simulating the evolution of business analytics at...

25
MIT Center for Digital Business CIO Conference May, 2011 Simulating the Evolution of Business Analytics at SAP

Upload: others

Post on 23-Mar-2020

0 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

MIT Center for Digital Business CIO Conference May, 2011

Simulating the Evolution of Business Analytics at

SAP

Page 2: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 2

AGENDA

Overview

Challenges & Opportunities: Innovating for Analytics

Partnering on Simulation Modeling for Strategic Management

MIT Simulation Approach

Recommendations

Conclusion

Page 3: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 3

EVOLUTION OF ENTERPRISE APPLICATIONS EMERGENCE OF BUSINESS ANALYTICS

1970s – 1980s 1990s – 2000s 2010 - Beyond

SAP invents ERP R3 and Business Suite Business Analytics

Automating Transactions

Automating Efficient Business

Processes

Optimizing Decision Making

THREE CONVERGING

FACTORS DRIVING THE

EMERGENCE OF

BUSINESS ANALYTICS

Page 4: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 4

VISION FOR BUSINESS ANALYTICS BRINGING ANALYTICS TO THE MASSES

■ Static Analytics for the Few

■ Focus on IT Requirements

■ Siloed Decisions

■ Information to Validate, Review and React

■ Proprietary Data Source

■ Traditional Deployment

■ Dynamic Analytics for Everyone

■ Focus on Business Needs

■ Collaborative Decisions

■ Insight to Anticipate, Share and Act

■ Any Data Source

■ On Premise, On Demand and On Device

To From

Page 5: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 5

Strategy Management Planning, Budgeting,

and Forecasting

Profitability and Cost Management

Financial Consolidation

Enterprise Performance Management

Disclosure Management

Enterprise Data Warehousing

Data Mart Solutions High-performance Analytic Solutions

Data Warehousing

Reporting and Analysis

Dashboards and Visualization

Data Exploration Mobile

BI Platform

Business Intelligence

Enterprise GRC Access Risk Management

Global Trade Services Continuous Transaction Monitoring

Governance, Risk, and Compliance

Data Services

Master Data Management

Event Processing

Content Management

Enterprise Information Management

Information Governance

Analytic Applications

By LoB Service, Sales, and Marketing

Procurement

Finance

Sustainability

IT, HR, and more…

By Industry Financial Services

Public Sector and Healthcare

Manufacturing

Consumer Products

Retail and Telco ….

COMPREHENSIVE PORTFOLIO NEW CATEGORY OF ANALYTIC APPLICATIONS

Page 6: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 6

Focused, purpose built niche applications

Tailored for the business user/ approved by IT

Solves specific use cases and user requirements

Captures deep domain expertise

Business Model Changes Required

Rapid development of new applications (6-8 month timeframes)

New pricing models

Volume play versus traditional play

Product lifecycle considerations

New channels & ecosystem approach

PURPOSE-BUILT ANALYTIC APPLICATIONS DIFFERENT APPROACH FROM CURRENT BUSINESS

Page 7: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 7

Executive Challenge “What Does Success Look Like?”

Partnership with MIT

MIT Center for Digital Business is the world's largest center for research focused on the digital

economy

Matched SAP resources with MIT researchers to form a collaborative research team

Simulation Modeling for Decision Making

Ability to validate overall approach while helping to make strategic decision

Allows for a common platform for communication and analysis of strategic options

Identifies high-leverage strategies that take organizational constraints into account

PARTNERSHIP WITH MIT ON SIMULATION MODELING

Page 8: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 8

A. Questions

What is the best speed, depth,

and quantity of applications?

What are the impacts of early

adopters on long-term revenue?

How long should the support

window be for short lifecycle products?

COLLABORATION WITH MIT

Page 9: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 9

A. Questions

What is the best speed, depth,

and quantity of applications?

What are the impacts of early

adopters on long-term revenue?

How long should the support

window be for short lifecycle products?

MIT APPROACH

Test

Base

Case time

Sales

What is the cause

of the difference?

Stage 1 Stage 2 Stage 3 Finished

ApplicationsBusiness Case

Identification

Business Case

Identification Rate

Advancing to Stage 2 Advancing to Stage 3 Packaging

Avg Time to DesignAvg Time to Produce

Avg Time to Package

Total Aware

New aware

Avg Deals per

Application

Total Revenue

New revenue

Avg Size of Deal

Avg cost of Stage 1

per Application

Total stage 1 costs

Avg cost of Stage 2

per Application

Total stage 2 costs

Avg cost of Stage 3

per Application

Total stage 3 costs

Total Costs

Increasing costs

Operating profit

Reference Time to

Design

Relative Time to

Design

Effect of Design

Time on Deals

Effect of Design

Time on Deals f

Base deals per

application

Removing stage 2

applications

Reference stage 2

applicationsRelative stage 2

applications removal

Effect of removing stage

2 applications on deals

Effect of removing stage

2 applications on deals f

Awareness Rate

per App

Leads

Generating leads

Installed BaseNew Sales

Support Costs

Fixed Costs per AppMarnigal Costs per App

Support Revenue

Support Revenue

per Deal

Supported Apps

Increasing

Supported Apps

Retiring

Supported Apps

Retirement Delay

Support profit

 

S2 = ò(AS2 - AS3)dt + S20

B. Modeling C. Analysis

Page 10: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 10

Background System Dynamics Modeling was developed at MIT in the 1950s.

Its been applied to numerous domains such as strategy, management, & process improvement;

its even been applied to insurgency & nation failure.

Approach It is designed to help addresses limitations of linear logic and over simplification caused by

typical human assumptions and behaviors.

In other words, its hard to manage complexity in our heads alone.

Key Features We can design simulations to experiment in complex systems.

For example, we can ramp up workforce faster than doable in the real world

MIT SIMULATION METHODOLOGY

Page 11: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 11

Conducted interviews with a wide variety of

stakeholders

Directed literature review across multidisciplinary topics: software

development, process improvement & strategy

Formulated dynamic models and simulation environments

Identified broad strategic concerns and articulate initial

recommendations

JUNE 2010 JANUARY 2011

MIT LIFECYCLE

PROGRESSION

Page 12: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 12

Enhance product quality and the ability to better select and cull applications to improve market performance

DEVELOPMENT

Early stage application development with alignment to sales improves outcomes and reduce risks

Balance product retirement and support window for positive ROI

FOCUS AREAS OF MIT MODEL FOR SAP

SALES

SUPPORT

Page 13: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 13

FOCUS AREAS OF MIT MODEL FOR SAP: DEVELOPMENT

Page 14: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 14

Total Deals

10,000

7,500

5,000

2,500

0

0 8 16 24 32 40 48 56 64 72

Time (Month)de

alTotal Deals : base Total Deals : test1

Result 1: Enhanced attention in early stages improves deal flow

Finished Applications

200

149.9

99.8

49.7

-0.4

0 8 16 24 32 40 48 56 64 72

Time (Month)

appl

icat

ion

Finished Applications : base

Finished Applications : test1

Total Deals

8,000

1,,000

1,,000

1,999

-0.8

0 8 16 24 32 40 48 56 64 72

Time (Month)

deal

Total Deals : base Total Deals : test1

Result 2: Even with fewer application, better application selection increases deals

10k

2010 2015

BASE

TEST

(Increase development time)

200

2010 2015

DEALS

APPLICATIONS

10k

2010 2015

DEALS

BASE

TEST

(Cull less

promising

applications)

BASE

TEST

SIMULATIONS

Page 15: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 15

DEMO

Page 16: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 16

Most weight (KPIs) Most applications Quality

Cost

Time

RELATIONSHIPS

Page 17: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 17

Applications

Deals Size

DYNAMICS

STRATEGY

LEVERS • Application weight

• Development time

• Time to market

OPTIONS

Page 18: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 18

DEVELOPMENT

RECOMMENDATIONS

Big opportunity for performance enhancement from improving early stage development

Information feedback from market into product development More KPIs and more accurate KPIs

There are advantages of not moving every application to SKU and selecting more promising applications

Reduced production costs More deals per application Lower support costs and support risk

Entering many applications into the pipeline can help mitigate risks of removing applications

More leeway for selection Less pressure to finish each application

Page 19: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 19

OTHER FOCUS AREAS OF MIT MODEL FOR SAP: SALES AND SUPPORT

Page 20: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 20

SALES

RECOMMENDATIONS

Need to seed and build the sales pipeline upfront Improved alignment of the sales plan to the execution plan Early planning needed to fill the 4x ratio

Early adopters help drive both product quality and awareness and leads

Collaboration helps to pre-set the revenue process Helps to close revenue via leads and application attractiveness

Page 21: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 21

SUPPORT

RECOMMENDATIONS

Maintenance and support of applications is a key input to overall ROI

Inputs to support include: Fixed costs per application Marginal cost per deal supported Revenue income per deal Lifespan of support costs (7 years) and revenue (years)

Dynamics of number of applications and number of customers are key to support performance

Page 22: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 22

A. Questions

What is the best speed, depth,

and quantity of applications?

What are the impacts of early

adopters on long-term revenue?

How long should the support

window be for short lifecycle products?

MIT COLLABORATION RESULTS

Test

Base

Case time

Sales

What is the cause

of the difference?

Stage 1 Stage 2 Stage 3 Finished

ApplicationsBusiness Case

Identification

Business Case

Identification Rate

Advancing to Stage 2 Advancing to Stage 3 Packaging

Avg Time to DesignAvg Time to Produce

Avg Time to Package

Total Aware

New aware

Avg Deals per

Application

Total Revenue

New revenue

Avg Size of Deal

Avg cost of Stage 1

per Application

Total stage 1 costs

Avg cost of Stage 2

per Application

Total stage 2 costs

Avg cost of Stage 3

per Application

Total stage 3 costs

Total Costs

Increasing costs

Operating profit

Reference Time to

Design

Relative Time to

Design

Effect of Design

Time on Deals

Effect of Design

Time on Deals f

Base deals per

application

Removing stage 2

applications

Reference stage 2

applicationsRelative stage 2

applications removal

Effect of removing stage

2 applications on deals

Effect of removing stage

2 applications on deals f

Awareness Rate

per App

Leads

Generating leads

Installed BaseNew Sales

Support Costs

Fixed Costs per AppMarnigal Costs per App

Support Revenue

Support Revenue

per Deal

Supported Apps

Increasing

Supported Apps

Retiring

Supported Apps

Retirement Delay

Support profit

 

S2 = ò(AS2 - AS3)dt + S20

B. Modeling C. Analysis D. Takeaways

Page 23: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 23

Enhanced existing portfolio process. Ideas that did not meet market needs would not be launched as a final application, but remain in demo/accelerator form

DEVELOPMENT

Validated importance of making co-innovation customers successful and referenceable prior to final launch

Carefully weighed product release and associated costs with years of support offered

APPLYING THE INSIGHTS TO THE BUSINESS

SALES

SUPPORT

Page 24: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 24

Simulation is a good approach for modeling a new business

Allows understanding of business and organizational changes required

Examines cost factors impacting the business

Captures critical linkages and dependencies across business functions

Strong partnership and executive sponsorship critical to success

Requires participation of senior leadership team in the process

Importance of dedicated champion that can bridge multiple domains

Results & Next Steps

SUCCESSFUL APPLICATION OF SIMULATION

Page 25: Simulating the Evolution of Business Analytics at SAPebusiness.mit.edu/sponsors/common/2011-AnnualConf/siegel.pdf · 2011-05-18 · Simulating the Evolution of Business Analytics

© 2011. All rights reserved. / Page 25

Thank you!

Contributors:

Daniel Goldsmith, Research Scientist

Michael Siegel, Principal Research Scientist

Shivani Govil, VP, Product & Bus. Strategy

Sandra Ballew, Sr. Operations Expert