Big Data Retail

State Of The Industry Research Series:
Big Data in Retail
SUMMER
12
EKN is part of the Edgell Family
Premium Sponsor
Supporting Sponsor
www.eknresearch.com
About Edgell Knowledge Network

Edgell Knowledge Network (EKN) is a practitioner driven, community reviewed approach to research and community. Our research focuses on delivering practical insights into key areas of interest for the industry, and come infused with pragmatic and actionable recommendations, contributed to and validated by the community.
Research Methodology
The central philosophy of our research is that it is Practitioner Driven. Community Reviewed. Our research agenda is developed using inputs from the end user community and the end user community extensively reviews the research before it is published. This ensures that we inject a healthy dose of pragmatism into the research and recommendations. This includes input of what research topics to pursue, incorporating heavy practitioner input via interviews etc., and ensuring that the bend of research takeaways are oriented towards a real-world, practical application of insights with community sign-off. Turn to page 48 for the detailed methodology.
Advisory Council
The EKN Advisory Council guides the research agenda, process and EKN strategy. The council is also closely involved in reviewing and validating our recommendations. Giri Durbhakula - VP, IT Business Partner for Merchandising, Marketing, Supply Chain and eCommerce, PETCO John Lauderbach - VP IT, Roche Bros Maura Hart - CIO and Group Vice President, Winn Dixie Neil Maizen - EVP & CIO, Canadian Tire Robert Fort - SVP IT, Application Services, Guitar Center Tom Scott COO / CIO, PetCare Rx Yul Vanek - CIO, Deckers Outdoor Corporation
Join the only community that is for retailers, by retailers.
Sign-up at
http://hello.eknresearch.com
In Person
4 Middlebury Blvd Randolph, NJ07869
Electronically
EKNinfo@edgellmail.com www.twitter.com/EKN www.linkedin.com/EKN
On the Phone
973-607-1306
Table of Contents
Report summary 4
What is Big Data?
Impact. Opportunities. Challenges.
10
Research findings
19
Recommendations 34
EKN Big Data Maturity Assessment Framework
38
EKN Big Data Roadmap
44
Research methodology and process
48
Report summary
State of the Industry Research Series: Big Data in Retail

Report summary
Our new Constitution is now established, and has an appearance that promises permanency; but in this world nothing can be said to be certain, except death and taxes. Benjamin Franklin, in a letter to Jean-Baptiste Leroy, 1789
Two centuries on, perhaps it is time to recognize another certainty of life exponential data growth. With the mass adoption of computers and digitization of the enterprise, data is constantly churned out by business processes and captured in company systems. All consumer interactions - online, mobile, and even physical interactions are getting converted into digital signals. Consumerization has unleashed a torrent of user generated and geo-location data that is growing so fast and is so varied that organizations are struggling to keep pace. From the videos of cats playing the piano, to the likes on the videos of cats playing the piano, to the 500,000 tweets sent each day, to the millions of bytes of information that more than 5 billion cellphones produce - this data is growing at a blistering pace. It doesnt fit into an organizations existing databases, and needs more processing power than most retailers possess. This phenomenon of increased data volume, velocity, variety, and the lack of infrastructure to adequately store, manage and analyze the data have led to the advent of Big Data.
Big Data Fast Facts 2.5 exabytes of data (exabyte = 1,000,000,000 GB) created each day; projected to grow at 40% every year. 15 out of 17 industry sectors in the U.S. (such as manufacturing, insurance, healthcare, utilities) have more data stored per company than the entire U.S. Library of Congress. 12TB of Tweets created every day. 2.7 billion average number of likes and comments posted on Facebook. Daily. 90% of the worlds data was created in the last 2 years.
Organizations have to capture, store, decipher and act on these millions of varied signals quickly, in order to stay relevant to their customers. They also have to manage this 40% data growth with a 5% budget increase. While managing this data growth is a big challenge, the opportunity to use this data to impact the business is significantly bigger. This is not news to retailers; they have the unenviable task of staying relevant to the needs and wants of an ever-changing consumer every day. They also have to deliver on this with technology budgets that are about 1.5% of sales, and often flat.
Edgell Knowledge Network

Report summary
Customer insights are the key to being relevant to consumers and Big Data can play a big role in achieving this rich relevance. Having adequate inventory of flare-cut deep-blue jeans before Back-to-School for Akron, OH is critical to Fall sales. This is a detailed exercise in understanding all the historical and current demand and supply signals and finalizing the buys and assortment. Now, modifying the assortment by incorporating new demand signals such as Twitterverse erupting with talk of those cut of Jeans since Justin Bieber started wearing them on his Believe 2012 tour - is critical to increased comp sales and profitability. There is a lot of hype associated with Big Data. The challenge lies in disaggregating the buzz and all the gamechanger talk from where the value lies; and more importantly, in answering the question of how and where do you get started. To help answer the question, the Edgell Knowledge Network (EKN) conducted extensive research using its Pulse360TM methodology. The research included a survey of 75+ retailers, in-person interviews with senior retail leaders, and guidance and validation of the findings and recommendations by our Advisory Council. The research focused on benchmarking the current state of understanding and adoption of Big Data among retailers, and providing pragmatic recommendations on how retailers should approach Big Data. Key findings from the research: 1. High awareness, nascent understanding of Big Data among retailers 2. Managing data variety biggest challenge for retailers with mature analytics capabilities 3. Retailers focusing on building Big Data strategy first 4. Business priorities, budgets and ROI cited as biggest inhibitors to investment 5. Retailers analytics maturity strongly co-related with Big Data adoption 6. Marketing, merchandising, and multi-channel cited as the biggest opportunity areas for Big Data initiatives 7. 50% of retailers have CXO or VP Big Data executive sponsor 8. Diverse and evolving vendor landscape
Research Findings Snapshot 80% of retailers aware of the concept of Big Data, but only 47% clear about its implications to their business 68% of retailers are at a nascent stage of analytics maturity 5% of retailers already have, or are creating, a Big Data strategy. 30% have executed a Big Data project 28% of Big Data initiatives being championed at the CXO level 50% of retailers plan to do their first Big Data project in Marketing or Merchandising 30% of retailers will run out of data storage capacity within the next 2 years

Report summary
Based on the research findings, anecdotal experience captured from retailer and vendor interviews, in addition to guidance from Edgell Knowledge Networks Advisory Council, EKN recommends the following approach towards Big Data: 1. Take time to understand this space. Start by assessing Big Data maturity and do proof of concepts before you invest big. 2. Identify areas of high impact opportunity, and build detailed use cases. Initially focus on 3 areas pricing, segmentation, and marketing effectiveness. 3. Capability building and analytics training are critical to Big Data success. Identify where gaps exist in current capability, build specific recruitment and training plans. 4. Create a comprehensive data strategy covering 3 core areas - customer data/master data management, data policy and process guidelines, and data pool use and sharing 5. Focus on, and plan for organizational change associated with Big Data and analytics adoption. The hype surrounding Big Data can sometimes be distracting, as if serving the needs of a particular type of technology or solution. However, the concept raises some pertinent questions that merit appropriate attention. Depending on how YOU define it, internalize it and implement it, Big Data can be the game-changer it is touted to be. The road to Big Data Nirvana is circuitous and less traveled. The first steps are often the hardest creating internal awareness, identifying high impact decisions that should drive your data strategy, creating analytical capability, demonstrating success, and investing in people, process and technology change. To end how we began, another iconic quote from a great American leader finds new meaning in our hands ask not what Big Data can do for you; ask what you can do with Big Data.
What Is Big Data?

What is Big Data?
The trouble with defining Big Data is that our minds expect clear, precise definitions, while Big Data is a relative concept. It is like being asked, What is fast? The answer to both questions depends on whos doing the defining. Any dogmatic definition of Big Data will be both flawed and incomplete. It helps, however, to have a common understanding of the overall concept, and towards that end, a rudimentary definition is certainly helpful.
Big Data
Collectively refers to the strategy, business processes, tools and technologies that pertain to datasets whose size and complexity is beyond the ability of typical database software tools to capture, store, manage, and analyze.
While this is an acceptable general definition, EKNs position is that there is no absolute specification of what Big Data is; each enterprise must define what it is for them. Following these guiding principles may be more useful than getting caught up in definitions: Decisions: Start your definition with the types of decisions and outcomes Big Data would enable. The purpose of Big Data must be to make sense of and gain knowledge from this data to create competitive advantage; and that must be reflected in its definition. Strategy: Big Data definition must address data strategy issues such as whether you are focused on storage, consumption, or both. Types of data: Typically Big Data strategies would address at least 2 of the 3 types of data structured, unstructured and semi-structured. Size of data sets: No absolute size qualifies, as this depends on how much of what kind of data can you currently store or analyze. Typically ranges from a few dozen terabytes to hundreds of petabytes. Tools and Technologies: While it is very likely that you will require specialized tools in your Big Data initiatives, no specific type of tool or technology needs to be included as part of your Big Data definition, as different tools (including your existing database software) may be appropriate for different circumstances. For instance, If you are able to add structured meta-data describing the unstructured portion of your data, and you want to run standard reports on the structured portion or retrieve individual unstructured elements (such as a single PDF document), then standard databases may do the job.
11

Big Data In Retail

With consumer technology (mobile, social, web, PC) adoption at an all time high the amount of data that is available freely is increasing exponentially. In addition to the consumer data, you have enterprise transaction and exhaust data that retailer systems are constantly producing - POS logs, inventory positions, RFID based product location, truck delivery logs. On its own, this data has little value. However, when combined with other sources it can offer important insights into internal operations and customer behavior. Big Data is going to play a major part in an organizations overall strategy in the coming future. Big Data inaction in the short-term may not be apocalyptic, as all the hype suggests. However, those who continue to ignore building internal consensus over the next 12 months, risk longer-term pain. EKNs interviews and feedback sessions with retailers brought forth a number of issues and challenges that highlight the divided opinion around the promise of Big Data.
Point
Vendors may well be exploiting the term Big Data but I think the basis is very real. Its the consumer. I think there is immense value in pursuing these large Big Data or analytics initiatives Retail CIO (EKN Peer Group Interactions, Apr May 2012)
Counter-Point
No one really knows what Big Data is. It is the latest buzzword vendors are using to sell us stuff. Everything is now Big Data or will now do Big Data. We are not even sure what it is or why we need it. Retail CIO (EKN Peer Group Interactions, Apr May 2012)

12
Opportunities
Opportunities are defined as clear actions or decisions that a retailer can take to impact its business. Capturing new data, such as Facebook likes or location data, or even doing predictive analytics, are not opportunities - they are important enablers. Opportunities must be defined in the context of a business outcome in a specific retail function area or associated business process. EKN sees opportunity identification as a four-step process.
Step 1: Identify an opportunity area based on your organizational goals, time and budget appetite, and ROI expectation.
The Big Data Opportunity-Impact Matrix is an illustrative framework that outlines Big Data opportunities for retailers, taking into account time taken to implement and potential impact of outcome.
Manage Merchandise Performance Improve Customer Insights Manage Sourcing
High
Develop & Sort Assortment Manage Loyalty Program
Manage Pricing Manage Digital Service
Medium
Impact
Manage Producement and Replenishment Execute Outbound Transporation & Logistics
Plan & Execute New Store Opening
Low
Manage Recall & Return to Vendor
Low
Time To Action Medium

Big Data Opportunity-Impact Matrix
High
EKN has mapped several business processes and created use cases, which can serve as a starting point for proof of concepts. This report is focused on providing the framework for building use cases, and a few illustrative examples are given below. Your retail peers will be contributing additional use cases on the members-only EKN portal (To learn more and sign up, go to http://hello.eknresearch.com). The Portal will also provide a Microsoft Excel model of the framework, in which you can plug in your own values to decide which use cases represent the best opportunity for you.
13

Step 2: Define an opportunity and the expected outcome for a business process or functional area
Business process or functional area Expected outcome Key decisions to be Example taken
Pricing
Improve pricing decisions
How can we use social Incorporate Twitter data about our competidata to improve pricing tors prices to price our products decisions?
Marketing
Optimize marketing spend
How can we use dif- Integrate POS data, public data (Census), ferent customer data syndicated data (Nielsen/IRI), CRM or loyalty sources to improve data, along with social data (Facebook, Twitresponse rates of recent campaigns. Categorize response by media form and geography, and then modify spend and outreach for future campaigns to study impact and lift. campaign effectiveness? ter etc). Overlay them on a map to look at

14
Step 3: Detail the various enablers that will be required for you to achieve the expected outcome
Enablers: Every opportunity that leads to an identifiable and finite scope for a proof of concept (POC), pilot, or project, will need to activate a combination of the following enablers. Establish new metrics o How will you measure success? Add and integrate new datasets o Twitter, Facebook, location data Use new analytical tools and techniques o HADOOP, sentiment analysis, genetic algorithms Document business process change o Map out as-is and to-be-determined processes and identify specific areas where new data sets and tools make an impact
15

Step 4: Synthesize your analysis to create a use case in the context of your business
As an illustrative example, let us assume that you are a specialty apparel and accessories retailer that is losing market share to an online only retailer. The following are sample Big Data use cases, across the different business processes EKN evaluated. You can use this framework for any retail format or expected outcome.
Business Process or Re- Use Case tail Functional Area Sales Forecasting Expected Outcome
This holiday season instead of using a subset of the seasonal Optimized inventory sales data to forecast sales you can use your entire sales history, through improved and thereby improve your overall forecast accuracy, and re- forecast duce inventory. By overlaying the sales and store cluster data you can achieve a more accurate and localized store assortment.
Customer Segmentation
You also do a complete analysis of your marketing campaigns, by Increased response product and by location using your entire customer data instead rate of campaigns of a subset. This can help you achieve more detailed customer segmentation and drive new campaigns towards products that add to gross margin, versus relying heavily on off-price promotions to drive sales. . and comp sales
Personalization
Online channels offer a plethora of data, the ability to do deep Increased conversion personalization, and provide very contextual products and of- rates fers. You can combine past purchases, navigation path, wish lists, demographic information and even CRM/Loyalty data to perform real time customer segmentation and personalization. You can also combine social activity (that you have access to) to create a broader set of interests that you might not have deciphered. Sensing the intensity and frequency of social activity recent tweets, Facebook likes - can help you position more seasonal items. In addition to the ability to combine these various data sources, you also have the ability to do real time segmentation to change the assortment for a consumer on the fly. Price elasticity can also be integrated into this process to provide different discount levels to different segments on the same site - Jill gets a 10% coupon, but Jennifer gets 15%.

16
Since base cases are never very exciting, and almost never the norm, let us assume that unplanned events force you and all of your competitors to reassess your assortments.
Unplanned Event 1:
Ms. Katy Perry decides to wear a yellow polka dotted dress on Good Morning America. Within minutes it is the new fashion statement, trending on twitter as the new black. With integration and advanced analytics on Twitter and Facebook (data that you are allowed to access) and feeds from leading fashion sources, you provide your buyers with a real time NYSE type ticker tape of fashion trends across the country. Buyers track these trends and also drill down into them by geography to evaluate what the possible impact to overall assortment could be by store cluster.
Unplanned Event 3:
Following your commendable responses to the two unplanned events, Wal-Mart and Amazon realize they are losing sales to you this season and bring out the ultimate weapon pricing. They drop the price on a national jeans label by 15%, and wait for the sales to come in. With the ability to do more accurate and real time price elasticity calculations, you calculate the overall impact of this price drop both from losing the sale of the pair of jeans or losing the entire basket of that shopping trip. Based on the analysis, you determine the amount of discount you want to give and campaigns you can run to offset the sales and margin impact.
17

Unplanned Event 2:
The schedule for the local basketball varsity team gets pushed up by 4 weeks because of construction, which throws the careful plans of the parents into disarray. Unlike the previous example, this isnt a national demand signal. Local demand signals are difficult to spot and react to quickly. In a typical situation, this would lead to stock-outs of varsity t-shirts and other related items. Through localized news and social data collation and analysis, you create a customized dashboard of whats new in the community. The Local Times lets you react to local trends by changing assortment, sending an input early enough to get the required stock; by delivering the appropriate marketing messaging at the right time; and by improving customer engagement through better knowledge of current events impacting local trends.
Unplanned Event 4:
Inclement weather forces a truck to get delayed. Using geospatial tracking data, analysis of trucker blogs, audio from your network, and traffic data, you are able to monitor and track the progress of the truck and are able to react quickly and guide it through the weather safely. Based on the delay and route changes, the delivery time for the goods is automatically updated. In the case of severe delays alternative fulfillment centers can be activated

18
Challenges
With great power comes great responsibility; and with big opportunity come big challenges. Integrating existing and new datasets Bar a few, most retailers havent invested in creating integrated data sets. Key data elements like customer, item, and pricing reside across transaction databases and warehouses. In addition, the datasets have different structures and are built on different technology solutions. The fact that most of the new data sources that retailers want to tap are torrents of unstructured data, adds to the complexity and magnitude of this challenge. Improving data quality Garbage-in, even-more-garbage-out. Retailers have struggled to get to the proverbial single version of the truth, from sales to stock positions. The challenge is that a multitude of data warehouses collect similar data, from different sources, with different periodicity. In addition to this, the new external datasets that retailers hope to tap also need to be scrubbed and baselined before they can be integrated. Significant investments with difficult to measure ROI Retailers justifiably tend to protect their thin margins, and Big Data infrastructure doesnt come cheap. Investments would be required in hardware, tools and training; and the business case definition is not easy. The return on a Big Data program, like a BI/analytics program is difficult to measure with hard numbers, as the program runs across functional areas and touches various P&Ls. Lack of internal capabilities to analyze data There is an acute shortage of resources that can work well with large quantities of data and analyze it. Resources that can actually interpret the analysis are even scarcer.
19
Research findings

Research findings
20
High awareness, nascent understanding of Big Data among retailers

Figure 1: Retailer awareness about Big Data
11.7%
16.9% Not aware

35.1% 16.9%
26.0% Aware of the concept 10.4% Well aware of the concept, but not sure of its implications for retail 35.1% Well aware of the concept, and its implications for retail
10.4%
26.0%
11.7% Big Data evangelist
Figure 2: Enterprise data currently stored by retailers

1.6% 8.2%
32.8% Dont know

32.8% 24.6%
32.8% Less than 50 TB 24.6% 51 TB - 500 TB 1.6% 500 TB 999 TB 8.2% More than 1 petabyte
32.8%
Figure 3: Unstructured data currently stored by retailers
24.6%
75.4% Dont know 24.6% Know the % of unstructured data
75.4%
21

Research findings
High awareness, nascent understanding of Big Data among retailers

While 8 out of 10 retailers responded they were aware of Big Data, the lack of a standard, concrete definition has made it difficult for retailers to understand what it really means, and how it is significantly different from what they are or have been doing (in data warehousing and analytics). More than half of the surveyed retailers were either unaware of the concept, or unclear of its implications to retail. Retailers awareness about the data they currently store is also low 1 in 3 respondents didnt know how much data they store. Lower still is their awareness about the amount of unstructured data they store. Unstructured data is the centerpiece of the Big Data challenge. It is all the data that doesnt fit neatly into our existing data models and databases; and it is huge. The text (email, logs, customer reviews, tweets, etc...), images, audio, video, location data that we generate at work and in our personal lives is all unstructured data. Only 1 in 4 surveyed retailers responded they know the amount of unstructured data they store as a % of total data. Effectively gathering and using unstructured data represents perhaps the greatest challenge, as well as opportunity, when dealing with Big Data. If not managed properly, it is worthless; but, it can be used to get insights into customer behavior, create a 360-degree profile of the consumer and improve operational efficiency.
Takeaway: Retailers need to improve their understanding of Big Data by baselining their current state of data and analytics maturity, and reviewing initiatives taken by leading and competing retailers.

Research findings
22
Managing data variety biggest challenge for retailers with mature analytics capabilities
Figure 4: Biggest challenge in managing data
50 46%
40 34% 30 20%
46% Handling data volume 34% Handling data variety 20% Handling data velocity
20
10
Figure 5: Retailers estimate of data growth (per year)

100 90 80 70 60 50 40 30 20 10 0 Overall Enterprice Data Structured Data Unstructured Data
5%
3%
5%
28% 34%
16%
More than 60%

20% 30%
Between 40% - 59% Between 20% - 39% Less than 19% I dont know
20%
23

Research findings
Managing data variety biggest challenge for retailers with mature analytics capabilities
The challenges of managing data can be looked at across three dimensions. Volume The amount of data generated Velocity The frequency with which data is generated and captured. Variety The different types of data generated In the EKN survey, retailers responded that volume of data is the major concern. They expect their data volume to grow between 20%-40% each year, and almost 50% of surveyed retailers cited that their data volume is doubling every 5 years. A third of surveyed retailers arent sure of how much data they have, or at what pace it is growing. This reemphasizes the recommendation to conduct a baseline of your current state. However, handling large data sets is not a new problem for retailers. Since the advent of the barcode, retailers have been challenged with managing large amounts of data and trying to create value from it. For years, the high cost of storage and data processing made handling even what may be considered as small pieces of data prohibitively expensive. Today, cheap storage, processing power, and new technologies like HADOOP have made it exponentially cheaper to handle larger data volume. So, EKN decided to dig a bit deeper. What did retailers with mature analytics capabilities, and retailers with data growth over 40%, indicate as their biggest challenge in managing data? Data variety. Both segments responded that handing the variety of data - structured, unstructured and semi-structured - is their biggest challenge.
Takeaway: Data growth is a challenge that retailers have faced before, and they have a fair idea about how to handle it. However, being asked to manage and make sense of the torrent of unstructured data is a bigger challenge for them. They need to evaluate their existing tools and their ability to manage unstructured data. They need to work on enterprise taxonomy to help them make sense of all this data.

Research findings
24
Retailers focusing on building Big Data strategy first

Figure 6: Created a Big Data strategy Figure 7: Executed a Big Data proof of concept (POC)
9% 33% 43% 13%
Yes In progress No
78%
Yes In progress No
24%
Figure 8: Executed a Big Data project
Figure 9: Budgeting for Big Data

1.3% 3.9%
23%
10.4%
40.3% I dont know

40.3%
Yes
7%
23.4% Already have a Big Data budget 20.8% Within the next 2 years 10.4% In the next 2-4 years 3.9% Beyond 4 years 1.3% Never
In progress No
20.8%
70%
23.4%
25

Research findings
Retailers focusing on building Big Data strategy first

Though only 30% of retailers have done a proof of concept (POC) or project in Big Data, close to 60% have, or are building, a Big Data strategy. It points to the fact that retailers are willing to take the time and make the effort to build a more considered approach. 24% of retailers currently have a budget for Big Data projects, and another 20% are planning to budget for them in the next 2 years. EKN expects Big Data projects to increase in the next year, with close to half of the retailers doing a project or proof of concept.
Takeaway: Retailers are taking their time to think about Big Data, and are working on building a strategy to approach it the right way. Experimenting with small proof of concepts (POC) to validate some key use cases will be a good addition to the strategy. EKN also encourages retailers to actively reach out to other retailers that have had some success in this space.

Research findings
26
Business priorities, budgets and ROI cited as biggest inhibitors to investment

Figure 10: Important elements in building Big Data capabilities
17%
Outsourced Analytics Services

27% 0% 9%
48% 27%
Internal Team Augmentation & Training (Analytics)
Unimportant
34% 56%
Somewhat important Important
1%
New Tools/Software
10% 36% 52% 10%
Most important
New Hardware
21%
29% 40%
0%
10%
20%
30%
40%
50%
60%
Figure 11: Challenges in starting Big Data initiatives

13%
No resources (People)
9%
17% 30% 28% 19% 28% 35%
41%
No Tools No Budget No Internal Ownership Champion Not A Business Priority

6% 7%
44%
Not a challenge Minimal challenge Signicant challenge

35%
9%
30% 30% 26%
9%
Biggest challenge
28% 31% 31% 17% 31% 46%
Unclear Benets/ROI 0% 10%
20%
30%
40%
50%
27

Research findings
Business priorities, budgets and ROI cited as biggest inhibitors to investment

Providing a watertight model for analytics ROI is inherently difficult. Unsurprisingly then, 31% of surveyed retailers responded that yet another analytics project will need a solid benefits argument. Importantly however, respondents cited Big Data not being a business priority and lack of budget as two equally important challenges. Retailers responded that new analytics tools or software, and augmenting and training internal analytics resources are the most important factors towards building big data capabilities. This speaks to an interesting realization by retailers that tools, while critical to Big Data, are only as important as the organizations capability to analyze and understand data. Retailers can have the fastest servers and the prettiest visualization in town, but if they dont have people who can make sense of the information or business processes that enable the decisions they want to make, they are no better off than before; except they may be out a few million dollars.
Takeaway (Retailer): In order to get management buy in, internal efforts are required to raise awareness with the different stakeholders around what Big Data is and how it impacts the different business units merchandising, marketing, etc. This needs to be followed by collaborative efforts to create metrics and ROI measures for Big Data. Retailers need to accelerate their efforts in analytics training; EKN recommends that a portion of the HR Training budget be set-aside for this in the next 6-9 months, and an executive program be put together for the leadership team. The creation of an analytics services team or Center Of Excellence needs to be evaluated.

Research findings
28
Retailers analytics maturity strongly co-related with Big Data adoption

Figure 12: Analytics maturity of retailers
14.3%
15.6%
15.6% We do reporting
18.2%
51.9% We do basic analytics 18.2% We do investigative analytics 14.3% We do predictive analytics

51.9%
Figure 13: Big Data awareness of retailers with Big Data initiatives
4% 23% 27%
4% Data evangelist 27% Aware of concept and implication 11% Aware of concept, but not implication 35% Conceptually aware
35%
11%
23% Not aware
29

Research findings
Retailers analytics maturity strongly co-related with Big Data adoption

Overall analytics maturity among retailers is low, with only a third of them conducting advanced analytics of any kind. The majority of retailers are still engaged in basic data reporting and analytics. This is important to note, as the promise of Big Data is delivered through high-end analytics. If the organization and the supporting infrastructure arent ready to handle the mining and interpretation of the data, then the promise will remain unfulfilled and the use case will not be realized. As a follow up to the above point, EKN analyzed the correlation between an organizations analytical capabilities and its awareness of, and existing efforts in, Big Data. EKNs analysis uncovered that the higher the analytical maturity, the more likely it is for a retailer to have existing Big Data efforts in place. Retailers that have already built capabilities and infrastructure to support deep analytics are now exploring the Big Data space. It therefore comes as no surprise, that some of the largest retailers globally are moving towards building or acquiring capabilities and platforms to help them conduct Big Data analytics. Companies like Wal-Mart, Target, and Tesco are marching ahead with efforts that are focused on creating a 360-degree consumer profile, and have the ability to conduct predictive analytics. Wal-Mart has been an early adopter of HADOOP, and is moving to a 250-node HADOOP cluster. The Wal-Mart Labs division has also come up with tools like ShopyCat that help recommend products to people by analyzing their social activity on Facebook. The fact that Target knows that you are pregnant even before your father does, is one of the astonishing examples of how Big Data can be leveraged. A summary of Big Data initiatives by leading global retailers will be available for download exclusively for EKN Portal members.
Takeaway: It is clear that the top retailers are already investigating this space, and it is a matter of time before they find Data Gold. Leading retailers have already put, or are putting in place, foundational platform and processes to improve their overall analytics infrastructure. Retailers that are currently behind should take steps to catch up. One option, beyond the traditional platforms and vendors, is to actively look at the new SaaS - based platforms. These can often provide a faster, lowercost option.

Research findings
30
Marketing, merchandising, and multi-channel cited as the biggest opportunity areas for Big Data initiatives
Figure 14: Functional areas that will benefit most from Big Data initiatives in retail
70 62% 60 50 40 30 20 10 0 25% 29% 60%
25% Store operations

44%
62% Merchandising 60% Marketing (including mobile, social) 29% Supply chain
14%
14% Operations (i.e. nance, IT, HR, etc.) 44% eCommerce or multi-channel
Figure 15: Retail functional areas experiencing the most data growth
35 30 25 20 15 10 5 0 13% 10% 33% 26% 33%
11% Store operations 26% Merchandising 33% Marketing (including mobile, social)
11%
13% Supply chain 10% Operations (i.e. nance, IT, HR, etc.) 33% eCommerce or multi-channel
Figure 16: Functional area of first planned Big Data project

3.7% 13.0% 20.4%
13.0% Store operations 20.4% Merchandising

20.4%
29.6% Marketing (including mobile, social) 9.3% Supply chain 3.7% Operations (i.e. nance, IT, HR, etc.)
3.7% 9.3%
20.4% eCommerce or multi-channel
29.6%
31

Research findings
Marketing, merchandising, and multi-channel cited as the biggest opportunity areas for Big Data initiatives
Marketing, merchandising and multi-channel are the 3 areas where retailers are experiencing the most data growth, and correspondingly see the biggest opportunity for doing their first project. 50% of surveyed retailers plan to do their first Big Data project in the areas of marketing or merchandising. Based on retailer feedback, EKN divided the functional areas where retailers perceive opportunities into 3 bands. Band 1 (Maximum Benefit): Marketing and Merchandising (60%+) Band 2 (Medium Benefit): Multi-Channel (40%+) Band 3 (Low Benefit): All Other Others (Below 39%) The Opportunities section of this report provides detailed use cases and examples of how Big Data can be leveraged in these, and other, Retail functional areas.
Takeaway: Retailers need to baseline their data (and its growth) in the 3 top areas, and assess the amount of unstructured data they have. While it is neither top of mind nor sexy, retailers shouldnt ignore the impact of Big Data on the supply chain and should explore use cases in that area.

Research findings
32
50% of retailers have CXO or VP Big Data executive sponsor

Figure 17: Executive sponsorship level of Big Data initiatives
14.8%
16.7%
9.3%
16.7% Not aware 27.8% CXO level 22.2% VP level 9.3% Director level
9.3% 27.8%
9.3% Manager level 14.8% No champion
22.2%
With 28% of Big Data initiatives being championed at the CXO level, and a further 22% at the VP level, retailers realize the importance of executive sponsorship for Big Data success. Cross-tabbing executive sponsorship responses with Big Data analytics maturity and ongoing Big Data initiatives data points, EKN found strong correlation between executive involvement and Big Data and analytics adoption. EKN interviews with retail executives validated the above finding. They mentioned, that for such strategic projects that are difficult to measure, executive support is critical for initial alignment, as well as long-term success. With Big Data getting coverage in the mainstream business media, a Big Data POC may very well be driven top down by a CXO. They also validated the need to have cross-functional teams and support from both business and IT leaders. Takeaway: To succeed in Big Data retailers need a Big Executive personally driving the strategy and organizational communication. Without that sponsorship it will be difficult for these initiatives to achieve enterprise wide adoption.
33

Research findings
Diverse and evolving vendor landscape

In EKNs survey, retailers were asked to name (unaided recall) the 3 top Big Data software vendors according to them. EKN found a clear group of leaders, followed by a long tail comprising of ERP, BI and Niche vendors. Clear leaders in retailer perception: 1. IBM (25%) 2. Oracle (18%) This is a new and evolving area with a lot of start-ups vying to be the next-big-thing, and the bigger, established vendors investing in branding and marketing to mark off their territory. Among the established players, only IBM and Oracle had more than 10% of recall. IBM has done a particularly admirable job of creating awareness, content, and generating visibility about Big Data. IBM Watson was mentioned a few times by some of the retailers EKN interviewed as an example of using high-end analytics. Oracles efforts to increase its presence in BI and Analytics seem to be working; as close to 20% of retailers mentioned it as a leading vendor for Big Data. Among niche vendors, most of the up and coming vendors were mentioned (Cloudera, Vertica, Greenplum (EMC), Splunk), but no one really stood out. In its secondary research, EKN found vendors contributing actively to thought leadership in Big Data. However, most vendors content focused on establishing why Big Data is important. There is an opportunity for vendors to clarify their Big Data message, especially as relates to what services or solutions they provide, their implementation approach and engagement models. A detailed Big Data vendor landscape is available on the EKN Portal for members.
Takeaway: Retailers need to develop a clearer understanding of the Big Data vendor landscape, learn more about how other retailers are using vendors and dig deeper into what implementation and execution challenges exist. They should start close to home, and ask their existing vendors to update them on their solutions in this space and the work that they are doing for other clients.
Recommendations
35

Recommendations
Start by assessing Big Data maturity, and do proof of concepts (POC) before you invest big
Even though Big Data as a concept is fairly new, and its definition nebulous, it has high potential for the entire enterprise. Data is being generated at an exponential rate; so, retailers ability to gather, synthesize, analyze and act on this data will be critical going forward. This is an evolving space, and other than the universally mentioned Amazon.com, no retailer has an unassailable advantage. Wait and see can be an effective strategy, only if combined with an aggressive ready to act orientation. Taking a few months to better understand Big Data and how your company can utilize it would likely be better than jumping into a pilot without a clear objective. Given that this is an evolving field with no clear and proven path to success, EKN suggests retailers start with proof of concepts (POC) and adopt the Fail Fast, Fail Cheap philosophy. The experience of the POC or pilot programs will help provide the right kind of grounding required to prove value on which future investment decision can based. However, a few of the top 50 retailers may be better positioned to take on larger Big Data initiatives as their first step. Even in this case, EKN recommends breaking up that larger vision into smaller, pilot or POC sized sprints. Not only will this help reduce complexity by distributing it over a number of smaller scopes, it will also ensure a deeper inspection and exploration of the larger initiative, its objectives and expected benefits. The EKN Roadmap on how retailers should approach Big Data is provided in the next section. Get Started: Use the EKN Big Data Maturity Assessment Framework, and follow Sprint 1 of the EKN Big Data Roadmap to understand and baseline your preparedness for Big Data.

Recommendations
36
Identify areas of high impact opportunity, and build detailed use cases
It is not about what data you have; rather, about what decisions you are trying to make, and how you can improve them. Towards this end, use cases of opportunity areas need to be defined in terms of objectives and outcomes you want to see achieved. Business and technology teams should work together to create a master list of such use cases, and use that as a starting point to identify potential proof of concepts: While enterprise-wide relevance and applicability is one of Big Datas core appeals, it also represents one of its key challenges i.e. the Big Data rhetoric doesnt usually define specific boundaries within which organizations can realize value. This boundary-less nature of the Big Data conversation is a challenge for those looking for quick wins and measurable ROI. EKN recommends retailers focus on 3 measurable problems to assess the benefit of Big Data in their organization: 1. Optimizing pricing 2. Improving micro-segmentation 3. Increasing marketing effectiveness. Get Started: Follow the 4 step EKN opportunity identification process in the Opportunity section of this report to identify and detail the use cases for a proof of concept.
Prioritize resourcing and training

You have all the data that you want and the latest tools in the market. Now what? Without the right analytical resources and training, the data and tools have little value. Invest in capability building across the board, from executive training to building a strong analytics team. Retailers need to accelerate their efforts in analytics training; EKN recommends that a portion of the HR Training budget needs to be set aside for this from 2012 itself. An executive program needs to be put together for the leadership team. The creation of an analytics services unit/Center of Excellence needs to be evaluated. Get Started: Assess the current analytics capability and resource strength in the organization. Build a resourcing and training plan based on the assessment. Create an executive program on business analytics in retail for the mid to senior executive levels.
37

Recommendations
Create a comprehensive data strategy covering 3 core areas

1. Customer Data/Master Data Management 2. Data Policy and Process Guidelines 3. Data Pool Use and Sharing (i.e. collaboration with manufacturers) The move towards understanding and harnessing Big Data will raise a lot of long buried questions around MDM, data privacy and collaboration. Fundamental strategy decisions will need to be taken in these areas, and these decisions will not go away. Data policy and processes regarding the collection, storage and use of consumer and social data are specific areas requiring attention. Global retailers need to be careful here because of the recent spate of legislation in Europe and US. Get Started: Start by making an operational plan to detail the capture, storage and consumption of the data that you plan to integrate for your POCs. Build detailed data policy and process guidelines.
Focus on, and plan for, organizational change associated with analytics adoption
The rise of analytics as a discipline will lead to structural organization changes. Analytics growing importance in retail is illustrated by the amount of investment that is being pumped in, and by the creation of senior leadership positions like Chief Insights or Chief Data Officer. However, even progressive retailers, that have such titles, struggle to answer the question of who owns analytics or insights, and how best can analytics fit into the current organization structure. As retailers augment their analytics capabilities and teams, EKN sees analytics being carved out as a separate department, or Business Insights Unit, that would lie at the intersection of Business and IT. Typically these teams sit in different departments across business and IT, but EKN sees the emergence of a shared analytics services center in the future. Get Started: Call out organization change as a focus area when you create your Big Data strategy. Initiate a dialogue with key stakeholders on the organization structure required to support your Big Data and analytics strategy.
38
39

The following factors, which emerged from EKNs research, suggest that retailers would benefit from being able to benchmark their Big Data maturity, and subsequently build a specific action plan that will help them move along the maturity curve. 1. Low awareness about what Big Data is and its impact on their business 2. Strong correlation between a retailers analytics maturity and readiness or likelihood of a Big Data initiative 3. Lack of clear ROI benefits from a Big Data initiative 4. No clear roadmap to Big Data success Towards this end, EKN has developed a model to assess the relative maturity of Big Data awareness and capabilities of an organization. Assessment Areas Strategic Intent (SI) Organization Readiness (OR) Opportunity Qualification (OQ) Investment Appetite (IA)
Definition of each stage:

Strategic Investment (SI) Maturity Level 1 Beginner Maturity Level 2 Tinkerer Maturity Level 3 Leader Maturity Level 4 Guru Low Medium High High Organization Readiness (OR) Low Low, Medium Medium High Oppotunity Qualification (OQ) Low Low Medium High Investment Appetite (IA) Low Low, Medium Medium High

40
Stage 1: Beginner
Strategic Intent: Low Organization Readiness: Low Opportunity Qualification: Low Investment Appetite: Low
Big Data Adoption:

There are no big data initiatives currently underway at the organization.
Characteristics:
There is low awareness of Big Data, including understanding its implications on their organization. It is very difficult to start a Big Data project, due to lack of awareness and an unwillingness to invest. There is no Big Data expert in the organization. Since the management and employees have low awareness of Big Data, there is very little chance of a Big Data initiative being undertaken by the organization in the short term.
Analytics Maturity:
Low. These organizations typically do not perform any analytics on data and have a rudimentary reporting system (excel or worksheet based), where data is extracted from relevant sources, and reports generated when required. Data is not managed or analyzed to provide further insights into issues.
41

Stage 2: Tinkerer
Strategic Intent: Medium Organization Readiness: Low, Medium Opportunity Qualification: Low Investment Appetite: Low, Medium
Big Data Adoption:

The organization is actively researching this space and has started doing proof of concepts (POCs).
Characteristics:
There is awareness and curiosity about Big Data, but the management is not convinced about Big Datas value. Limited budgets are available for analytics projects, and it is difficult to get funding approved for Big Data. The organization is investing or has invested a minimal amount in understanding Big Data better use cases, competitive moves and applicability to their format. The initiative is driven at the Middle to Senior Executive level (Director, Senior Director, or VP). The organization is on the cusp of starting to build a strategy or executing small proof of concepts to validate the use case and ROI. This is the longest maturity phase and firms at this level will execute proof of concepts within 6 to 9 months and decide on a longer-term Big Data strategy within the next 12 to 18 months.
Analytics Maturity:
The firm stores a huge amount of data, but performs basic analytics. Big Data analytics are championed at the Manager or Senior Manager level, and operated as silos within a particular function; there is limited top management involvement. Parts of the organization are pushing the existing technology and solutions to the brink, however there is no corporate or top driven transformation program for them to hop on to.

42
Stage 3: Leader
Strategic Intent: High Organization Readiness: Medium Opportunity Qualification: Medium Investment Appetite: Medium
Big Data Adoption:

The organization has executed proof of concepts, and captured the lessons from the experience. They have built a Big Data strategy and identified key focus areas for projects.
Awareness:
The CIO, CTO or CMO is the Big Data champion in the organization. The organization may still struggle with Big Data being looked at as an IT initiative, however there is generally widespread awareness of the concept among key stakeholders.
Analytics Maturity:
Medium to high. The firms analytics strategies include a wider and more complex set of data sources than a Tinkerers, typically integrating all 3 types of data (structured, unstructured, semi-structured). They have the ability to perform investigative analytics and determine patterns from data that can help drive better business decisions. The Big Data pilot or proof of concept included an upgrade of the organizations Big Data infrastructure including software, hardware and tools.
43

Stage 4: Guru
Strategic Intent: High Organization Readiness: High Opportunity Qualification: High Investment Appetite: High
Big Data Adoption:

A Big Data initiative with multi-function impact has been implemented and delivered measurable benefits. There is a separate Big Data division, a strong bench of data scientists, and best-of-breed Big Data technology tightly integrated with enterprise systems. The firm is positioned to provide strategic advice to other firms in the industry on big data initiatives.
Awareness:
An orientation towards Big Data is integrated within the firms strategy, being driven into the organization from the CEO down. The management team is aware of and aligned with the need to leverage Big Data for competitive advantage. Big Data spending is planned for at the beginning of the year, as part of the business planning process.
Analytics Maturity:
High. The organization has a demonstrable track record of successfully leveraging Big Data to influence or deliver positive business outcomes. The organization is capable of conducting investigative and predictive analytics.
44
45

The EKN Big Data Roadmap is built keeping in mind how retailers can graduate up the maturity ladder by focusing on the four levers of Strategic Intent, Organization Readiness, Opportunity Qualification and Investment Appetite. This is not a one-size-fits-all recommendation, but is meant as a starting point for retailers that benchmark themselves as Beginners or Tinkerers.
Sprint 1: Months 0 3
Focus: Understand and baseline your preparedness Expected Outcome: A. Baseline your data How much data you currently have How is the data growing o Types of data: structured, unstructured, semi-structured o Domain Areas At what rate is that data growing Document your data governance and privacy policy Call out social media and consumer location data o How much do you have o Who owns it (for e.g. Retailer, marketing agency, user) o What are the governance and privacy requirements B. Understand the landscape What is your immediate competitive pool doing What are the top 5 pure-play ecommerce retailers doing Create a list of use cases based on the above analysis C. Baseline your current analytics capabilities How many resources do you currently have Which department are they currently in
Execution Model: For this initial discovery and benchmarking phase, EKN recommends you rely more heavily on internal resources. There are many consultants who can help, however for this particular phase, EKN recommends engaging them only if available bandwidth and opportunity cost are significant barriers to forming a crossfunctional internal team. This phase will help you get a stronger hold of your environment, and be better placed to direct and manage consultants in future phases.

46
Ideally, a lean, 3-member, cross-functional team from IT and Business, working part-time on this project, should drive this. If resourcing is a challenge at this stage, try to have at least a single dedicated and experienced resource that you can support with part-time analysts. As established previously, Big Data success requires senior level buy-in from the get go. Ensure you have a senior (reports directly to CEO) executive sponsor identified during this stage.
Focus: Get buy-in and pick a few areas in which to do a proof of concept (POC) A. Communicate and generate buy-in Publish your findings to key stakeholders Create a CXO note on Big Data Get your business partner involved Arrange a joint business-IT workshop Based on business buy-in, explore a small POC Create a working group on Big Data B. Crystallize POC ideas Focus on these areas o Price improvement o Micro segmentation o Marketing effectiveness Note: These are analysis areas. Data inputs like Facebook, Twitter, and geo-location will feed into these and should be tapped. Multi-channel can be covered across these 3 analysis areas. Evaluate solution and service providers C. Identify capability and training requirements Identify skills and capabilities required for analytics Create competence levels for analytics
A. Execute POC Finalize vendor partnerships Set clear measures for what a win is Execute the most promising POC and use cases Constantly communicate progress to Business and IT Leadership Create a ROI measurement framework for the POC
47

B. Assess current capability levels and conduct executive training Conduct an assessment center on analytics for the current resources Benchmark all current resources on the competence scale Create a training plan on business analytics for senior executives Conduct executive training on business analytics for senior executives
A. Incorporate feedback and decide strategy Publish performance of POC to key stakeholders Organize a formal feedback session with stakeholders and vendors Using all the data collected, and based on the experience, do a visioning session Conduct a resource and training forecasting exercise Initiate stakeholder discussion on organization structure Take a call on whether any Big Data program is worth focusing on Detail a single pilot program that you would like to focus on Do formal budgeting for a Big Data program Initiate partner selection process Execution Model: EKNs recommendation on reliance on internal resources from Sprint 1 applies to Sprint 2 as well. Look at partnering with consultants more actively as you move towards defining the areas for a POC. For training and capability building, explore partnerships with niche firms or universities. For consulting and technology, a partner that is strong in the corresponding domain and has strong analytical capabilities would be a good fit. As you move to the POC and execution stage, start inviting and testing out vendors. While enterprise vendors seem easy to work with, EKN suggests taking a good look at some of the niche vendors. The Big Data Vendor Landscape on the EKN Portal is a good place to start. What Next: Decision on go / no-go for Big Data If it is a go, start working on a comprehensive strategy Continue to execute on the pilot
48
49

The central philosophy of our research is that it is Practitioner Driven. Community Reviewed. EKNs research agenda is developed using inputs from the end user community and the end user community extensively reviews the research before it is published. This ensures that we inject a healthy dose of pragmatism into the research and recommendations. This includes input of what research topics to pursue, incorporating heavy practitioner input via interviews etc., and ensuring that the bend of research takeaways are oriented towards a real-world, practical application of insights with community sign-off.
Identify: Practitioner Driven

Research topic and focus. Decided based on advisory and industry feedback
Gather: Pulse 360TM

Quantitative and qualitative research conducted across relevant stakeholders for 360 degree view Primary Research (i.e. Web Surveys, Tele-Surveys, Interviews, Workshops/ Focus Groups) Secondary Research (i.e. Galaxy, Reseach subscriptions, public data) Signal Repository (i.e. Social Sentiment - blogs, twitter, user groups, community)
Analyze:
Leverage frameworks, expertise & experience to analyze the data and develop context and options
Synthesize:
Incorporate all analysis into a framework that provides a clear understanding of the topic.
Review: Practitioner Reviewed

Apply real world pragmatism and get quantative feedback from the practitioner community
Recommend
Exercise judgment to create final set of recommendations and roadmaps
Our research methodology lends from our fundamental philosophy and recognizes that an enlightened opinion though important cannot be the primary basis for a recommendation

50
The Pulse 360TM Methodology: Incorporates a mix of quantitative research, qualitative research and experience across stakeholders for a 360 degree view. Key data points for the EKN Big Data State of the Industry report: Methodology: Pulse 360TM Primary Research: o 75 retail executives from North America surveyed o 3 senior retail technology executives interviewed at length CIO interviews: 2 Director interviews: 1 Extensive Secondary Research: o Analysis of top 100 retailers o Analysis of Big Data Vendors o Analysis of Big Data news and literature Peer Review & Feedback Loop o Closed door feedback from 10 retail CIOs o Detailed feedback on the entire report from 4 retail CIOs
Premium Sponsor
About Microstrategy
MicroStrategy is a global leader in business intelligence (BI), enabling organizations worldwide to analyze the vast amounts of data stored across their enterprises to make better business decisions. The MicroStrategy platform delivers actionable information to business users via the web and mobile devices, including the iPad, iPhone, BlackBerry, and Android. Top global retailers use MicroStrategy for developing a 360 view of the customer (including merging existing CRM systems with Social Intelligence data), improving operating efficiency and the customer shopping experience through the use of mobile store operations BI, improving sales and decreasing costs by reducing stock-outs and overstocks, and enabling rapid decisions and immediate action anytime, anywhere. Experience the power of MicroStrategy Clouds data visualization, deliver a mobile application in 10 days with a Mobile Quickstrike program, and download free reporting software by visiting www.microstrategy.com. Email us at info@microstrategy.com
Supporting Sponsor
About Tata Consultancy Services

Tata Consultancy Services is an IT services, consulting and business solutions organization that delivers real results to global business, ensuring a level of certainty no other firm can match. TCS offers a consulting-led, integrated portfolio of IT, BPO, infrastructure, engineering and assurance services. This is delivered through its unique Global Network Delivery Model, recognized as the benchmark of excellence in software development. A part of the Tata group, Indias largest industrial conglomerate, TCS has over 243,000 of the worlds best-trained consultants in 42 countries. For more information, visit us at www.tcs.com/retail. Email us at retail.solutions@tcs.com
About Edgell Knowledge Network

Our research agenda is developed using inputs from the end user community and the end user community extensively reviews the research before it is published. This ensures that we inject a healthy dose of pragmatism into the research and recommendations. This includes input of what research topics to pursue, incorporating heavy practitioner input via interviews etc., and ensuring that the bend of research takeaways are oriented towards a real-world, practical application of insights with community sign-off. For more information, visit www.eknresearch.com Email us at EKNinfo@edgellmail.com
EKN is part of the Edgell Family
Disclaimer: EKN does not make any warranties, express or implied, including, without limitation, those of merchantability and fitness for a particular purpose. The information and opinions in research reports constitute judgments as at the date indicated and are subject to change without notice. The information provided is not intended as financial or investment advice and should not be relied upon as such. The information is not a substitute for independent professional advice before making any investment decisions.
Copyright 2012 Edgell Knowledge Network Registered Office: 4 Middlebury Blvd. Randolph, NJ 07869 Ph: (973) 607 1300

Big Data Retail

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Big Data Retail

Uploaded by

Copyright:

Available Formats

State Of The Industry Research Series:

Big Data in Retail

EKN is part of the Edgell Family

About Edgell Knowledge Network

Join the only community that is for retailers, by retailers.

What is Big Data?

Impact. Opportunities. Challenges.

EKN Big Data Maturity Assessment Framework

EKN Big Data Roadmap

Research methodology and process

State of the Industry Research Series: Big Data in Retail

Edgell Knowledge Network

State of the Industry Research Series: Big Data in Retail

What Is Big Data?

State of the Industry Research Series: Big Data in Retail

Impact. Opportunities. Challenges.

State of the Industry Research Series: Big Data in Retail

Big Data In Retail

Edgell Knowledge Network

Develop & Sort Assortment Manage Loyalty Program

Manage Pricing Manage Digital Service

Manage Producement and Replenishment Execute Outbound Transporation & Logistics

Plan & Execute New Store Opening

Manage Recall & Return to Vendor

Time To Action Medium

State of the Industry Research Series: Big Data in Retail

Improve pricing decisions

Optimize marketing spend

Edgell Knowledge Network

State of the Industry Research Series: Big Data in Retail

Edgell Knowledge Network

State of the Industry Research Series: Big Data in Retail

Edgell Knowledge Network

State of the Industry Research Series: Big Data in Retail

Edgell Knowledge Network

High awareness, nascent understanding of Big Data among retailers

16.9% Not aware

11.7% Big Data evangelist

Figure 2: Enterprise data currently stored by retailers

32.8% Dont know

Figure 3: Unstructured data currently stored by retailers

75.4% Dont know 24.6% Know the % of unstructured data

State of the Industry Research Series: Big Data in Retail

High awareness, nascent understanding of Big Data among retailers

Edgell Knowledge Network

Figure 5: Retailers estimate of data growth (per year)

More than 60%

State of the Industry Research Series: Big Data in Retail

Edgell Knowledge Network

Retailers focusing on building Big Data strategy first

Figure 8: Executed a Big Data project

Figure 9: Budgeting for Big Data

40.3% I dont know

State of the Industry Research Series: Big Data in Retail

Retailers focusing on building Big Data strategy first

Edgell Knowledge Network

Business priorities, budgets and ROI cited as biggest inhibitors to investment

Outsourced Analytics Services

Internal Team Augmentation & Training (Analytics)

Somewhat important Important

10% 36% 52% 10%

Figure 11: Challenges in starting Big Data initiatives

17% 30% 28% 19% 28% 35%

No Tools No Budget No Internal Ownership Champion Not A Business Priority