevancy, thereby optimizing the CRM channels and increasing revenue per customer.
Matt enjoys live music and over-enthusiastically follows University of Michigan football and basketball. She provides direction to strategic offerings, including omnichannel and digital channels, and she is an expert in engagement, digital, web, mobile, social and eCRM.Hotel transforms customer relationships, wins guests’ hearts with the help of Microsoft Dynamics 365
Be it functionally, architecturally or creatively, Jeannette digs deep to find the right solution for every client and individual project.
Jeannette is a regular contributor for many publications, and also presents frequently at industry events and webinars on website conversion metrics and optimization, mobile marketing, search marketing, and social media marketing. Throughout her career, Jessica has been a pioneer in the growth of eCommerce and the impact technology has had on how brands market to consumers. In her role with Vitamin World, Jessica is responsible for growing the online business as part of a broader omnichannel strategy.
To achieve that goal, Jessica manages a team that leverages online data and incorporates strategic analytics to maximize customer engagement and ROI.
In addition to her professional expertise, Jessica has also spent time as an adjunct instructor of Advertising at the New York Institute of Technology. Rich Lynch is responsible for creating and implementing marketing programs and education offerings for over 3, Do it Best memberowners around the world.
Prior to joining Do it Best Corp. He served in a variety of roles at Ace, including leadership roles in category management, graphics and creative, store development, training, equipment rental, and business to business marketing.
Rich is a graduate of Northern Illinois University with a B. He and his family reside in Huntertown, Indiana. What if you could take action based on direct knowledge of actual consumer spend? By applying sophisticated analytics to 95 billion anonymized, aggregated transactions annually, MasterCard delivers extraordinary insights into consumer spend with you and your competition. Learn about all the ways we can help you better understand your market, and reach the exact customer and prospect segments you need.
For more insight from MasterCard, visit us at www. At Epsilon, in connection with sister company Alliance Data Retail, Kelly helps New York and Company stay engaged with its loyal customer base through implementing PLCC loyalty programs and targeting solutions that embrace constantly changing customer, marketing and technology landscapes. Before joining Epsilon inKelly worked as Strategic Marketing Consultant for Equifax for five years; prior to that she served as retail sector Client Development Manager at Harte Hanks for eight.
Over years of groundbreaking experiments, he has distilled remarkable discoveries significant enough to become the bedrock of a whole new scientific field: We can now predict and change the social structures of companies, governments, communities and much more, to solve some of our most difficult issues.
Social physics is about idea flow, the way human social networks spread ideas and transform those ideas into behaviors. It will change the way we think about how we learn and how our social groups work—and can be made to work better, at every level of society.
In his newest book, Social Physics: How Good Ideas Spread—The Lessons from a New Science, Sandy leads you to the edge of the most important revolution in the study of social behavior in a generation, an entirely new way to look at life itself. He is among the most-cited computational scientists in the world and a pioneer in computational social science, organizational engineering, wearable computing GoogleGlassimage understanding and modern biometrics. Through passionate speeches, he offers revolutionary insights into the mysteries of collective intelligence and social influence.
She continues to play a pivotal role in the ongoing success and evolution of the program, managing the strategic vision and integrated execution across creative, communications, technology and reporting. Jeff is also a co-founder of Warby Parker, a brand offering designer-like eyewear at a revolutionary price while leading the way for socially conscious businesses.