Big Data In Retail Shantanu Goswami
Big Data In Retail Shantanu Goswami
1990 20132000 2005 2010
ANALYTICS(CIRCA 1980)
PREDICTIVEANALYTICS(CIRCA 1980)
SEMANTICANALYTICS(CIRCA 1980)
How we got to Big Data
REAL TIMEIN THECLOUD
PERSONAL COMPUTERAND CLIENT SERVER
DATABASE(CIRCA 1980)
WWW MOBILE
SOCIAL
B2B/B2C
BIGDATA
REAL TIME
1,000,000+SOLD
3,000,000people had access toInternet worldwide
More people havemobile phones thanelectricity or safedrinking water
Facebook: 1 billion users; 600 mobile users; morethan 42 million pages and 9 million appsYoutube: 4 billion views per dayGoogle+: 400 million registered usersSkype: 250 million monthly connected users
2015
Opportunity to unlock value from Big Data
* “Big Data: Next frontier for innovation, competition, and productivity”McKinsey study
Customer DataCompetitive Prices
GPS
Promotion Planning
Speed
Hot Trends
Omni Channel
Store Enhancements
Employee Records
Suppliers
Purchase Orders
MerchandiseEmails
Tweets
Planning
Social Media
Mobile
Instant Messages
Optimal Channel Assortment
60%potential increase inretailers’ operating
margin with big data*
What if you could turnnew signals into business value?
:-)Brand Sentiment
360O Customer View
ProductRecommendation
Propensity to Churn Real-time Demand/Supply Forecast
Predictive Maintenance
Fraud Detection
Network Optimization Insider Threats
Risk Mitigation,Real-time
Asset Tracking Personalized Care
4
© 2013 SAP AG. All rights reserved. 5
DIGITALLYCONNECTED
SOCIALLYNETWORKED
HIGHLYINFORMED
A new echelon of customers havetransformed the business landscape for retailers
While retail objectives remain the same
Provide seamlesscustomer experienceacross all channels
Deliver optimal offersbased on the right
product, at the rightprice, via the right
channel, at the righttime
Improve saleseffectiveness, increaserevenue potential forcross-sell and up-sell
SAP makes Big Data Real
Big DataPlatform
DataScience
Accelerate Apply Achieve
Big DataAnalytics &
Apps
Big Data successdemands full coverage
Accessible
Deep
SimpleRealTime
Broad
• Answer complex questionson granular data
• Predict the best next action
• On any device orto any user
• Self service and intuitiveinteractions
• No data preparation• No pre-aggregates• No tuning
• Real-time streams of data• Ask a question, get an immediate
answer
• Massive data scale• Many data types
Next generation businessplatform for real-time retail
One PlatformBringing It AllTogether
In-Memory
BusinessTransactions
DigitalConnections
CollaborativeBusiness
CloudSocial
Big DataMobile
AdvancedAnalytics
SAP Big Data Platform
Big Data Platform Big Data Science
Real Time Real Value Real Results
Big DataAnalytics & Apps
SAP: Real-time, with real results
10
Big Data Retail Applications
Make Big Data
insights actionable
via industry specific,
business focused
applications from
SAP and its
partners.
11
HybrisOmniChannel
Insights
:-)AudienceDiscovery
(ADT)
PromotionsManagement
(PMR)
FraudManagement
SalesInsights
Social ContactIntelligence
(CEI)
SentimentIntelligence (RDS)
Forecast &Replenishment
Planning for Retail
SAP Analytics for Big Data
ENGAGE PREDICTVISUALIZE
SAP BI SAP Lumira SAP Predictive Analysis &SAP InfiniteInsightRapidly connect
individuals, data, andprocesses to understandthe business and drive
better decisions
Intuitively explore andpresent data to reveal
new insights at-a-glance
Confidently anticipatewhat comes next to drivebetter business outcomes
EMPOWER ALLPEOPLE TO GET
THEIR QUESTIONSANSWERED
SAP MAKES BIG DATA REAL
ANTICIPATESCONSUMERBEHAVIOR
SAP MAKES BIG DATA REAL
BIG DATA GIVESSHOPPERSFASHION
ADVICE TOFIT THEIR
STYLE
SAP MAKES BIG DATA REAL
LISTEN.TAP INTO THE
STREAMING VOICES OFPEOPLE, PROCESSES,
AND THINGS
SENTIMENT.
PERSONAL
OFFERS AND
CUSTOMERCARE
BASED ONTHOUGHTS
EXPRESSED
PERSONALIZE.
BUILD LOYALTY WITHUNIQUE EXPERIENCES
TAIILOREDFOR EACHCUSTOMER
RECOMMEND.
PREDICT INDIVIDUAL’SFUTURE PURCHASES
ON BROWSINGAND SHOPPING
BEHAVIOUR
APPLY TOOPERATIONS.
SHOPPERSGET FASHIONADVICE THATFITS THEIR
STYLE
ACHIEVEBREAK-OUTRESULTS.
PROGRESSMEASURED
USING MODELOPTIMAL FOR
MANAGER’S KPI
© 2013 SAP AG. All rights reserved. 22
How to start ?
© 2013 SAP AG. All rights reserved. 23
People & Skills GovernanceStandards &Processes
Information &Application ArchitectureUse Cases
How mature are your capabilities?Take our online assessment here: https://valuemanagement.sap.com/BigData2
No Big DataCapabilities
Search for asignificant UseCase in big data
Understand thebig picture fromall available data
Generate changein response to shiftsin data automaticallyor manually
Use big Data to predictoutcomes and adjustprocesses accordingly
INC
RE
AS
ING
MAT
UR
ITY
BUSINESS VALUE EXTRACTION THROUGH DATA INSIGHT
Big Data maturity model
© 2013 SAP AG. All rights reserved. 24
Practical 5 step guide to getting value from Big Data
1. Define Your Use Cases
• Don’t forget to look outside, apply design thinking, prioritize
2. Discover Your Data
• Check under the bed, don’t overlook the ordinary, mine opportunities in machine data
3. Design Your Business Case and Technical Architecture
• Source, acquire, transform, stage, model, view, exploit
4. Do an Initial Deployment
• Actualize the benefit calculation; determine core and edge issues; learn and iterate
5. Make Big Data a Business Capability
• Integrate into processes, measure effectiveness
© 2013 SAP AG. All rights reserved. 25
A practical example
© 2013 SAP AG. All rights reserved. 26
Latency, Confusion and the lack of Granular Insights ..
Data Scientists
Detail levelNo transactional data available, only aggregated?IT support and overnight processes required?Details missing for ad-hoc analysis per stores andproduct/category?
Time & costSales analysis only available with time lag, not in real-time?Data analysis done only by specialists or externally?Time-consuming KPI calculations, low flexibility?
Business impactsReactive business decisions due totime lag in analysis availability?Deeper insights for assortmentoptimization delayed or missing?Higher costs for time-consuming dataanalysis?
CorporateOperations
Marketing
Planning
Buying
Supply Chain
Web, Mobile,Catalog
Stores
StoreOperations
SAP Sales Insights for RetailBASKET ANALYSIS. ASSORTMENT ANALYSIS . PROMOTIONS ANALYSIS
• Assortment Analysis - Spotting Unsuccessful products
• Basket Analysis - Drag along effect
• Evaluation of promotions – Vendor Analysis
© 2013 SAP AG. All rights reserved. 28
Customer Challenges
Which is the best / worst selling product
Which is the best running / worst store?
How successful are the promotions ?
How to change the assortment ?
What are the Root causes for revenue changes ?
Key Queries that are difficult to answer immediately ..
© 2013 SAP AG. All rights reserved. 29
SAP Sales Insights for Retail
SAP Sales Insights for Retail is an analysis toolbox tailor-made for POSdata models and retail business problems
© 2013 SAP AG. All rights reserved. 30
Where can we co-innovate with you ?
Promotions Analyzer
Fraud Management
Shelf Space Optimization
Repeat Purchase Analysis
Slow moving goods
Price Elasticity
Cannibalization
Hoarding Behaviour
Zone Optimization
Size Optimization
© 2013 SAP AG. All rights reserved. 31
SAP’s Retail Innovation Platform
Pricing
Fraud Detection
Assortment
Cannibalization
Hoarding
New Customers
Shelf Space
SAP CAR Data Model SAP Sales Insights for RetailLoyalty data Promotion Stock
HANA procedures, calculation view and specialized algorithms from SAPPredictive Analysis Library (PAL)
on-the-fly
Predefined Content – SalesInsights for Retail
SAP Innovation Extensions
External Extension and Innovation
Ad-hoc Analysis
POS data Vendor Funding
© 2013 SAP AG. All rights reserved. 32
Business Impacts
The 3 topics of Assortments, Basket Analysis & Promotions Analysis contributes to150 – 200 BPS to the topline.
The solution addresses
• Reduced analysis effort
• Fixing problems at underperforming stores faster
• Fixing problems at underperforming product categories faster
• Identifying success & failure earlier in new product introductions
• Ability to have out-of-the-box , retail specific content
• Extensions/Innovation ideas from SAP, free to innovate independently
• Lower entry point (LREA) and can extend with value selling (Enterprise)
• Flexible deployment models ( Both On-premise & HEC )
© 2013 SAP AG. All rights reserved. 33
Customer Co-Innovation Stories
Denmark
€7.5B Turnover.
43000
employees.
Departmental
store.
1000+ stores.
Customer runs a series of differentpromotion aimed at generating moretransactions and customer loyaltyacross the different store chains.Subsequently, it runs various types ofpromotions in their retail chains: Weeklyleaflets, Hit of the month product, TVcampaigns or in store promotions.
By gathering data about POStransaction and promotional data, theSAP Promo Excellence Tool tries toclassify each promotion into severaldistinct classes (Good promotion,Expensive promotion, Low impactpromotion).
This is a Co-Innovation we are pilotingon top of the Sales Insights on Hanaimplementation.
© 2013 SAP AG. All rights reserved. 34
Customer Co-Innovation Stories in Mexico
Mexico
€1B Turnover
5000
employees
176
Supermarts.
Grupo Merza uses SAP Sales Insights for Retailto identify and monitor their most importantproducts, categories, or stores to relevant KPIs,such as distinct buyers, units sales, profit, orrevenue.
With this information they can better understandthe impact of each SKU in their productassortment so they can drive market basket sizeand profit.
SAP Sales Insights for Retail also helps GrupoMerza compute how key performanceindicators (KPIs) such as average basket size,number of items per basket, shopping frequency,or promotions impact each other.
By monitoring these KPIs, Grupo Merza canquickly perform root-cause analysis wheneverthey observe unexpected revenue changes in astore or product category. With rapid insight,Grupo Merza can make adjustments quickly inpromotions or product assortment to ensureprofit.
InnovationPlatform
DataScience
Accelerate Apply Achieve
Analytics &Apps
Why SAP ?
© 2013 SAP AG. All rights reserved. 36
Real TimeAnalytics
MultisourceData Model
Retail Specific
Content
Customextensions
Key Differentiation
Gracias !!Shantanu GoswamiDirector – Retail Sales Motions
This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review,copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.
Reúnete conmigo para discutir