Outline 1. Components of Enterprise Systems 2. Dell: Pioneer of Virtual Integration 3. Cisco Did It Right ! 4. Spotlight on ERP Disasters: What Went Wrong? Hidden Costs of ERP Implementation A Success Story (Asian Paints) vs. A Failure (Nike)
Related searches for Enterprise Systems
1. Components of Enterprise Systems
2. Dell: Pioneer of Virtual Integration
3. Cisco Did It Right !
4. Spotlight on ERP Disasters: What Went Wrong?
A Failure (Nike)
An integrated suite of information systemsthat form the backbone of the enterprise for runningand managing its operations
Enterprise Systems Encompass
Seamless and Transparent
Flow of Data
Across the Entire Value Chain
Transaction over, customer support
begins with information and updates
being provided, typically via call
Electronic funds transfer happens
if buyer and seller are with the
same bank; online credit card
payment soon possible
Delivery systems take over from here
on. They are being increasingly
outsourced today to specialist firms
Since you hang on to the customer
for life, you profile the customer’s
data to create product updates
The customer configures his
own product. The configurator
software works out the possible
combination and price
Customers are lured into Net
space either by company
websites or by general portals,
kicking off the chain
With the customization process initiated,
the back end machinery
starts whirring to fulfill
the customer’s needs
Customization adds greater
complexity to the process,
making scheduling key to better
Get a logistics company to pick up
finished goods either from your
premises or outsourced premises
With the value chain transparent,
vendors know demand the moment
a customer logs in and start
delivering according to
the production schedule
The one bricks and mortar process
that still can’t be avoided, though
increasingly, this too is
being outsourced to others
Adapted from Businessworld, May 22, 2000
…“Islands of Automation”
Most Important: Emergence of the Internet
A cost-effective means to connect
the back-office and front-end systems
IT Architecture Should Integrate Information, Processes and Functions
of Enterprise Systems
of Enterprise Systems(…contd)
It is all about:
with the OUTSIDE world
…Not just speeding up and automating a company’s internal processes
…Spread the efficiency gains to the business systems of its suppliers and customers
…Collaboration is the competitive advantage
The dominant model in the personal computer industry – a value chain with arms-length transactions from one layer to the next:
Dell’s direct model eliminates the time and cost of third-party distribution:
Virtual integration works even faster by blurring the traditional boundaries and roles in the value chain:
Dell VirtuallyIntegratedthe Value Chain with its Customers and Suppliers
Benefits of Virtual Integration value chain with arms-length transactions from one layer to the next:
- Dell’s Suppliers
… Connected to Dell’s customer data through the corporate extranet
… Have real-time access to Dell’s customer orders
… Can schedule production and delivery to ensure Dell’s production line moves smoothly
- Dell’s Customers
… Connected to Dell’s supply chain via website
… Can track the progress of their order from Dell’s factory to their doorstep
… Save time and cost on telephone or fax inquires
- Dell: Big Savings
… Low Inventory: 13 days in 1998 vs. 40+ for Compaq – Now counting inventory in “hours”
… Customer Self-Service: Frees up Dell personnel
Implementation of Enterprise Systems value chain with arms-length transactions from one layer to the next:
- Two Big “Non-IT” Hurdles:
1. Data Problems
2. Organizational Resistance to Change
(Lewin-Schein Model of Change)
1. Legacy system could not handle 80% annual growth rate of Cisco.
Constant band-aids to meet business needs resulted in the application
becoming “too much spaghetti.”
Systems outages became routine exacerbated by the difficulties of
recovering from outages.
Problem: None of us was going to throw out the legacies and do something big.
Autonomous approach to replacing legacy systems in Order Entry, Finance and Manufacturing will not work.
SVP of Manufacturing took the lead - put together a team in February to investigate the replacement
1. Could not be an IT-only initiative
2. Need a strong integration partner to assist in both selection and implementation of the ERP solution
3. Selected KPMG because they brought experienced people to the engagement, not “greenies”
4. Team of 20 people tapped the actual experience of large corporations and knowledge of sources such as Gartner Group
5. After a week of evaluating the packages at a high level, two prime candidates were selected - Oracle was one.
6. Team Spent 10 days on the Request for Proposals
7. Visited reference clients of each vendor as part of “due diligence”
8. Vendors invited for a 3-day software demo
9. Oracle selected because
Total Time: 75 days
Time clock: May 1, 1994
Next step: Board Approval
1. Cisco’s Financial Year: August 1 - July 31
2. Constraint: Cannot implement in Quarter 4
3. One option: July/August 1995
4. Rejected it - too late
5. Worked backwards:
Target Date: February 1995
Project Time: 9 months
1. Upgrade legacy system
2. Replace it in parts
3. Big bang implementation
- One ERP solution for all systems in 9 months time
- Single largest capital project at the time
- No Cost-Benefit Analysis
- “We are going to do business this way.”
ERP System became one of Cisco’s top 7 goals for the year
“come in with an 80 -20 recommendation on how to configure the system.”
Versions 2 & 3
Reward for the ERP Team - Over $200,000 cash bonus
System Integration: 38%
Total Cost $15 Million
Note: Cost of Cisco personnel time not included beyond some members of the core team
Key Steps taken by Cisco Management value chain with arms-length transactions from one layer to the next:
Key Steps taken by Cisco Management value chain with arms-length transactions from one layer to the next:
e.g.: R/3 miscounted inventory, which in turn screwed up customer orders - Outright crashes were routine
- Allied Waste Industries
- Waste Management Inc.
A Notorious Disaster value chain with arms-length transactions from one layer to the next:
Hershey Foods – October 1999
No End In Sight . . . value chain with arms-length transactions from one layer to the next:Goodyear – November 2003
ERP Implementation: A Real Pain value chain with arms-length transactions from one layer to the next:
ERP implementation costs fall in the range of $3 to $10 per dollar spent on the software itself
- Meta Group
4. Data Conversion
5. Data Analysis
6. Getting rid of your consultants
Workers have to learn new processes
Not just a new software interface
e.g., A receiving clerk at the plant’s loading dock now becomes an accountant. Because the clerk is keying new inventory directly into a live system, mistakes have an immediate impact on the books. And the plant’s number crunchers can no longer simply look at their data in batches, now they need to be able to pinpoint the origin of each data entry to verify its accuracy.
otherwise, pay more later
Links have to be built between ERP and other corporate software on a case-by-case basis
Monsanto has add-on applications for logistics, tax, production planning and bar coding. Integrating them with SAP has consumed more time and money than estimated
If the ERP’s core code has to be modified to fit the business process, costs will skyrocket.
…Use DUMMY DATA
…And move it from one application to another
Run a real purchase-order through the system, from order entry to shipping and receipt of the payment -- the whole “order-to-cash” cycle - preferably with the employees that will eventually do the jobs.
Company received about 500,000 orders daily from thousands of pharmacies, each of which ordered hundreds of items.
…SAP could only handle a few thousand items a day
No way to test in advance…ran some simulations, but not with the level of data we have in an operating environment.
e.g. EI du Pont Nemours might also appear as Du Pont, DuPont and so on, giving the illusion of several customers - in reality, only one.
Reports in ERP package will NOT meet management information needs because …
… ERP data has to be combined with external and soft data such as goals, budgets, etc.
Cost of data analysis is often overlooked in project budget because of misconception that ERP package provides all the analysis users need
Users fail to plan for disengagement
e.g. A specific number of staff should be able to pass a project management leadership test - similar to what Big 5 consultants have to pass to lead an ERP engagement
AND lower-level end users who will have daily contact with the ERP systems and provide level of detail.
A Big Splash In Wall Street in March 2001... value chain with arms-length transactions from one layer to the next:
Nike Says Profit Woes Due To IT
Philip Knight, Nike’s Chairman and CEO, blamed the “complications arising from the impact of implementing our new demand-and-supply planning systems and processes” for the shortages of some products and excess amounts of others as well as late deliveries.
Result: Profits Fell Short of Estimate by 33%
I guess my immediate reaction is: This is what we get for $400 million?
Source: Computerworld, March 5, 2001
Some Causes for the Nike Problem value chain with arms-length transactions from one layer to the next:
A Success Story: Asian Paints value chain with arms-length transactions from one layer to the next: IT Funded Global Acquisitions
$20 million investment in IT
@ ASIAN PAINTS value chain with arms-length transactions from one layer to the next:
Restricted to India
Number of locations
Suppliers across the globe
Decided to install SCM software before ERP software
Top management insight
Did not recognize the complexity of a global Supply Chain Project
Phased – First SCM, then ERP, last CRM
i2 played a proactive role - suggested implementing smaller modules one at a time.
i2 did not adhere to what it did usually – it adopted a big-bang rollout approach
Three packages simultaneously – Nike IT staff spread thin
Restructured in 1998, before SCM Project – only modest customization needed in the software
Heavy customization of i2 software to fit Nike’s business processes – no pilot test due to aggressive time-table
The Big Hurdle: Change Management
Learning from Failure value chain with arms-length transactions from one layer to the next:- Case of BMC Software
Succeeded the third time after two failed attempts
What Went Wrong?
The Third Time value chain with arms-length transactions from one layer to the next:- BMC did it Right!
The Data You HAVE
The Data You WANT
The Data You NEED
Data problems are more difficult to solve than hardware and software problems.
“We found … that management … didn’t even have a good profile of its market and customers. It didn’t really know in summary form what (its position was) with respect to discrete market segments … There was very little account profitability and not even market segment profitability information.”
“Information on orders, sales and margins … are of maximum value when tied to … meaningful market segments. And segment-based data are of limited use to finance, hence the common misalignment problem between finance and marketing.”
. . . Field Sales, Telesales, Service,
Call Centers, Storefront, . . .
1. Careful design of the form to collect data
2. Train the Data Providers
3. Motivate the Data Providers
Inaccurate and low-quality data costs US businesses $611 B each year in bad mailings and staff overhead alone… More injurious than the unnecessary printing, postage and staffing costs is the slow but steady erosion of an organization’s credibility among customers and suppliers as well as its inability to make sound decisions based on accurate information.
Source: The Data Warehousing Institute Report, 2001
CIF requires data from a disparate set of databases, located in various parts ofthe enterprise containing data of varying ages collected from various sources and channels, and stored in a multitude of different architectures and platforms
Ex: Shell Exploration and Production took 7 months to map data from 27 data sources in a 450 GB data warehouse, using a tool from Kalido Ltd.
Every system has its own internal set of codes. Going back and cleansing the data in those host systems wasn’t an option. It would have taken too much time and been too expensive. Corporate politics was not too bad because no single business unit lost control of its data. And now they all contribute to a greater understanding of the company as a whole. Once the concept was proved, we had pressure from the top to integrate other applications as well. They could see themselves what information they could now get and how powerful it is.
Source: Computerworld, Apr. 15,2002
Form a data mapping committee – but keep it small or risk never reaching agreement
Problem: Poor quality of customer master data in the ERP system a manufacturer - a subset of large customers was labeled with an incorrect industry classification code.
Result: Overlooked in market segmentation analysis performed by the Marketing dept. - this customer segment received no promotions, which would have generated an estimated $ 5M in revenue within one year.
Good data quality does not drive value in and of itself but it is the
means to achieve high-value benefits. Although data quality
maintenance is not the front-facing functional module in a CRM
project, it is a necessity to get value from the CRM investment.
* Source: Gartner Viewpoint, Nov.29, 2001
Information = Power
Who has access
to What Data ?
The politics of competition within the
company is a real obstacle to developing a
common, shared CRM database.
Employees, especially touch-point personnel, have
to change the way they work
Big roadblock for CRM implementation
A very useful framework for thinking about the change process is problem solving. Managing change is seen as a matter of moving from one state to another, specifically, from the problem state to the solved state in a planned, orderly fashion.
The Lewin-Schein Model of Change
I wasn’t smart enough about that. I was reflecting my engineering background and was insufficiently appreciative of the human dimension. I’ve learned this is critical.
Benchmark goals aren't met
IT Fiascoes… and How to Avoid Them value chain with arms-length transactions from one layer to the next: