European Data Market SMART 2013/0063 D.3.1 Data-driven innovation in the Retail Industry

May 16, 2014

Author(s)

Rosanna Lifonti, Alis Woodward, Giorgio Micheletti, Gabriella Cattaneo, (IDC)

Deliverable

D 3.1 Quarterly Stories – Story 1

Date of delivery

May 16, 2014

Version

2.0

Addressee officer

Katalin IMREI Policy Officer European Commission - DG CONNECT Unit G3 – Data Value Chain EUFO 1/178, L-2557 Luxembourg/Gasperich [email protected]

Contract ref.

N. 30-CE-0599839/00-39

2

TABLE OF CONTENTS

1

SHORT SUMMARY ....................................................................................................... 5

2

OVERVIEW .................................................................................................................... 6 2.1

INTRODUCTION ................................................................................................................. 6

2.2

THE DATA-DRIVEN EVOLUTION OF THE RETAIL INDUSTRY................................................... 7

2.3

OVERVIEW OF THE CASE STUDIES ..................................................................................... 8

2.3.1

The case of Morrison’s ................................................................................................................. 8

2.3.2

The case of SportCheck ............................................................................................................... 9

2.3.3

The case of OTTO ......................................................................................................................... 9

2.4

TECHNOLOGY INNOVATION AND THE ROLE OF STAKEHOLDERS ............................................ 9

2.4.1

Morrison’s solution: Smart Steps by Telefonica .......................................................................... 9

2.4.2

Sportscheck and Otto’s solution: Blue Yonder’s Predictive Analytics........................................ 10

2.5

DATA-DRIVEN INNOVATION BENEFITS AND IMPACTS.......................................................... 11

2.5.1

Forecasting demand and returns ............................................................................................... 11

2.5.2

Data-driven marketing and loyalty systems .............................................................................. 13

2.6

CONCLUSIONS................................................................................................................ 14

MAIN SOURCES ......................................................................................................................... 15

3

4

1

Short Summary The retail industry is pioneering the evolution towards data-driven innovation, showing how data can become a core asset creating significant competitive advantages. According to an IBM study together with the University of Oxford, 62% of retailers in 2012 agreed that the use of advanced data analytics was improving their competitiveness. Keeping pace with the new technology-savvy consumers and influencing their choices through personalized marketing is becoming a new standard of competition. As shown by the case studies presented in this story, European retailers are rising to the challenge, with the help of emerging newcomers (such as Blue Yonders, created by former CERN scientists) and traditional operators entering the new market (Telefonica’s new Dynamic Insights division). Positive effects range from 150% increases in customers visits for British supermarket chain Morrison’s, to 40% increases in the accuracy of sales forecasts improving the efficiency of stock management for the German chain Sportscheck, to 6 million pounds in annual cost savings for global retailer Tesco, to the halving of product return rates for German retailer Otto. These are all strong signals of the digital transformation of the retail industry towards data-driven processes. The retail industry has long been focused on data usage in many parts of its businesses: managing transactions in detail via electronic points of sale (EPOS), analyzing these transactions in data warehouses, tracking responses to advertising, optimizing supply chain and store restocking systems, and many more. However, the volume, variety and velocity of the data being used by the new data-driven applications represent a change of scale with disruptive impacts. Retailers are facing disruptive innovation trends which are re-shaping the way they relate with customers, as well as the way they work and compete. Consumers can look for the desired article on the Internet, can go to try and touch it in a store and buy it from a mobile device at the best price they can find. Consumers are much more informed than they were in the past, loyalty involves many more factors than it did and demand forecasts become strategic to efficiently manage storage. Today’s demand forecasting, data-driven delivery services, pricing analytics require the processing of a very large amount of data, making the traditional forecasting methods less reliable than in the past. In contrast, the innovative use of data technologies is improving the accuracy of forecasting by 20% to 40%, as experienced by some large retailers. Goods returns and returns management have become the number one critical-success factor in online retail commerce. Returns are in fact a fundamental part of the online business model: mail order affords the customer the advantage of being able to try things at home and then being able to decide whether to buy them, and, if so desired, be able to send part of them back. For example Otto, a large German retailer, prevented 2 million article returns and saved €10m, thanks to accurate demand forecasting based on predictive analytics and greater efficiency in stocks management. Finally, last generation data-driven, cloud-based solutions are starting to displace expensive CRM systems, because they require lower initial investments and rely on flexible, pay-as-yougo delivery systems with immediate returns.

5

2

Overview

2.1

Introduction

The present story focuses on the data technologies’ impacts on the European retail industry with specific reference to competitiveness’ improvements. The table below outlines the story’s general information and the key elements of the story’s description. The document is structured in three main parts:  



In the first paragraph an overview of the technology-driven evolution of the European retail industry will be provided; In the second paragraph a few case studies from the European retail industry will be presented: the stakeholders involved in those case studies, the technology used and the impacts obtained by the innovative use of this technologies will be investigated; The third paragraph will summarize the key messages and conclusions to be taken from the case studies.

GENERAL INFORMATION Title

Link with /Sources

The Impact of Data Technologies on the Retail Industry - Innovative Uses of Information to improve Competitiveness information

http://dynamicinsights.telefonica.com/488/smart-steps http://blog.digital.telefonica.com/?press-release=telefonica-dynamic-insightslaunches-smart-steps-in-the-uk http://www.blue-yonder.com/en/

Interviews

Dunja Riehemann, Marketing Director, Blue Yonder Nicky Guy, Independent Marketing Consultant, Guy Marketing, working under contract for Blue Yonder

STORY DESCRIPTION Topic/object of story

How innovative data exploitation is impacting on the retail industry

Main examples

The relevant business cases taken as example are Morrison’s, Otto, Tesco, SportScheck Cost saving

Main impacts identified in this story

Revenue generation Profitability Advanced predictive analytics company (Blue Yonder)

Main stakeholders

Telecom company (Telefonica, as data holder) Retailers (Morrison’s, Otto, Tesco, SportScheck) Market research company (GFK)

Key words

Advanced predictive analytics, Big data, demand forecasting, data-driven decisions, data-driven marketing, data-driven pricing

6

2.2

The Data-Driven Evolution of the Retail Industry

The retail industry has long been focused on data usage in many parts of its businesses: managing transactions in detail via EPOS, analyzing these transactions in data warehouses, tracking responses to advertising, optimizing supply chain and store restocking systems, and many more. Despite this strong use of data, the volume, variety and velocity of the data being used by the new data-driven applications represent a change of scale with disruptive impacts. The increase of mobility in terms of ubiquitous access to the mobile internet and the ability to perform transactions on a mobile device, coupled with the use of social media are all clear examples of how more access to more data represents both an opportunity and a challenge to retailers who are trying to truly understand and influence their customers across a world of different channels. A powerful framework to apprehend the current transformation of retail consumers is represented by the “Five-Is”. According to this framework, consumers are:     

instrumented with mobile devices, informed with access to the Internet on their devices, interconnected in social communities, always in place in stores or wherever else they might be, immediate in their ability to take action.

The different aspects covered by the Five Is show the complexity of trying to connect consumers' actions together across different channels (mobile devices, Internet, social communities, stores). What is more, all these channels play a direct role in the so called “buying journey”, that is the consumer’s activities devoted to research, comparison, choice and final purchase of products and services. Within this context, the innovative use of information provided by data technology becomes of paramount importance for the retail industry as it allows retailers to:   

leverage data shared by both consumers and retailers; connect the different stages of the consumers’ buyer journey; develop effective ways to influence consumers’ purchases.

More and more consumers in Europe are in fact beginning to research a product on a mobile app, purchase it online, ask for home delivery or pick it up at a store. Marketing means coordinating this multi-channel interaction which will require new data competencies based on managing, integrating, understanding wide sets of data flows. A recent study conducted by the IBM Institute for Business Value partnered with the Saïd Business School at the University of Oxford showed that 62 percent of retailers report that the use of information (including big data) and analytics is creating a competitive advantage for their organizations. The percentage of retail respondents reporting a competitive advantage rose from 30 percent in 2010 to 62 percent in 2012, a 107 percent increase in two years after a peak in 2011 at 66 percent. Most organizations are currently in the early stages of big data planning and development efforts with retailers slightly lagging the global pool of cross-industry counterparts (IBM Institute, 2012). Some evidence of data-driven decisions and operations are available mainly in the USA market but some examples are now available in Europe too. The current story will therefore present and expand on some of the most relevant examples in the European retail industry and will investigate some of the few key themes and related issues that are shaping the evolution of the emerging European data market today, namely: process and process innovation, structure and behaviour of the demand and costs and cost-related issues.

7

Table 1 Benefits and impacts of data exploitation in the retail industry subdivided by retailer function

Retailer function

Effect description

Impact on competitiveness and or performance

Supply chain analytics

Optimisation of supply chain flows

Cost savings

Demand forecasting

Forecast demand accurately and flexibly in order to meet demand as fully as possible while also minimizing overstocks.

Cost savings, revenue generation, optimized profitability, customer loyalty

Customer management analytics

Customer selection and customer loyalty: identify customers needing the products or services delivered, identify the customers with the greatest profit potential,

Revenues increase, profit increase, market share increase

Marketing analytics

Data-driven marketing: provide targeted advertisement and recommendations (cross-selling and up-selling)

Revenue increase, market share increase

Pricing analytics

Identify the price that will maximize return on investment or profit

Revenues increase, profit increase

Data-driven Production and delivery services

Detect and minimize quality problem

Cost savings, customer satisfaction improvement

2.3

Overview of the Case Studies

The case studies selected to illustrate the impact of data technologies on the retailer industry concern three European retailers: Morrison’s, the fourth-largest supermarket retailer in the UK, SportScheck, one of the leading leisure and sport retailers in Germany and Otto, a German multi-channel retailer. In the first case, the innovative solution is a cloud-based service directed to identify and target customers. In the other two cases, the innovative solution is based on predictive analytics. 2.3.1

The case of Morrison’s

With more than 400 stores in the highly competitive UK market, Morrison’s wanted to increase the number of visitors to its stores without investing in a huge loyalty card program. The company decided for the mechanism of supplying coupons to people in selected postal areas. The usual challenge faced by supermarkets that use this method and do not have loyalty schemes is the identification of the potential customer to target. Morrison’s initially started to distribute the coupons by post code in the UK basing this decision on a mathematical (estimated) model. In 2012, Morrison’s decided to adopt Smart Steps, a cloud-based innovative service from Dynamic Insights - a Telefonica Business Unit. Smart Steps uses customer location data in an aggregated and anonymized form to monitor crowd movements or "footfall". By using this service, the British supermarket was able to improve strongly their ability to target and identify customers and therefore to increase footfall.

8

2.3.2

The case of SportCheck

SportScheck, one of the most important leisure and sporting retailers in Germany, deals with more than 30,000 items from 400 brands; many of its staff are athletes themselves, and regularly review the quality of the equipment in joint tests with customers. It also deals with 4 main catalogs, 16 stores located across Germany and an online shop which counts 52 million visits per year. SportScheck’s business success is heavily dependent on accurate sales predictions, which are to be based on both the off-line and the on-line channels; the online channel is increasing complexity to the demand analysis and predictions become therefore more critical and less reliable. To overcome this problem, SportScheck started using the Blue Yonder (http://www.blueyonder.com/en/) data pattern and predictive analytics to make accurate predictions on demand and plan supplies. 2.3.3

The case of OTTO

OTTO, a German multi-channel retailer, employed data-driven services to better manage its ecommerce activity. Otto’s online shop is the focus of the retailer‘s business accounting for 80% of its annual sales from over 2 billion Euros. Alongside fashion items and technical products, OTTO also sells furniture, sports articles, shoes, and toys. The online shop has a total of about 4,000 brands and more than two million article items. OTTO started using demand forecasting from software vendor Blue Yonder to reduce return rates on fashion items. If fashion items don't sell in the right season, they have to be discounted, which reduces profitability. Blue Yonder's technology ingests the customers own past sales data, and uses powerful predictive analytics to forecast future demand. It also integrates external data, where appropriate; for example, when forecasting demand for perishable goods the technology integrates weather information. The underlying analytics technology was originally developed at CERN, where it was used to make predictions for the physics of particle acceleration.

2.4

Technology innovation and the role of stakeholders

The three case studies illustrate different approaches towards data-driven processes, where stakeholders play different roles in the development and deployment of innovation. 2.4.1

Morrison’s solution: Smart Steps by Telefonica

In the Morrison’s case, the service was provided by European telecom operators Telefonica with the collaboration of market research company GFK. Telefónica launched Telefónica Dynamic Insights (Dynamic Insights) in October 2012- a business unit it describes as "Big Data". Organizationally the new business unit sits within the Telefónica Digital, an innovation-focused business division, with the purpose was to build and market solutions based on insights gleaned from its customer location data. Dynamic Insights solutions are based on mobile data that has been anonymised and aggregated, to respect data privacy regulation. The solutions measure observed behavior of crowds rather than individuals, to demonstrate the impact of a marketing campaign, a competitor store opening or even a change in a store’s opening hours on society. This helps businesses and local government to make better decisions.

9

The first product from Dynamic Insights is called Smart Steps, a cloud-based service. Released in November 2012, it uses customer location data in an aggregated and anonymized form to monitor crowd movements or "footfall". Smart Steps is a product in partnership with the Consumer Choices unit of market research firm GfK, which had a long history in conducting footfall research for retailers. The product is available in three countries (Brazil, Germany and the U.K.). By using innovative technologies and data sciences, GfK turns big data into analytical knowledge, supporting clients in the decision making processes. The product delivered by Telefonica Dynamic Insights is a dashboard of anonymized customer footfall data, updated hourly, and delivered over the cloud as a monthly subscription service with a 12-months minimum subscription period. The Smart Steps dashboard shows the previous day's data in hourly totals, and shows a measure of footfall that users can then explore across a number of dimensions including demographics, and other events happening in the local area. The footfall measure is not the raw count of Telefónica customers in the specific area, rather it is an extrapolation, based on a number of algorithms. The core data used for the retailer Morrison’s is supplied by the main Telefonica business. The cost of integrating the data into the Smart Steps product is partly covered by the initial investment to set up the business unit, and there is an ongoing operational cost as well.

Figure 1 Screenshot from Telefónica Dynamic Insights’ Smart Steps offering, showing the footfall grid over a week’s view

2.4.2

Sportscheck and Otto’s solution: Blue Yonder’s Predictive Analytics

Predictive Analytics is a specific category of the new business intelligence solutions based on advanced data technologies. Sportscheck and Otto used a SaaS (software as a service) solution designed and delivered by Blue Yonder. Founded in 2008, Blue Yonder is a privately held company headquartered in Karlsruhe, Germany, with additional offices in Hamburg (Germany) and London (the U.K.). The founder, Dr. Michael Feindt, is a particle physicist from the University of Karlsruhe who worked at the European research center CERN (European Organization for Nuclear Research). The company has around 100 employees, of whom more than 70 are focused on the product in technical roles such as PhD data scientists, software engineers, and UI/UX designers. This is truly a 10

company run by data scientists. Blue Yonder does not disclose revenue, but in the past two years it is estimated to have grown software-related revenues at 150% or more (IDC, 2014). Blue Yonder's solution used in our case studies is named Blue Yonder Predictive Analytics Suite. It combines neural networks (for calculating the outcome compared to a number of factors) and Bayesian algorithms. The product combines data extraction automation software for internal and external data sources, statistical algorithms, including probability density functions, and machine learning. This leads to dynamic and adaptive decision making and event processing functionality. Also it produces forecasts that can be embedded in business processes. However, for effective usage it requires the client to have a clear understanding of his/her needs: to be able to ask the right questions. So, even if the service is automated via cloud, in the initial implementation process Blue Yonder’s technicians act as consultants with the client to focus their needs. A key advantage of this technology is to use machine learning in order to avoid errors induced by human actions. This approach builds on CERN’s experience, when machine learning was used in a way that generated incorrect results due to human interventions, rather than wrong algorythms. Thus, the BlueYonder solution has been deliberately built to be self-contained, without the need for trained statisticians to use it: the deep statistical intelligence is in the platform and is not required by the end user. Blue Yonder has experienced success in Germany within retail and manufacturing, and its U.K. business expanded rapidly since its launch in October 2013. More recently, the company has moved beyond the focus on the above mentioned industries into what Blue Yonder calls "industrial data analytics": looking at forecasting related to processes in heavy industry, such as automotive and engineering, where traditional techniques have been able to optimize the performance of a single machine in great depth, but have struggled to optimize a range of interrelated machines. What Blue Yonder delivers to its customers is a set of industry-specific predictive applications, with the objective of giving answers to business users. Forecast results can be, and are expected to be, embedded in the customer's core systems so business users can see an integrated view of actual results and forecast or calculated data. The product is not a toolset and does not support free form or ad hoc exploratory use cases. Blue Yonder understands well that this is an emerging market and must educate its clients. Coherently with their original vocation as scientists, the company managers have set up a European Data Science Academy, a series of courses designed to help companies bring data science knowledge and understanding into their organization.

2.5

Data-driven Innovation Benefits and Impacts

Our case studies are recent, since the retailers started using the new data-driven services on average less than one year ago. As a consequence, there is some evidence about early returns and benefits, but consolidated impacts data is not yet available. We analyse here the most visible and relevant effects. 2.5.1

Forecasting demand and returns

Multi-channel retail enterprises are finding themselves in a competitive environment characterized by increasingly faster changing market conditions. Decision-making processes are more and more characterized by such relevant data volumes and by such a relevant number of influential factors that conventional statistic processes are no longer able to meet the new requirements. For the retail industry, with its low margins, demand forecasting is critical to maximize efficiency and reduce waste. For online commerce, this is even more strategic since the share of goods returns is typically much higher than in traditional commerce. Returns are in fact a fundamental part of the e-

11

commerce business model, since customers like to order, try out deliveries at home, and then keep or return their orders. Minimizing goods returns and managing well the interaction with the customer can arguably be considered as the number one critical success factor in online retail commerce, because of their impact on costs and profitability. Reducing returns optimizes both profitability, because the full price can be obtained for goods, and revenues, because all customer demand can be satisfied. Predictive analytics basically helps to improve the accuracy of forecasting orders and returns, leading to improved revenues and profitability, by combining data on customers socio-demographic characteristics and past purchase behavior with their current orders. For example (Table 2), German retailer Otto used Blue Yonder’s predictive analytics system to forecast return rates per product by analysing data on customer behaviour, product characteristics, delivery times, pricing and costs going back to 2007. Otto’s analytics team ran through 10 different hypotheses for different situations in which shoppers might return clothes. For example, the project found a correlation between late deliveries and returns. As a result, Otto has worked with its logistic providers to tighten delivery times and reduce delays. Thanks to this service, Otto estimated cost savings between €10m and €15m in one year, from October 2012 to October 2013, due to the reduction of the cost of returns and overstocking. The company also declared an improvement of forecasting accuracy by 40%, meaning that in the same period the forecast demand was 40% closer to the actual sales than previous estimates (Retail Week, October 2013). Otto also improved their gross profitability on men’s fashion items by introducing dynamic pricing using software from Blue Yonder. The technology means prices can be changed based on demand, and it can forecast what prices customers will accept on a particular day. This is a sophisticated application of predictive analytics which represents a fundamental revolution of pricing mechanisms, previously applied only for online services such as airline tickets sales.

“Using Blue Yonder, our forecast quality constantly improves and the forecasted sales volumes are becoming more and more accurate. They support us in getting ready for future developments, early.” Michael Sinn, Director of Category Support at OTTO Extracted from Customer Case Study, by Blue Yonder

Otto regularly uses Blue Yonder sales forecasting software. For each article, per color and size and based on 200 different input variables (for example: brand, price, online placement, stock situation, weather), an up-to-date forecast is made on a daily basis. This means that OTTO provides 300 million data records to Blue Yonder each week. One of the main impacts of this forecasting system is that overstocks at the end of the season were reduced by 20%. Blue Yonder and OTTO began a common project to identify the “drivers” and reasons for returns, to quantify them, and based on that, to put measures in place to reduce the returns rate. A system was developed to forecast the returns rate of an article at an early time, and to remove high-return, unprofitable articles from the online shop in a fully automated manner, based on an exactly defined system of rules. In total, due to the knowledge generated by the project, approximately two million article returns have been prevented to date. In the case of SportScheck (table 2), their online retail service is an increasingly important part of the business, with 52 million individual visits each year. SportScheck started using predictive analysis from Blue Yonder to improve the accuracy of sale predictions, to allow SportScheck managing its stock levels efficiently and anticipating demand for its individual products. Similarly to Otto’s case,

12

forecasts were improved by between 20% and 40% over the traditional approach with correspondingly high savings in process efficiency and logistics. Data-driven innovation is spreading fast in the retail industry. Big data are also used for example by large food retailers to combine data from weather records with detailed sales data, broken down by store and products, to build computer models that predict future demand for product lines according to weather forecasts. A more accurate picture of demand means the retailer can avoid holding too much stock, or running out of stock all together. As an example, Tesco is running several cost-cutting projects which led to around £6m saves each year (Retail Week, October 2013). 2.5.2

Data-driven marketing and loyalty systems

Innovative data-driven solutions can also substitute more cheaply expensive CRM systems. For example, in the case of Morrison’s, the company used Telefonica’s Smart Steps cloud-based solution to deliver coupons in specific areas, by analysing potential customers’ journeys and their likeliness to walk by a Morrison’s store. This was done by using anonymised and aggregated mobile network data to provide extrapolated trends about footfall by time, gender, and age. The result was a much higher rate of coupons redemption that is a 150% increase in new or reactivated customers visiting Morrisons' stores to spend their coupons, compared to previous campaigns. This was done with a much lower investment than comparable CRM solutions.

“We also needed to avoid effectively wasting budget talking to customers who simply cannot get access to our stores. Instead, we needed to focus our resources on the customers who are in the vicinity or within good transport links. Smart Steps helped us target our investment on our customers better than we have been able to before, really levelling the playing field.” “What Smart Steps offered us was a means to get real customer insight, instead of estimates. Using weighting and with Telefónica Dynamic Insight’s expertise, we could clearly see the best areas to target, and also the areas that we should avoid.” Crawford Davidson Customer Director at Morrisons Supermarkets Extracted from Telefonica Dynamic Insights website Retailers can use Smart Steps to understand footfall across their estate, tailor product promotions in

Table 2 – Summary of impacts of data-driven innovation Function

Retailer example

Effects

Impact

Marketing program: delivery of coupons in selected areas to target customers

Morrison’s

150% increase in new or reactivated customers visiting Morrisons' stores

Increased revenues

Prediction of online sales

SportScheck

20 to 40% of forecasts were closer to actual demand (where compared with traditional forecasts)

Cost savings because of a more efficient management of stock level

13

Demand forecast and data-driven delivery

Otto

Two million article returns prevented

Decrease of return rates by at least 1% or 2%, which represents €10m savings each year

Demand forecasts of food articles (including weather forecasts)

Tesco

Efficiency in the stock management, which also avoid too many discount in case of over stock

£6m saves each year

2.6

Conclusions

The diffusion of data-driven services and processes among European retailers is still at an early stage of development. Our case studies outline both the potential advantages and the transformation power of these new services for the retail industry. It is very much clear that retailers are competing in a consumer-empowered industry and that the only way to keep the pace is to leverage their information assets. The competition is more and more a matter of understanding markets and customers, offering the right article at the right time at the price the consumer is willing to pay. To do so, retailers have to take advantage from the large amount of data and information that is made available in the economic and social system. Companies specialized in data analytics can provide valuable information by unleashing the power of data. This, said an interviewee, is the real challenge of the data emerging market. Retailers and other businesses in general need to be aware of both the power of data and of the large number of data they generate. Every single transaction, event, process, website visit generates data. The business critical data that can make a profitable difference to key decision-making processes are in the businesses’ systems. This is true for large businesses but for small businesses as well. Exploitation of data is not a privilege of the large businesses only. However, the challenge is to make these data available for use, that is to implement the internal “data supply 1 chain” as suggested by Accenture in its last Technology Vision report . On the other hand, thanks to last generation technologies and to cloud delivery, the cost of these datadriven services is affordable by most businesses especially because of “pay-as-you-go” business models, which do not require high investments upfront. But even if the technology tools and the data are available, data-driven innovation would be impossible without skilled data workers and managers able to envisage the potential use of these services. Retailers and businesses in general do not need many data scientists with highly technical and specialist skills, but they need professionals able to understand the market evolution, to identify the potential application of data-driven services and to interpret and use the results for business growth. Big data analytics can provide significant support to improve brand performance, drive customer loyalty, adjust pricing, and improve customer satisfaction. An intensive use of data analytics by retailers assumes that they are willing to know their customers in depth and to adapt to their needs as quickly as possible, while in the past retailers used to launch investments and initiatives hoping their customers would reward them with loyalty. This is certainly a Copernican revolution for the retail industry approach.

1

Technology Vision 2014, Accenture http://www.accenture.com/microsites/it-technology-trends2014/Pages/tech-vision-report.aspx 14

Main Sources http://dynamicinsights.telefonica.com/488/smart-steps http://blog.digital.telefonica.com/?press-release=telefonica-dynamic-insights-launches-smart-steps-in-the-uk http://www.blue-yonder.com/en/

Economist Intelligence Unit, Retail 2022 - How the Economist Intelligence Unit sees the retail landscape changing over the next decade, 2012 IDC Retail Insights, Perspective: Risk, Reward, and Resilience for Retailers and the Role of New Technologies, April 2014 IDC Retail Insights, Business Strategy: A Framework for Mobile Transformation Retail, March 2013 IDC, Advanced and Predictive Analytics Vendor Profile: Blue Yonder, March 2014 IBM Global Business Services Business Analytics and Optimization, Analytics: The real-world use of big data in retail - How innovative retailers extract value from uncertain data, 2013 OECD, Exploring Data-Driven Innovation as a New Source of Growth, 2013 Retail Week, Analysis: how Tesco and Otto are using using data to forecast demand, October 2013

15