HDP_EXPRESS_LATEST
Download as pptx, pdf0 likes170 views
- EXPRESS is a men's and women's fashion brand that has upgraded its technology systems and is now using Hadoop and Hortonworks to gain insights from customer data. - The goals are to better understand customers, increase loyalty program participation and personalization, and ensure a seamless omnichannel experience. - EXPRESS has loaded customer and web data onto Hadoop and developed APIs to access customer profiles in real-time, allowing tests like comparing weather to online sales. While challenges remain, Hadoop is helping EXPRESS improve operations and marketing.
1 of 12
Download to read offline
Ad
Recommended
Web intelligence
Web intelligenceKoray Kocabas
油
This document discusses big data and analytics. It notes that 70% of customer data is generated but only 0.5% or less is analyzed. It advocates storing all data in one place and using tools like Apache Hive to analyze large datasets stored in Hadoop. Hive allows users to run SQL-like queries on Hadoop. The document provides examples of data sources like sensors, social media, and transactions. It also gives an overview of building an analytics platform with Hadoop, Hive, and related tools to extract insights from big data.How to migrate data from Shopify to Shopify by LitExtension
How to migrate data from Shopify to Shopify by LitExtensionLitExtension
油
LitExtension offers a service to migrate data such as customers, orders, and products between Shopify stores without downtime or technical skills required. The migration process is automated, involves no third-party, and comes with a money-back guarantee along with one year of free technical support. Users can perform the migration by entering source and target cart information, selecting data to migrate, and starting the migration process.Introduction to Cin7 Webinar Presentation
Introduction to Cin7 Webinar PresentationMarc Zabern
油
Cin7 is a cloud-based inventory management and warehouse management system that helps businesses better manage inventory and improve profit margins. It offers modules for contact management, product management, purchase orders, sales orders, reporting, and more. The presentation included a demonstration of the Cin7 system and discussed the company's background, product overview, industries served, solution capabilities, and client testimonials.Sales deck (generic) (5)
Sales deck (generic) (5)Marc Zabern
油
Cin7 provides inventory and POS software to help businesses better manage inventory, leading to improved profit margins. The software offers accurate, visible, and integrated inventory tracking online or from mobile devices. It is integrated with accounting, CRM, and e-commerce systems. Customers say Cin7 helped them avoid inventory problems like stock-outs, reduce overheads, and grow their business by 30%.Measurement Roadmap
Measurement RoadmapAni Lopez
油
This document outlines a measurement roadmap with 7 key steps: 1) Understanding the business and customer lifecycle, 2) Defining a measurement framework, 3) Acquiring data, 4) Establishing data governance, 5) Reporting, 6) Analyzing insights, and 7) Advocating for a measurement culture. Each step provides goals and considerations for identifying relevant data sources and metrics, implementing tools, designing reports and dashboards, and analyzing data to improve strategies and achieve goals. The overall aim is to establish a process for gathering and utilizing measurement data to help organizations make better decisions.Bangalore Executive Seminar 2015: MongoDB - Your database of choice for real ...
Bangalore Executive Seminar 2015: MongoDB - Your database of choice for real ...MongoDB
油
This document discusses real time big data analytics (RTBDA). It explains that data is now generated from many sources and in real time, and analytics needs to be performed on this unstructured data from multiple sources. Real time analytics allows questions to be answered and validated in real time using all available data to make better decisions. It requires a data store that can scale to store large amounts of data and process it quickly, as well as a real time analytics engine and dashboarding technology. The data store needs to store and aggregate data from different sources and structures, scale horizontally as data grows, and support complex queries and aggregations. MongoDB is presented as a data store option that provides the necessary flexibility, agility, scalability,Advanced Analytics Implementations at EA scale
Advanced Analytics Implementations at EA scaleAni Lopez
油
The document discusses the challenges and solutions of advanced analytics implementations at EA, focusing on the transition from traditional to digital analytics amidst complex multi-platform environments. Key initiatives include migrating to Google Analytics, improving global revenue attribution, and enhancing customer data integration through streamlined processes and tools like Tealium. Overall, the presentation highlights the importance of measurement strategies and the value of agility and standardization in managing analytics across multiple properties.Data import and widening in Google Analytics
Data import and widening in Google AnalyticsZorin Radovancevic
油
The document discusses widening in Google Analytics and provides guidance on whether, how, when, and what to widen. It recommends widening by importing reasonable amounts of useful data and using consistent naming conventions while considering privacy, sustainability, and other factors. Widening provides benefits for analyzing, testing, learning, reporting and consuming data. It is best to start widening as soon as possible once data is available. There are some limitations around data import volumes and sampling.Hortonworks laurie maclachlan
Hortonworks laurie maclachlanBigDataExpo
油
This document provides an overview of big data use cases and the Hortonworks Connected Data Platform. It discusses how traditional data constraints like siloed storage environments and stale data limited business insights. The new world of data allows gaining pre-transaction insights, creating new revenue sources, and delighting customers. Case studies describe how Centrica and Royal Mail used Hortonworks to transform their data journeys, gaining a single customer view, optimizing operations, and improving customer retention. Hortonworks supports the full data lifecycle from architecture to operations and provides proactive support through its SmartSense program.Hortonworks - How Hadoop makes the successful Retailer.
Hortonworks - How Hadoop makes the successful Retailer. Mats Johansson
油
The document discusses how Hortonworks' open-source Apache Hadoop platform supports retailers in navigating data challenges and optimizing operations amid a data explosion. It highlights the importance of data-driven insights for enhancing customer engagement, improving inventory management, and fostering personalized experiences, with Zulily as a case study example. Furthermore, it points to the rise of mobile commerce and consumer control over purchasing decisions as critical factors in the retail landscape.Big data analytics
Big data analyticsAmr Kamel Deklel
油
This document provides an overview of big data concepts. It discusses how Gartner's hype cycle model applies to new technologies like big data. It then covers the 3 Vs of big data - volume, velocity and variety. Examples of big data use cases are given, such as using customer log files and social media analytics. The document also discusses how big data fits with business intelligence and provides examples of IBM and other companies' big data solutions.Agile data science
Agile data scienceJoel Horwitz
油
The document discusses the principles of Agile Data Science, emphasizing the importance of collaboration, flexibility, and continuous improvement in data analytics. It outlines the role of business champions, integrated environments, and analytics experts in creating successful data products while highlighting the significance of efficient data handling and interpretation. Through a case study approach, it illustrates the process from data acquisition to actionable insights, promoting a culture of data democracy and shared learning.Paris HUG - Agile Analytics Applications on Hadoop
Paris HUG - Agile Analytics Applications on HadoopHortonworks
油
Russell Jurney discusses strategies for developing agile analytics applications using Hadoop. He advocates for an iterative approach where insights are discovered through exploration of data in an interactive web application from day one. The data model should be consistent end-to-end to minimize impedance between layers and allow insights to grow in scope and depth. Insights formed through this process can then be used to build out the application.Utrecht NL-HUG/Data Science-NL - Agile Data 際際滷s
Utrecht NL-HUG/Data Science-NL - Agile Data 際際滷sHortonworks
油
The document discusses agile analytics applications using Hadoop, emphasizing the importance of iterative development, collaboration among diverse teams, and the need for exploratory data application design. It highlights the advantages of using Hadoop for data processing, including cost efficiency and flexibility, while advocating for the integration of design and data in developing insights. Key strategies for successful analytics include maintaining an interactive work environment and embracing an iterative approach to understanding data.UK - Agile Data Applications on Hadoop
UK - Agile Data Applications on HadoopHortonworks
油
Here are the steps to load the Avro serialized events into Pig:
1) Load the Avro data using AvroStorage():
enron_emails = LOAD '/enron/emails.avro' USING AvroStorage();
2) Describe the schema:
describe enron_emails;
This will show the schema including the fields like message_id, date, from, etc. that were serialized in the Avro data.
3) You can now use Pig operations like FILTER, FOREACH, etc. on the enron_emails relation to extract/transform the data as needed before exporting it for display.
So in summary - LOAD Avro data, DESCRIBAgile Big Data Analytics Development: An Architecture-Centric Approach
Agile Big Data Analytics Development: An Architecture-Centric ApproachSoftServe
油
The document presents an architecture-centric approach to agile big data analytics, emphasizing the integration of architectural practices with agile methodologies to tackle the challenges of big data management. It discusses the 5V requirements for big data and proposes a refined methodology called AABA (Architecture-Centric Agile Big Data Analytics) based on case studies that demonstrate effective agile principles combined with architectural design. The research highlights the need for continuous architecture support and the involvement of data scientists early in the development process to respond to rapid technology changes.Hortonworks: Agile Analytics Applications
Hortonworks: Agile Analytics Applicationsrussell_jurney
油
The document discusses building agile analytics applications using Hadoop. It recommends setting up an environment where insights can be repeatedly produced through iterative and interactive exploration of data. The document emphasizes making an application for exploring data rather than trying to design insights directly. Insights are discovered through many iterations of refining the data and interacting with it.Agile analytics applications on hadoop
Agile analytics applications on hadoopHortonworks
油
The document discusses strategies for developing agile analytics applications using Hadoop, emphasizing an iterative approach where data is explored interactively to discover insights which then form the basis for shipped applications, rather than trying to design insights up front. It recommends setting up an environment where insights are repeatedly produced and shared with the team using an interactive application from the start to facilitate collaboration between data scientists and developers.Deep-Dive: Predicting Customer Behavior with Apigee Insights
Deep-Dive: Predicting Customer Behavior with Apigee InsightsApigee | Google Cloud
油
The document discusses using Apigee Insights to enable personalized experiences through predictive analytics. Insights uses a GRASP technology to analyze sequential customer behavior patterns at scale from big data sources. This allows building predictive models to anticipate customer needs and adapt interactions in real-time across channels. The platform provides segmentation, predictions, and an interaction layer to deliver the right offer to the right customer at the right time.Big Data and User Segmentation in Mobile Context
Big Data and User Segmentation in Mobile ContextInMobi Technology
油
The document discusses the rapid growth of mobile data and user segmentation, highlighting InMobi's expansive reach and technology infrastructure. It details unique aspects of mobile data, including rich context, short half-life, and fragmentation challenges, while proposing solutions for scalable user segmentation and analytics. The document also emphasizes the importance of real-time data processing and machine learning in enhancing mobile advertising efforts.Hortonworks and HP Vertica Webinar
Hortonworks and HP Vertica WebinarHortonworks
油
The document discusses the integration of HP Vertica with Apache Hadoop for enhancing customer analytics by leveraging big data. It highlights market trends, the capabilities of modern data architectures, and various applications in industries like retail, healthcare, and telecommunications. Additionally, it emphasizes the importance of using Hadoop and Vertica to handle vast amounts of data and improve business intelligence through actionable insights.Big Data LDN 2016: Case Studies of Business Transformation through Big Data
Big Data LDN 2016: Case Studies of Business Transformation through Big DataMatt Stubbs
油
The document discusses the future of data and how organizations can maximize the value of data through connected data platforms. It notes that traditional data constraints like siloed storage environments and difficulty combining data formats limited business insights. The new approach of connected data platforms powered by open source technologies like Hortonworks can help organizations gain customer insights, create new revenue sources, and act on opportunities faster by breaking down data silos and enabling analytics across all data in motion and at rest. Case studies of companies like Centrica and Royal Mail are provided that demonstrate how they used connected data platforms to improve customer service, reduce costs, and unlock new business opportunities from their data.LA HUG - Agile Analytics Applications on HDP
LA HUG - Agile Analytics Applications on HDPHortonworks
油
The document discusses publishing structured event data to applications for analytics. It describes serializing event data from sources like emails and logs into Avro documents and loading them into Pig for processing. The data is then published from Pig to databases and analytics stacks using various options like ElasticSearch, MongoDB, HBase, and Hive/HCatalog for exploration and building analytics applications. Code examples demonstrate loading Avro data into Pig, illustrating the data schema, and publishing the data from Pig to MongoDB. The overall approach emphasizes agility, iteration, and flexibility in building analytics applications on Hadoop.Leverage Big Data to Enhance Customer Experience in Telecommunications with...
Leverage Big Data to Enhance Customer Experience in Telecommunications with...Hortonworks
油
The document discusses leveraging big data in telecommunications to enhance customer experience, focusing on managing customer data from various sources. It highlights the need for a centralized architecture via the Hortonworks Data Platform to handle data efficiently and derive insights for customer engagement. The presentation also covers the roles of machine learning in predicting customer behaviors and improving satisfaction metrics like the Net Promoter Score.Hadoop Data Reservoir Webinar
Hadoop Data Reservoir WebinarPlatfora
油
The document outlines the concept of the Hadoop Data Reservoir (HDR), which serves as a central cluster for enterprise data storage and analytics, aiming to eliminate data silos and promote agile business intelligence. It highlights the critical requirements for HDR, including performance, self-service capabilities, and robust security measures. Through various use cases, it demonstrates how organizations can leverage HDR to optimize data access and analytics processes.Analyzing Multi-Structured Data
Analyzing Multi-Structured DataDataWorks Summit
油
The document outlines the application of big data analytics, particularly using Hadoop for analyzing multi-structured data. It describes various analytical models, business opportunities, and challenges faced in leveraging big data, highlighting specific examples from companies like Amazon and Netflix. The text emphasizes the integration of Hadoop with RDBMS and the use of advanced analytics to enhance performance and decision-making in diverse sectors.Orange County HUG - Agile Data on HDP
Orange County HUG - Agile Data on HDPHortonworks
油
The document discusses strategies for developing agile analytics applications on Hadoop, emphasizing an iterative approach where the data model and insights evolve through exploration of data in an interactive web application, rather than trying to design insights up front, in order to discover insights rather than define them. It also highlights using techniques like storing data in documents rather than relational structures and using Pig for its ability to handle diverse data formats.Taming data lake - scalable metrics model
Taming data lake - scalable metrics modelBharathiRaja Chandrasekaran
油
The document discusses setting up a Scalable Metrics Model (SMM) for a data lake. The SMM uses pre-aggregated metrics and dimension keys to provide a flexible and scalable approach. An example model from retail is described, including defining customer-level dimensions, metrics, and value-added metrics. The document outlines the ETL process, including writing queries to populate metrics for different granularities and integrating them into a framework. Reporting options are also discussed, including using views, tables, and reporting tools to provide accessibility and analysis capabilities. Finally, the extensibility of the model by adding new dimensions like sellers and transactions is highlighted.More Related Content
Similar to HDP_EXPRESS_LATEST (20)
Hortonworks laurie maclachlan
Hortonworks laurie maclachlanBigDataExpo
油
This document provides an overview of big data use cases and the Hortonworks Connected Data Platform. It discusses how traditional data constraints like siloed storage environments and stale data limited business insights. The new world of data allows gaining pre-transaction insights, creating new revenue sources, and delighting customers. Case studies describe how Centrica and Royal Mail used Hortonworks to transform their data journeys, gaining a single customer view, optimizing operations, and improving customer retention. Hortonworks supports the full data lifecycle from architecture to operations and provides proactive support through its SmartSense program.Hortonworks - How Hadoop makes the successful Retailer.
Hortonworks - How Hadoop makes the successful Retailer. Mats Johansson
油
The document discusses how Hortonworks' open-source Apache Hadoop platform supports retailers in navigating data challenges and optimizing operations amid a data explosion. It highlights the importance of data-driven insights for enhancing customer engagement, improving inventory management, and fostering personalized experiences, with Zulily as a case study example. Furthermore, it points to the rise of mobile commerce and consumer control over purchasing decisions as critical factors in the retail landscape.Big data analytics
Big data analyticsAmr Kamel Deklel
油
This document provides an overview of big data concepts. It discusses how Gartner's hype cycle model applies to new technologies like big data. It then covers the 3 Vs of big data - volume, velocity and variety. Examples of big data use cases are given, such as using customer log files and social media analytics. The document also discusses how big data fits with business intelligence and provides examples of IBM and other companies' big data solutions.Agile data science
Agile data scienceJoel Horwitz
油
The document discusses the principles of Agile Data Science, emphasizing the importance of collaboration, flexibility, and continuous improvement in data analytics. It outlines the role of business champions, integrated environments, and analytics experts in creating successful data products while highlighting the significance of efficient data handling and interpretation. Through a case study approach, it illustrates the process from data acquisition to actionable insights, promoting a culture of data democracy and shared learning.Paris HUG - Agile Analytics Applications on Hadoop
Paris HUG - Agile Analytics Applications on HadoopHortonworks
油
Russell Jurney discusses strategies for developing agile analytics applications using Hadoop. He advocates for an iterative approach where insights are discovered through exploration of data in an interactive web application from day one. The data model should be consistent end-to-end to minimize impedance between layers and allow insights to grow in scope and depth. Insights formed through this process can then be used to build out the application.Utrecht NL-HUG/Data Science-NL - Agile Data 際際滷s
Utrecht NL-HUG/Data Science-NL - Agile Data 際際滷sHortonworks
油
The document discusses agile analytics applications using Hadoop, emphasizing the importance of iterative development, collaboration among diverse teams, and the need for exploratory data application design. It highlights the advantages of using Hadoop for data processing, including cost efficiency and flexibility, while advocating for the integration of design and data in developing insights. Key strategies for successful analytics include maintaining an interactive work environment and embracing an iterative approach to understanding data.UK - Agile Data Applications on Hadoop
UK - Agile Data Applications on HadoopHortonworks
油
Here are the steps to load the Avro serialized events into Pig:
1) Load the Avro data using AvroStorage():
enron_emails = LOAD '/enron/emails.avro' USING AvroStorage();
2) Describe the schema:
describe enron_emails;
This will show the schema including the fields like message_id, date, from, etc. that were serialized in the Avro data.
3) You can now use Pig operations like FILTER, FOREACH, etc. on the enron_emails relation to extract/transform the data as needed before exporting it for display.
So in summary - LOAD Avro data, DESCRIBAgile Big Data Analytics Development: An Architecture-Centric Approach
Agile Big Data Analytics Development: An Architecture-Centric ApproachSoftServe
油
The document presents an architecture-centric approach to agile big data analytics, emphasizing the integration of architectural practices with agile methodologies to tackle the challenges of big data management. It discusses the 5V requirements for big data and proposes a refined methodology called AABA (Architecture-Centric Agile Big Data Analytics) based on case studies that demonstrate effective agile principles combined with architectural design. The research highlights the need for continuous architecture support and the involvement of data scientists early in the development process to respond to rapid technology changes.Hortonworks: Agile Analytics Applications
Hortonworks: Agile Analytics Applicationsrussell_jurney
油
The document discusses building agile analytics applications using Hadoop. It recommends setting up an environment where insights can be repeatedly produced through iterative and interactive exploration of data. The document emphasizes making an application for exploring data rather than trying to design insights directly. Insights are discovered through many iterations of refining the data and interacting with it.Agile analytics applications on hadoop
Agile analytics applications on hadoopHortonworks
油
The document discusses strategies for developing agile analytics applications using Hadoop, emphasizing an iterative approach where data is explored interactively to discover insights which then form the basis for shipped applications, rather than trying to design insights up front. It recommends setting up an environment where insights are repeatedly produced and shared with the team using an interactive application from the start to facilitate collaboration between data scientists and developers.Deep-Dive: Predicting Customer Behavior with Apigee Insights
Deep-Dive: Predicting Customer Behavior with Apigee InsightsApigee | Google Cloud
油
The document discusses using Apigee Insights to enable personalized experiences through predictive analytics. Insights uses a GRASP technology to analyze sequential customer behavior patterns at scale from big data sources. This allows building predictive models to anticipate customer needs and adapt interactions in real-time across channels. The platform provides segmentation, predictions, and an interaction layer to deliver the right offer to the right customer at the right time.Big Data and User Segmentation in Mobile Context
Big Data and User Segmentation in Mobile ContextInMobi Technology
油
The document discusses the rapid growth of mobile data and user segmentation, highlighting InMobi's expansive reach and technology infrastructure. It details unique aspects of mobile data, including rich context, short half-life, and fragmentation challenges, while proposing solutions for scalable user segmentation and analytics. The document also emphasizes the importance of real-time data processing and machine learning in enhancing mobile advertising efforts.Hortonworks and HP Vertica Webinar
Hortonworks and HP Vertica WebinarHortonworks
油
The document discusses the integration of HP Vertica with Apache Hadoop for enhancing customer analytics by leveraging big data. It highlights market trends, the capabilities of modern data architectures, and various applications in industries like retail, healthcare, and telecommunications. Additionally, it emphasizes the importance of using Hadoop and Vertica to handle vast amounts of data and improve business intelligence through actionable insights.Big Data LDN 2016: Case Studies of Business Transformation through Big Data
Big Data LDN 2016: Case Studies of Business Transformation through Big DataMatt Stubbs
油
The document discusses the future of data and how organizations can maximize the value of data through connected data platforms. It notes that traditional data constraints like siloed storage environments and difficulty combining data formats limited business insights. The new approach of connected data platforms powered by open source technologies like Hortonworks can help organizations gain customer insights, create new revenue sources, and act on opportunities faster by breaking down data silos and enabling analytics across all data in motion and at rest. Case studies of companies like Centrica and Royal Mail are provided that demonstrate how they used connected data platforms to improve customer service, reduce costs, and unlock new business opportunities from their data.LA HUG - Agile Analytics Applications on HDP
LA HUG - Agile Analytics Applications on HDPHortonworks
油
The document discusses publishing structured event data to applications for analytics. It describes serializing event data from sources like emails and logs into Avro documents and loading them into Pig for processing. The data is then published from Pig to databases and analytics stacks using various options like ElasticSearch, MongoDB, HBase, and Hive/HCatalog for exploration and building analytics applications. Code examples demonstrate loading Avro data into Pig, illustrating the data schema, and publishing the data from Pig to MongoDB. The overall approach emphasizes agility, iteration, and flexibility in building analytics applications on Hadoop.Leverage Big Data to Enhance Customer Experience in Telecommunications with...
Leverage Big Data to Enhance Customer Experience in Telecommunications with...Hortonworks
油
The document discusses leveraging big data in telecommunications to enhance customer experience, focusing on managing customer data from various sources. It highlights the need for a centralized architecture via the Hortonworks Data Platform to handle data efficiently and derive insights for customer engagement. The presentation also covers the roles of machine learning in predicting customer behaviors and improving satisfaction metrics like the Net Promoter Score.Hadoop Data Reservoir Webinar
Hadoop Data Reservoir WebinarPlatfora
油
The document outlines the concept of the Hadoop Data Reservoir (HDR), which serves as a central cluster for enterprise data storage and analytics, aiming to eliminate data silos and promote agile business intelligence. It highlights the critical requirements for HDR, including performance, self-service capabilities, and robust security measures. Through various use cases, it demonstrates how organizations can leverage HDR to optimize data access and analytics processes.Analyzing Multi-Structured Data
Analyzing Multi-Structured DataDataWorks Summit
油
The document outlines the application of big data analytics, particularly using Hadoop for analyzing multi-structured data. It describes various analytical models, business opportunities, and challenges faced in leveraging big data, highlighting specific examples from companies like Amazon and Netflix. The text emphasizes the integration of Hadoop with RDBMS and the use of advanced analytics to enhance performance and decision-making in diverse sectors.Orange County HUG - Agile Data on HDP
Orange County HUG - Agile Data on HDPHortonworks
油
The document discusses strategies for developing agile analytics applications on Hadoop, emphasizing an iterative approach where the data model and insights evolve through exploration of data in an interactive web application, rather than trying to design insights up front, in order to discover insights rather than define them. It also highlights using techniques like storing data in documents rather than relational structures and using Pig for its ability to handle diverse data formats.Taming data lake - scalable metrics model
Taming data lake - scalable metrics modelBharathiRaja Chandrasekaran
油
The document discusses setting up a Scalable Metrics Model (SMM) for a data lake. The SMM uses pre-aggregated metrics and dimension keys to provide a flexible and scalable approach. An example model from retail is described, including defining customer-level dimensions, metrics, and value-added metrics. The document outlines the ETL process, including writing queries to populate metrics for different granularities and integrating them into a framework. Reporting options are also discussed, including using views, tables, and reporting tools to provide accessibility and analysis capabilities. Finally, the extensibility of the model by adding new dimensions like sellers and transactions is highlighted.HDP_EXPRESS_LATEST
- 1. Big Data with Hortonworks Steve Howard, EXPRESS Eric Thorsen, Hortonworks
- 2. Who are we? Womens fashion brand started by Limited Brands in 1980 Added a mens line in 1986 Mens spun off into its own brand, Structure, in 1989 Reintegrated Structure back into EXPRESS in 2000 as a dual gender brand Limited Brands sold a majority stake in 2007 to GoldenGate Capital The company issued an IPO of its stock in May, 2010 Target market is 20 30 year olds
- 3. Where have we been? Upgrading application and technology portfolio Upgrading internal processing systems such as GL and HR Brought ecommerce in-house from hosted platform Installed allocation and size profile systems Brought order management system in-house Currently installing new merchandising and planning systems Allows us to have: More flexible merchandise and organization hierarchy Robust reporting Supported technology stack to allow for organizational and associate engagement and growth
- 4. Business Goals Collecting stats on Millennial shopping behavior Increase purchases through Loyalty program Improve conversion to loyalty program Seeking greater personalization of marketing messages Visibility of customer across all channels Single cart 360属 view of customer Better positioned to offer: what they want when they want it where they want it
- 5. What are we doing? Need a processing platform that allows us to better understand our customers at scale, in as close to real time as possible Use Hive for large scale analytics Use same Tableau reporting solution as existing datastores Use HBase for storing customer profiles with real time access Custom spring based API
- 6. Hadoop/Hortonworks Initially hosted our customer data warehouse with an external partner Large capital and expense outlay Inflexibility with reporting changes 2015 Began buildout of HDP cluster that would allow us to: Bring data warehouse in house Provide data directly to external partners for deep data analytics Run operational reports on customer activity Load website request logs to Hadoop for omnichannel analysis 2016 Provide data to campaign management partner Develop API layer to provide real time access to customer profile records
- 7. How did we get started? Install software Secure it Quick POCs/quick wins
- 8. Weather compared to online checkouts Story Are customers more likely to shop online when bad weather hits a location close to them Plan Identify data sources Load data Construct query Results No correlation whatsoeverthats OK! Sometimes the answer no is as good as the answer yes. Actually, that is always true.
- 9. Semantic analysis of the twitter feed Story - Could we build a language scoring tool to determine if a given tweet was positive or negative given a search filter? Plan How to obtain data? Twitter firehose Twitter gardenhose Results Good We were able to prove we could load the data realtime using Apache Storm Bad We found that semantic analysis is very hard to get right For example, yeah, Im going to shop there again real soon! Is that very positive, or incredibly sarcastic?
- 10. Using web logs to tie shopper behavior across channels Story Customers purchase in store, browse online, how to track Plan Customers dont always log in, so how to track? Omniture cookies? Added our own cookie Ensure we protect data at all times Results We were able to identify shoppers online vs in store Looking at operationalization of data flow
- 11. Lessons learned Identify all data sources Security early Allow people time to experiment with the technology Semantic analysis of twitter feed - used python natural language toolkit to score words in tweets with search Communicate early and often Identify all data sources (web logs, transaction systems, social feeds, etc.) Come up with a list of POCs for important business questions Failure is OK, just communicate it and document it Utilize Hortonworks support for problems