short presentation titleassets.timoelliott.com/docs/sapsa_track.pdf · 2011-09-13 · +13.4%....

71
BETTER RUN Business Analytics: The Big Leap Forward

Upload: nguyenbao

Post on 10-Jun-2018

215 views

Category:

Documents


0 download

TRANSCRIPT

BETTERRUN

Business Analytics: The Big Leap Forward

2

Top Business Issues 2011

3

Everybody Has Questions At Every LevelO

PER

ATIO

NS

l H

R l

FIN

ANC

E |

IT

| S

ALES

l M

ARKE

TIN

G

MANUFACTURING l RETAIL l HEALTHCARE l BANKING l UTILITIES l TELCO | PUBLIC SECTOR

Tom Davenport International Institute for Analytics

How and whydid it happen?

What is the risk if it does/doesn’t happen?

How do you prevent / ensure it happens again?

Whathappened?

What is happening now?

What willhappen?

4

Business Analytics Provides Great Value

Data is extremely important for competitiveadvantage

Data makes an important contribution to customer relations efforts

Business information has helped manage costs or improve operations

Executives believe companies can benefit greatly from using data, especially information generated within the company

Agree: 69% Agree: 77% Agree: 70%

5

Surging Growth in Business Analytics

2009 2010

+3.8%

+13.4%

Gartner: worldwide BI, analytics and performance management software revenue

BI Growth more than tripled between 2009 and 2010!

6

Analytics is an Ever-Increasing Share of IT Budget

2009 2010 2011

3.9%

+4.1%

+4.3%

Gartner: worldwide BI, analytics and performance management software revenue

“BI spending has far surpassed IT budget growth overall for several years”

Dan Sommer, Gartner

7

Business Analytics Around the World

Business Analytics MarketGrowth 2010

3.0%

3.7%

6.7%

17.8%

18.3%

19.5%

22.9%

Eastern Europe

Japan

Western Europe

North America

Middle East and Africa

Latin America

Asia/Pacific

13.2%

11.6%

Gartner Market Share Analysis: Business Intelligence, Analytics and Performance Management Software, Worldwide, published March 2011

88

Market Consolidation Continues

9

Business Analytics Market (BI, EPM, Analytic Applications)Share of Market, 2010

Worldwide Leader in Business Analytics

SAP

Oracle

SAS Institute

IBM

15.6%

13.2%

11.6%

23%

Gartner Market Share Analysis: Business Intelligence, Analytics and Performance Management Software, Worldwide, published May 2011

The 4 “Megavendors” continued to increase their market share– smaller vendors took from each other

10

“In-Memory Business Analytics” is Nothing New

The “What” doesn’t fundamentally change — but the “How” does

11

Key Data Challenges Today

Poor organization of data is a challenge

Agree: 88% Agree: 81% Agree: 77%

Poor processes for sharing data between departments and employees is a hurdle

Technical issues—such as data silos or incompatible systems—represent at least some challenge

12

Today’s Disks Can’t Keep Up with Processing Power

13

SAP HANA™■ In-Memory software + hardware

(HP, IBM, Fujitsu, Cisco, Dell)■ Data Modeling and Data Management■ Real-time Data Replication via Sybase Replication

Server■ SAP BusinessObjects Data Services for ETL

capabilities from SAP Business Suite, SAP NetWeaver Business Warehouse (SAP NetWeaver BW), and 3rd Party Systems

Capabilities Enabled■ Analyze information in real-time at

unprecedented speeds on large volumes of non-aggregated data

■ Create flexible analytic models based on real-time and historic business data

■ Foundation for new category of applications (e.g., planning, simulation) to significantly outperform current applications in category

■ Minimizes data duplication

SAP In-Memory Appliance (SAP HANA™)Architecture

SQL MDXBICSSQL

SAP In-Memory Computing studio

Sybase Replication

Server

SAP Business Objects Data

Services

SAP HANA™

Other ApplicationsSAP BusinessObjects BI Solutions

SAP NetWeaverBW

SAP Business Suite 3rd Party

SAP In-Memory Database

Calculation and Planning Engine

Row & Column Storage

14

Simplified Memory Hierarchy Main memory vs. Cache (Based on Intel Nehalem)

CPU

Core

1st Level Cache

2nd Level Cache

Shared 3rd Level Cache

Core

1st LevelCache

2nd Level Cache

Core

1st LevelCache

2nd Level Cache

Core

1st Level Cache

2nd Level Cache

Size Latency

64KB 1-2 cycles

256KB 6-20 cycles

8MB 30-60 cycles

SeveralGBs up to TBs

100-400cycles

Main Memory

Flash – 5000 cycles; Disk seek – 10,000,000 cycles

15

Columnar data store benefitsü Optimizes load of data to CPUü High data compressionü Very fast data aggregationü Makes use of real-life fill of tables (few fields filled, few

updates)

Can be joined with row-based data

Making Use of Columnar Data Store

A 10 € B 35 $ C 2 € D 40 € E 12 $

A B C D E 10 35 2 40 12 € $ € € $

memory address

organize by row

organize by column

A 10 €

B 35 $

C 2 €

D 40 €

E 12 $

conceptual view

mapping to memory

16

Row vs. Column Databases

My Filing System

My Wife’s Filing System

Row-based Column-based

17

Row-Based Data

Wasted space, and a full scan to aggregate any particular field

18

Column Data

More efficient data storage, better compression, faster queries

19

§ Regression§ Correlation and Covariance§ Analysis of Variance and Designed

Experiments§ Categorical and Discrete Data

Analysis§ Nonparametric Statistics§ Tests of Goodness of Fit§ Time Series and Forecasting§ Multivariate Analysis§ Survival and Reliability Analysis§ Probability Distribution Functions

and Inverses§ Random Number Generation§ Data Mining

§ Linear Systems § Eigensystem Analysis § Interpolation and Approximation § Quadrature§ Differential Equations § Transforms § Nonlinear Equations§ Optimization § Special Functions § Statistics and Random Number

Generation

In-Database Analytics

20

Predictive Analytics

Forecasting ClusteringAnomalies

Influencers Trends Meaningful or Random?

21

A Database Designed for Business

Volume DriverCyclesDriverForecast DriverForecast AgentsGrowSeasonal ComplexAssortment PlanningCumulateDaysDays OutstandingDiscounted Cash FlowDe-cumulateDelayDelay Debt

Delay StockAnnual DepreciationAnnual DepreciationDiminishing Balance

DepreciationSum of Year DepreciationYear To Date StatisticalYOY/ YOY DifferenceForecast Dual DriverForecast SensitivityFeedFeed OverflowForecastFundsFuture Value

Inflated Cash FlowInternal Rate of ReturnMoving MedianNumber of PeriodsNet Present ValueOutlookPaymentPresent ValueLagLastLeaseLease VariableLinear AverageForecast MixMoving Average/Sum

ProportionRateRepeatSeasonal SimpleSeasonal SimulationStock FlowStock Flow ReverseStock Flow BatchTimeTime SumMax ValueMinimum ValueTransformRounding

Up until now, there’s been a false separation between application logic and database functionality

22

Extending HANA

Business ApplicationsAnalytic Appliance

Business Intelligence

Cloud computingUnstructured and personal dataMobile revolutionCollaboration

23

And What About Hadoop?

24

What about Hadoop?

Complementary technology

Very real value, but immature

Primarily used today for preprocessing unstructured data

Velocity

Volume Variety

New analytic

platforms

HADOOP

25

HANA, Sybase IQ, And HADOOP

Event Driven Transactional Processing EDWOperational

Data Store

Multi-Dimensional

OLAP

Real-Time Real-Time Intraday+Intra-hour Intraday+

Small< 1GB

Small< 1GB

Large1 TB+

Medium100 GB+

Medium100 GB+

Events Parameterized ParameterizedParameterized Ad-HocPredictive

Analysis

Data Volume

Latency

Batch Processing

Intraday+

Very Large10 TB+

Ad-HocPredictive

Structured Data Analytics Un-StructuredSemi-Structured

Event Insight SAP ECC

Sybase IQ

HANA HADOOP

+

26

Reality Is, and Always Will be, Messy

Different information sources

Different levels of expertise

Different access devices

Different time horizons

Different levels of analytic need

Differentproject phases

RiskPolitics

But new architectures mean simplification and new opportunities

27

28

Data Quality

Data Integration

Master Data Mgt

Meta Data Mgt

Business IntelligenceTop-to-bottom

visibility required

Enterprise Information Management

29

Integrating Flows of Data

Incremental loads, replication

30

Integrating Flows of Data

31

Streaming Data

33

Real-Time Data Quality

If everything’s incremental, when do we do data cleansing?

Levels of quality

In-db cleansing

35

Social Data

36

Unstructured Data

Column stores are good at storing text data.

Can push the text analytics algorithms into the appliance, more flexibility

38

Text Data Processing for Unstructured Data

http://experience.sap.com/twitterta/sapsummit.jsp

40

Bridging the Gap

Corporate Local

41

Centralized Infrastructure, Full Autonomy

Data Warehouse

ApplicationData Department

Data

Personal Data

Ease of Use is The #1 Barrier to Deployment

Top Roadblocks to BI Success

Challenge Rank

Complexity of BI tools and interfaces 1Cost of BI software and per-user licenses 2

Difficulty accessing relevant, timely, or reliable data 3

Insufficient IT staffing or excessive software requirements for IT support 4

Difficulty identifying applications or decisions that can be supported by BI 5

Lack of appropriate BI technical expertise within IT 6

Lack of support from executives or business management 7

Poor planning or management of BI programs 8

Lack of BI technology standards and best practices 9

Lack of training for end users 10

1. Doug Henschen, InformationWeek, “BI Efforts Take Flight”, Oct 13, 2008

43

Use the Power to Improve Ease of Use

No longer query –wait – analyze –format …

Donkey Kong

Grand Theft Auto

46

Progressive Expertise

View Reports Strategic Analysis

47

Use the Power to Create New Applications

49

Services Analysis PurchasingFinancials Sales

Sales Overview• Sales Order per Customer (List

reporting)• Fulfillment rate (static, per value and

per quantity)• Credit Memo List• Billing Document List• Sales Organization Analysis

Master Data• Material List• Customer List• Vendor List

Generic

Accounting Overview• Flexible customer open item reporting

(Debitor)• Flexible vendor open items reporting

(Creditor)• Flexible open item reporting (New

General ledger)• Customer open item analysis (Day

sales outstanding)• New General Ledger line item

reporting

Shipping

Purchasing Overview• Order History Overview• Purchase Orders• Goods Receipts / Service Entries• Return Delivery Rate• Vendor Invoices

Shipping Overview• Outbound Deliveryies• Outbound Deliveries for Picking• Outbound Delivery Items• Outbound Delivery Items for

Picking• Stock Overview

HANA Rapid Deployment Solution for SAPHANA 1.0 SAP Operational Content – Included with HANA

50

Real-Time Financial Applications

Financial applications are the lowest-hanging fruit for the new architectures

Operational, but very analytic – push down budget algorithms, forecasting, ABC costing, etc. down into the underlying architecture

Strategy & RiskManagement

Business Planning& Consolidation

Execute withCompliance

PerformanceOptimization &Sustainability

AnalyticPlatform

51

SAP HANA ERP CO-PA Accelerator

n Accelerated reporting and month-end closing processes

n Flexible CO-PA Reporting, not limited in data scope

n Super-fast processing of queries and drill downs against large data volumes

n Easy integration of legacy CO-PA data with flexible modeling in PCM

n Better commercial decision making - pricing, discounts, sales strategy etc.

n Optimized month end close

n Timely self-service analytics for business users

n Rapid deployment and low cost of ownership

Expected benefitsNeeds

Faster processes and deeper insight for competitive differentiation

SAP In-Memory Computing - potential business impact

§ Business performance is impacted by poor reporting and month-end closing runtimes

§ Limited profitability insight and root-cause understanding due to large or incomplete data sets

Typical Business ChallengesSAP In-Memory Computing

600 million records

Drill-down to detail in seconds

Analyze any SKU, product family, region, time period …

HANA CO-PA Analysis

0 200 400 600 800 1,000

Standard System

In-Memory System

52

Structure of the CO-PA model in HANA

Keys:-Object no.-Customer-Product-Sales Org-...

Realigned keys:-Object no.-Customer-Product-Sales org-...

Persisted key figures:-Gross sales-Sales deductions-Variable cost-..

Line items CE1xxxx Objects CE4xxxx

Master data and texts

Master data and textse.g. for product

Analytic View for Actuals

Calculated key figures:-Contribution Margin-Net Sales-...

Keys:-Object no.-Customer-Product-Sales Org-...

Realigned keys:-Object no.-Customer-Product-Sales org-...

Persisted key figures:-Gross sales-Sales deductions-Variable cost-..

Line items CE2xxxx Objects CE4xxxx

Analytic View for Plan Data

Calculated key figures:-Contribution Margin-Net Sales-...

Calculation View for Plan & Actuals

HANA CO-PA AcceleratorERP CO-PA with HANA as secondary database

The solution

HANA

HANA 1.0

~30x compression100x faster

ERP

Traditional DB

CO-PAProfitability Analysis

Report Writer

BOBJ BI

BOBJ BI 4.0

Flexible reporting

Unabridged data

Archived data

Fasterallocations

Fasterreporting

Accelerated month-end closing ü Accelerated CO-PA Reportingü Accelerated ERP CO-PA Allocations

Improved Reportingü Business user driven data analysisü Instant response times

Eliminated data boundariesü No pre-defined data aggregation levelsü Complete life-time, line-item analysis

Deeper Insightü High-volume and ad-hoc data queriesü No limitations to reporting dimensions

54

HANA CO-PA

55

Line of Business: Finance SAP Dynamic Cash Management

§ Consolidated visibility and powerful calculations of massive volumes of transactional data (millions of documents) from heterogeneous source systems

§ Powerful cash forecasting based on critical drivers such as open shipments, open PO’s and customer/company payment behavior

§ Real-time visibility into a company’s cash position by customer, product, supplier, region, and broader time horizon

§ Easy-to-use interface based on Excel and SAP BI tools

n Maximize return on liquid/cash assets and optimize the company’s cash flow positions

n Improve forecast accuracy of cash flows to ensure critical business operations

Key BenefitsCapabilities

§ Ineffective management of cash, accounts receivables/payables leading to sub-optimal return on liquid assets

§ Limited visibility of cash due to challenges in consolidating cash flow data and restrictions in current systems (e.g., limited to 30-day window, lack of customer/document-level details)

§ High forecast error of cash due to insufficient visibility of drivers such as open shipments, PO’s, customer payment behavior

Business Challenges

Advanced by SAP In-Memory Computing

A new SAP application, advanced by SAP In-Memory Computing, that delivers powerful forecasting capabilities and real-time visibility for a company’s cash flow management processes

Solution Overview

67

Line of Business: Finance SAP Dynamic Cash Management

§ Consolidated visibility and powerful calculations of massive volumes of transactional data (millions of documents) from heterogeneous source systems

§ Powerful cash forecasting based on critical drivers such as open shipments, open PO’s and customer/company payment behavior

§ Real-time visibility into a company’s cash position by customer, product, supplier, region, and broader time horizon

§ Easy-to-use interface based on Excel and SAP BI tools

n Maximize return on liquid/cash assets and optimize the company’s cash flow positions

n Improve forecast accuracy of cash flows to ensure critical business operations

Key BenefitsCapabilities

§ Ineffective management of cash, accounts receivables/payables leading to sub-optimal return on liquid assets

§ Limited visibility of cash due to challenges in consolidating cash flow data and restrictions in current systems (e.g., limited to 30-day window, lack of customer/document-level details)

§ High forecast error of cash due to insufficient visibility of drivers such as open shipments, PO’s, customer payment behavior

Business Challenges

Advanced by SAP In-Memory Computing

A new SAP application, advanced by SAP In-Memory Computing, that delivers powerful forecasting capabilities and real-time visibility for a company’s cash flow management processes

Solution Overview

68

WYN-WYN-WYNMobile Opportunities

More People, More Often, More Context

69

Mobile Isn’t Only About “Mobile”

71

Adobe Flash Dashboards on Android

73

10k m

De NHM kijker

Eerste Romeinsenederzetting: “OppidumBatavorum”Jaartal: 12 voor Chr.Afstand: 300 meter

0.3

Augmented Reality

74

Filter by: Branch

HighstreetOperations +23%

NE 0.1km

Augmented Corporate Reality

75

Augmented Corporate Reality

76

Filter by: Maintenance History

Tower Pipe 3Last Maintenance: 2 Weeks

E 0.1km

Photo by Thomas Hawk, Flickr

77

78

Store 23Current sales: $15k

SE 0.1km

Filter by: Store Performance

80

“Computers are useless.

- Pablo Picasso

They can only give you

answers.”

91

Did You Know…

92

93

How Not To Stock Up For Promotions

SKU ProductAverage items sold prior 3 weeks

Items sold during special promotion % increase

120595 Kams Mint Toothpaste 8 oz 72 112 56%593300 Peepers Size 5 Diapers 32 pack 134 170 27%309454 Pata Negra Ham Sandwich 35 43 23%139913 Closers Breath Mints 40 112 180%149292 Bboy Barbecue Charcoal 2lbs 17 98 476%249200 Lindas Cookie Ice cream kids treats 26 65 150%202184 Giant Corn Chowder Soup 12 oz can 43 84 95%233120 Silly String Cheese, Lunch pack 12 55 358%210653 Green Label 6-pack beer 120 115 -4%

499 854 151%71%

Average of % Increase

Better: Ratio of total items sold provides different % increase

94

The REAL Big Leap Forward

© SAP 2008 / Page 94

Breadth and Sophistication of Possible Analytical Tasks

Perc

enta

ge o

f Use

rs D

oing

or

Thin

king

abo

ut th

ese

task

s

Quantitative Thinking Gap

Huge opportunity to make business people more productive and efficient, increase their satisfaction, save money for the company, and drive more revenue.

95

The SAP difference

Lightning Fast EasyTrusted

Anytime Industry/LoB Expertise

Collaborative

Thanks!

Email:[email protected]

BI Blog:timoelliott.com

assets.timoelliott.com/docs/sapsa_track.zip

You Should Follow Me on Twitter: @timoelliott