Successfully reported this slideshow.

ERP Considerations


Published on

  • Be the first to comment

  • Be the first to like this

ERP Considerations

  1. 1. ERP Considerations Enterprise Resource Planning (ERP) Considerations1 Peter Aiken, Bob Dudas & Susan Carter VCU/Institute for Data Research Executive Summary ERPs offer organizations the potential of unprecedented integration. However if management does not avoid certain common pitfalls, success can be elusive and costly. Reading this paper you will discover that ERP implementations are more likely to: o Fail; o Be delayed; o Cost more than forecast; or o Fail to deliver full functionality than they are to succeed. It is important to be aware of how ERP as a technology evolved, what its strengths and weaknesses are, the nature of important implementation challenges. Further, understanding of how ERP and legacy system metadata can be used to simplify implementation and help organizations best plan for this exciting new challenge. Contents Introduction: ERP Defined.................................................................................................. 2 ERPs Complex MRP II Roots ............................................................................................. 3 ERP Capabilities ................................................................................................................. 3 Achieving Organizational Understanding:........................................................................... 4 Other ERP Challenges........................................................................................................ 5 Inaccurate Expectations......................................................................................... 5 Customization Challenges ..................................................................................... 7 Long payback period.............................................................................................. 7 Data quality costs................................................................................................... 7 Other ERP Hidden Costs ....................................................................................... 8 ERP Metadata Value in Support of ERP Implementation................................................... 9 1 This paper was initiated in the Fall of 2002 when the first author was a Visiting Scientist as part of an Internal Research and Development (IRAD) project for the Software Engineering Institute (SEI) at Carnegie Mellon University. The Institute for Data Research (IDR) is a research and development center located in Richmond, Virginia and affiliated with Virginia Commonwealth University. The Institute is an interdisciplinary research organization founded to further the study of data and metadata as valuable organizational assets. The key to IDR's successful practice of creating business value is the application of innovative data engineering and management techniques for recovering, understanding, improving and re-using organizational metadata assets. Institute for Data Research • Maggie L. Walker Business and Technology Center 501 East Franklin Street, Suite 522 • Richmond, VA 23219 • – – 1.804.521.4056
  2. 2. ERP Considerations (page 2 of 11) Enterprise Resource Planning (ERP) Considerations Introduction: ERP Defined Replace this: (N*(N-1))/2 Logistics Application Ordering Application Finance Application HR/Payroll Application The acronym "ERP" stands for Enterprise Resource Planning. As illustrated, ERPs are popularly considered to be modern, (8*(8-1))/2=28 integrative, technology-based solutions that Data Data Data Data permit organizations to replace: With this: Integrated Applications Logistics Ordering Finance HR/Payroll • Disparate, legacy applications and stove-piped data; and the Shared Data • Associated organizational inefficiencies that arise from the use of the disparate data and applications. Because legacy systems were typically not designed to easily share information with other systems – they don't. This leads to confusing situations where the organization doesn't know the answer to questions such as how many customers it has! Each disparate legacy application/data combination typically supports an individual functional-area such as finance, ordering, or logistics, etc. Only recently have organizations become aware of that the magnitude of these disintegration costs – ranging from 20-40% of organizational IT budgets! ERP solutions provide integrated applications that access standardized enterprise data – replacing the hodgepodge of legacy systems. Major ERP vendors attempting to help organizations reduce their organizational disintegration costs include: PeopleSoft (, SAP (, Baan (, Oracle (, Lawson (, J D Edwards ( When considering enterprise resource planning as a potential solution to specific organizational challenges, major considerations include: • Understanding ERP's complex MRP II origins; • ERP capabilities; • Five major ERP implementation challenges as: o Unrealistic ERP timeline expectations, Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  3. 3. ERP Considerations (page 3 of 11) o Customization challenges, o Long payback periods; o Data quality, cleansing, & conversion obstacles; o Lack of understanding of ERP capabilities; and o "Hidden" implementation costs • ERP metadata value in support for ERP implementation. ERPs Complex MRP II Roots ERP's roots emanate from the same engineering-based solutions developed as "MRP" (or Manufacturing Resources Planning). MRP solutions integrated the data used to produce bill of materials, inventory, and production material requirements. Functional area coverage included areas such as business planning, sales and operations planning, production planning, master production scheduling, material requirements planning, capacity requirements planning, and the execution support systems for capacity and material. Additional MRP capabilities included the ability to forecast the quantity and timing of raw material deliveries; simulation capabilities to answer "what if" questions; and integrated financial reporting such as the business plans, purchase commitments, shipping budgets, and inventory projections. ERP's MRP roots provide an engineering-focused solution development perspective that can be perceived as technically detailed and complex. As we shall see, these complexities lead to implementation difficulties based on misunderstanding of what an ERP can and cannot do and what it takes to make an implementation successful. ERP Capabilities Generally ERPs are considered to possess more robust technology/integration capabilities, and increased functional area coverage (for example adding human resource functionality, maintenance, project planning and document management capabilities) than MPR II solutions. According to one vendor's website ERPs are commercial-off-the-shelf (COTS) enterprise software applications that are marketed as "global, enterprise-wide software solutions ... (consolidating) fragmented resources formally devoted to supporting functional stand-alone systems. (By investing in ERPs) … resources formerly devoted to the tedious process of converting and integrating data among individual applications are freed to apply to analysis. Increased access to better quality and more integrated information can be powerful tools with which to run an organization."2 In short, ERPs are intended to be software systems that permit coordinated exchange of data using common data structures and coordinated procedures among a wide-variety of functional organizational areas. In spite of the size of the investment, ERPs are popular with management especially those with complex and/or costly legacy environments. ERPS can be purchased by 2 – accessed 10/98. Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  4. 4. ERP Considerations (page 4 of 11) functional area module permitting organizations a degree of flexibility when determining the scope of a package. For example, an organization might typically implement a combination of pay and personnel management modules to replace two or more legacy payroll and HR systems. Other popular ERP solutions: integrate financial information; customer order/relationship information; standardize and speed up manufacturing processes; reduce inventory; or standardize HR information. However it is rare that management fully understands the total cost associated with ERPs prior to implementation. The increases costs (which typically include: o Longer that expected implementation schedules; o More complex customization requirements; and o Unexpected data quality problems and data structure mis-matches) can be avoided with additional planning and a complete assessment of the current environment and an in-depth understanding of exactly what the ERP can do for the organization. Another often-overlooked factor in implementation is having the resources that completely understand the offerings and component interaction of the ERP systems. Generally speaking, most organizations gather their legacy systems experts together and attempt to use them as their ERP experts. This can be a risky (and ultimately costly) mistake. A better approach is to create a team of legacy experts and ERP experts together to create a complete analysis, general design and detailed design not only of what the new system will look like and include, but additionally exactly how the conversion will take place. This extra time and cost spent up front can greatly reduce the schedule and budget overruns and will create a better-integrated system. When correctly implemented, ERPs make sound arguments for data-centric enterprise software development. (While implementing the complex logic associated with enterprise-wide referential integrity in their applications layer – at least this form of implementation permits organizations to effectively "outsource" complex data integration functions to the package manufacturers.) They shield the enterprise from the complexities of having to develop standard data structures for sharing and operating procedures for coordinating activities among organizational components. PeopleSoft and SAP (for example) each contain thousands of data tables and individual procedures for interacting with the enterprise software. Because ERPs attempt to permit organizations to capitalize on planned information sharing cost avoidance, they make sense when existing organization procedures and data structures can be successfully adopted to match those implemented by the ERP – the process of ERP customization or tailoring. Achieving Organizational Understanding: A major ERP challenge faced by organizations is recognizing that the integration of previously un-integrated job functions, requires that knowledge workers supporting Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  5. 5. ERP Considerations (page 5 of 11) different organizational functions will now be using the same software and they will both be entering information that affects each other. To do this accurately, these knowledge workers must have a much broader understanding than they did before the ERP solution of how others in the organization perform their functions. Understanding has several specific definitions in an ERP context: "Understanding an architecture" indicates that the organizational legacy systems and the ERP are documented and articulated as a digital blueprint illustrating the commonalities and interconnections among the component metadata. Understanding is a shorthand reference for use of a data centric technique to represent, manage, and develop manipulable system component models within a formally defined common metadata model. Common metadata model components are represented using standardized notation and should be sufficiently detailed to permit both business analysts and technical personnel to separately read the same model, and come away with a common understanding of the model. Other ERP Challenges As mentioned above, most ERP implementations suffer from two common problems: inaccurate expectations and customization/tailoring challenges. Inaccurate Expectations The first thing to realize when considering ERPs is that inaccurate expectations are the norm but they can be avoided. Most ERP implementations today result in cost and schedule overruns. Consider the following statistics:3 3 Standish Group web site – – accessed 10/98. Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  6. 6. ERP Considerations (page 6 of 11) • 10% of ERP implementations On time, within budget, as planned 10% 350% succeed with full functionality, within 300% forecast cost and time frames. 250% 230% • Cost overruns average 178%; 200% Overrun 55% 178% • Schedule overruns average 230%; Cancelled 35% 150% • 100% Implemented functionality averages 50% 100% 100% 59% 41% of what was desired. 0% 41% Cost Schedule Planned Functionality Other recent ERP related findings include4: • A 2001 Robins-Gioia Survey of ERP implementers found that 51% viewed their ERP implementation as unsuccessful and 46% characterizing their organizations as not understanding how to use the system to improve the way they conducted business. • 2001 Conference Board Survey participants indicated only 34% of respondents were very "satisfied" with what they got and that 40% of the projects failed to achieve their business case within one year of going live. These outcomes indicate a lack understanding of ERP implementation complexities. Routinely, the cost of implementing and the time required to implement are underestimated while the scope of what organizations are able to implement are routinely overestimated. By focusing more attention and resources on the analysis and design phases and creating a cross-functional team in the critical planning stages, organizations can better understand, estimate and prepare for a successful implementation. One of the largest avoidable costs is that associated with poorly or incompletely analyzed data. If organizations allotted more time and resources to analysis of the state and format of their existing legacy data, the constraints and validations exercised against the data, and precisely how that data can be and will be mapped to the new system, much pain and cost could be avoided. Typically, organizations do not research the structures of the legacy systems and the high level structures implemented by the ERP providers. However, generally the quality of data (especially historical data stored over long periods of time), the different roles that the individual data fields have played over time, and the various processes that access the data are often overlooked. Additionally, an exact mapping of fully analyzed legacy data to actual ERP structures is rarely executed to the extent to which it should be. 4 IT Cortex web site – – accessed 10/02. Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  7. 7. ERP Considerations (page 7 of 11) Customization Challenges Because ERPs attempt to permit organizations to capitalize on planned information sharing cost avoidance, they make sense when existing organization procedures and data structures can be successfully adopted to match those implemented by the ERP. Most organizations discover how compatible the new ERP is with their existing systems and processes when the turn on the new system. When it is discovered that the ERP differs from the system it is replacing the organization is faced with possibility of customization or tailoring the ERP. There are four basic choices to 1. Modify the ERP to match the organizational processes and/or data structures; 2. Modify the organizational processes and/or data structures to match the ERP (Note: that 1 & 2 require understanding of the ERP and organizational processes and data structures); 3. Perform some of choice 1 and choice 2; or 4. Ignore the problem. Most organizations approach the customization/tailoring decision with out proper information required to reach a good decision. Taking the time and allocating the resources up front can help organizations to avoid these costly surprises. Long payback period Another not well understood statistic is that the return-on-investment for most organizations runs to almost three years or in CIO terms – no benefits will be realized until your successor's, successor is in the job of the CIO who implemented the ERP project. A recent Meta Group study indicated the median annual savings from the new ERP average $1.6 million/annually on (roughly) $30 million dollar investment one year after implementation is completed. And a Deloitte survey of Fortune 500 companies found one in four suffering a drop in performance when their ERP system went live. Data quality costs A recent data management study performed by Price, Waterhouse, Coopers survey revealed two troubling facts about organizational data quality: • "Only 15% of companies are very confident of the data received from other organizations" Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  8. 8. ERP Considerations (page 8 of 11) • "Only one in three companies are very confident in the quality of their own data"5 Garbage in – garbage out as they say. Poor quality data input can be fatal to ERP projects. Just imagine the confidence that the new system engenders as they get bad data out of the system more easily than the legacy system. This too can be avoided with proper analysis, design and cleansing prior to implementation. Most organizations discover data and process anomalies after attempting to implement what they believe is a compete system. Through gap analysis, risk assessment and a detailed plan of attack is clearly the best way to avoid these potential failures. Other ERP Hidden Costs Other common "hidden" ERP implementation costs6 include the following: • Training – the most underestimated budget item because knowledge workers almost invariably have to learn a new set of processes, and must understand the "new rules" and interpretations not just a new software interface. • Integration and testing – analyzing, implementing and testing literally hundreds of links between ERP and other corporate software. This is often done on an ad- hoc basis after the conversion. • Customization – of the ERP happens when it can't support one or more of your business processes and you decide to try to make it do what you want. Guess what, typically you'll have to do it all over again when a new version of the ERP is released (past experience indicates this is approximately every eighteen months). • Data analysis & conversion – ERP data must be combined with external data for analysis and gap identification and often organizations don't perform these functions thoroughly enough. Organizations are likely to underestimate the size and cost of the conversion process. • Consultants ad infinitum – When users fail to plan for disengagement, consulting fees run wild. To avoid this, companies should identify objectives for which its consulting partners must aim when training internal staff. Include metrics in the consultants' contract; for example, a specific number of the user company's staff should be able to pass a project-management leadership test—similar to what Big Five consultants have to pass to lead an ERP engagement and a 5 PriceWaterhouseCoopers, Global Risk Management Solutions, Global Data Management Survey 2001: the new economy is the data economy, 5/23/01 accessed on 8/8/01 – 0cae4d32ccbaba2380256a0d004e454b/$FILE/Data+Management+brochure.pdf 6 accessed 10/9/02. Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  9. 9. ERP Considerations (page 9 of 11) knowledgeable transfer is critical in these scenarios. The more you get, the better you will survive when the consultants leave. • Replacing your best and brightest – staffing the project with the best and brightest is required because the software is complex and the business changes dramatic. Huddle with HR early on to develop a retention bonus program and create new salary strata for ERP veterans. Keep in mind that a diverse team is typically more successful. Having a balance of legacy and ERP experience will afford the organization the optimal combination covering the data and process live cycles. Additionally, sometimes have a combination of veteran and new team members typically help to balance both of the natural but dangerous tendencies of "doing it the old way and scrapping it all and starting fresh." • Implementation teams can never stop – After ERP, you can't go home again - the implementers are too valuable. Because they have worked intimately with ERP, organizations can't afford to send their project people back into the business because there's so much to do after the ERP software is installed. Teams must be created and focused on continually monitoring, documenting, and revising the ERP to keep up with changes in the business environment without compromising the work done during ERP creating. ERP Metadata Value in Support of ERP Implementation Using a semi-automated approach, you can support your ERP initiative in a range of ways – knowledge of ERP metadata can help you to: • Understand your existing environment in terms of organizational strengths and weaknesses. • Objectively assess individual ERP solutions in terms of their fit with your organizational practices and their flexibility to let you expand in the future. • Assess various software versus procedural change options as you try to respond to identified gaps. • Really understand the logical and physical data structures and their corresponding accessibility and exactly how they can and should correlate to your legacy data and processes. • Properly cleanse and otherwise prepare your data for conversion to the ERP. • Speed up interface development based on knowledge of the system metadata. • Lower overall costs associated with ERP planning, implementation, project management, and conversion. Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  10. 10. ERP Considerations (page 10 of 11) Understanding the ERP metadata structures enables you to develop and apply effective, engineering-based approaches as opposed to the brute force approaches adopted by many organizations. Extracting the metadata of ERP systems permits organizations to more rapidly gain a fact-based understanding of their ERP challenge. For organizations that want to avoid ERP implementation discomfort, ERP metadata can be used by the organization for a variety of purposes ranging from gaining an architectural perspective to accomplishing physical data mapping and migration. The figure on the right illustrates the basic processes homepages menugroups metadata architecture for a version of a major (39) (7) (8) (41) (8) enterprise resource planning (ERP) vendor's (182) (86) package.7 It is interesting to understand what components (180) (949) stepnames (822) menunames (86) this simple metamodel tells us about this (847) (281) vendor's system. panels menuitems menubars (1421) (1149) (31) (1916) (1259) • Direct interactions happen between four (25906) (5873) primary components – users select fields (7073) records (2706) (264) parents (264) MENUITEMS to navigate to PANELS in order (708) (647) to access data from specific FIELDS and (347) (647) RECORDS. reports children (347) (647) • Specific steps must be executed in order to achieve access to the desired panel – literally hundreds of MENUITEM and business process STEPNAME combinations are required! • Having access to this level of detail makes it possible to determine MENUITEM navigation information required to implement a specific business process STEPNAME. • It is also possible to obtain detailed descriptions of the vendor's perspective on specific business processes. • Typically, only a subset of the total number of records is addressed by reports – usually indicating a requirement for additional report writing capabilities. The dotted line connecting RECORDS, REPORTS and PANELS did not exist in the operational system; IDR created these lines by uncovering linking metadata and increasing the ERP metadata value. IDR has developed baseline metadata descriptions of multiple ERP applications. This "out of the box" metadata has proven to be the key to effective strategic systems implementation and has proven to have considerable value. These valuable metadata assets often aid in jump-starting new 7 Extracted from Peter Aiken et. al. Reverse Engineering New Systems for Smooth Implementation. IEEE Software. March/April 1999 16(2):36-43.and other sources copyrighted by the author. Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056
  11. 11. ERP Considerations (page 11 of 11) initiatives, restarting stalled projects, and helping organizations avoid unnecessary pain and cost. Understanding the metadata of both your legacy systems and your desired ERP target can save the implementation effort in at least five ways: 1. Cost: Compared to the standard conversion practice, the metadata solution has saved an order of magnitude in total conversion costs, to include significant amounts of employee time and resources. 2. Efficiency: Metadata assets enable the implementers to greatly outperform other analysts using manual stand-alone analysis approaches. 3. Effectiveness: By approaching data migration with an engineering-based solution data quality measures are able to become integrated conversion process components. Not only is the initial investment less costly, but its payoff increases with respect to maintenance and upgrades. 4. Communication Administer Workforce Effectiveness: While not Recruit Workforce (62%) really a separate category from #3 above, Manage Competencies (20%) management is able to Plan Successions (~5% Develop Workforce (29.9%) Administer Workforce (28.8%) make more informed Compensate Employees (23.7%) Monitor Workplace (8.1%) decisions when the ERP Administer Training (~5%) Define Business (4.4%) implementation is presented Plan Careers (~5%) Target System (3.9%) EDI Manager (.9%) using fact-based metadata Target System Tools (.3%) (see figure at right Manage Positions (2%) illustrating the relative 0 500 1000 1500 2000 2500 complexities of various software modules). 5. Risk Mitigation: three primary categories include: a. Error reduction: Human interpretation errors are significantly reduced and through-put is increased using our metadata extraction technologies. b. Conversion measures: Instead of applying a cookie cutter approach to your ERP challenges, our fact-based analysis has enabled us to optimize for tasks variations as the target analysis focus changes. c. Low implementation cost: Proof of concept leading directly to business cases requires a minimal investment and can return a positive cash flow. Only through careful planning of a metadata-based implementation strategy can organizations hope to avoid standard ERP pitfalls. Institute for Data Research • Maggie L. Walker Business and Technology Center • 501 East Franklin Street – Suite 522 • Richmond, VA 23219 • • +1.804.521.4056