Don’t know what to do with your legacy system?

Legacy systems often contain functionality not found in ERP systems, and would be expensive to replace. As well, the legacy system often gets the job done, so many would argue that "if it’s not broken, don’t fix it".

Introduction

On the other hand, legacy systems lack the improvements made to software over the years such as a graphical user interface, ad-hoc reporting tools and easy customizations such as adding user defined fields. As well, the legacy systems are often maintained by aging boomers, who may not be readily available now and will certainly be less available in the future.

The legacy system could be enhanced by putting on a graphical user interface or screen scraper. If the underlying database can be accessed directly or through tools such as ODBC, a modern report writer can be deployed. Another possibility is simply moving data to a data warehouse and using Business Intelligence tools to get at the information.

A compromise solution is a best of breed approach maintaining some components of the legacy system. The legacy system may contain strategic functionality that enhances an organization’s differentiation from the competition. The strategic component could potentially be maintained and non strategic components could be replaced with an ERP system. There are many tools available for integration between the legacy system and the ERP system. But don’t be fooled by anyone who says integration is a piece of cake because of a certain technology such as SOA (Service-Oriented Architecture). There can be multiple points of integration between systems. The simple case is transferring a transaction from one system to the other. It gets more complicated when trying to synchronize master files such as inventory or supplier files. Is the synchronization in real time? Is it two-way? Does one system take precedence over the other if both systems have changed the same data?

Sooner or later, the legacy system is going to be replaced. The question is when. Today’s difficult economy can be considered the ideal time to replace the legacy system. One of the biggest challenges in the implementation of an ERP system is getting the right people spending enough time on the project. If things have slowed down because of the economy, employees should have more time to devote to an implementation of a new system. But even if there is incentive and time to replace the legacy system, a cautious approach is still advised. The implementation of a new system is a challenge at the best of times.

A cost/benefit study should be conducted on all the alternatives. The cost/benefit study should be conducted by someone that does not have a vested interest in the outcome of the study. The cost/benefit study should contain a 5 year projection of all costs and savings with calculations for ROI (Return on Investment) based on NPV (Net Present Value) as well as payback and IRR (Internal Rate of Return). These calculations are the easy part.

So what do you do with an ERP system that has lots of problems that you don’t want to replace? Many reluctantly live with the inefficiency and ineffectiveness. Others want to make changes but where do you go for help? The vendors are eager and capable to help but the perception is that their recommendations will be self-serving. It’s a shame to do nothing as there are many ways to extend the useful life of the existing ERP system including system upgrades, improving business process, improving infrastructure, 3rd party solutions, customization, integration, work flow and business intelligence and reporting tools.

Requirements Analysis

By providing potential vendors with requirements and user counts, it is possible to get reasonable cost estimates to replace the legacy system or portions of it. In order to get good numbers from the vendors, requirements need to unambiguous. As well, the vendor must not be allowed to respond with a Yes or a No to each requirement. A Yes could mean many things including out-of-the box functionality, third party, customization or work around. The vendors are forced to be honest in their responses if the requirements are not unambiguous. They know that the ultimate decision is based on trust and if they misrepresent their software, they will lose trust and have no hope of being awarded the contract.

Requirements gathering starts with defining scope, critical success factors (CSFs) and key requirements. CSFs are defined as what an organization must do well in order to be successful. Mapping key requirements to CSFs helps ensure that the most important requirements are identified. Next interviews should be conducted with employees to obtain a good understanding of the current system and As-Is Business Process. The expression the devil is in the details applies to requirements gathering but don’t waste time on the basics. If a low entry system such as QuickBooks can do it, there should be no need to document the related business process or requirement.

Business Process Review

The reason for documenting business process is that changing systems is the best opportunity to improve business process. By documenting the existing process along with the current problems will be extremely useful in subsequent steps. The As-Is process could be used as a script should a detailed demonstration by vendors be required. The vendor could be given the business process, the problems, requirements and sample documents. The vendor could be instructed to show or explain how they would overcome the existing problems.

Creating the optimal or To-Be Business Process is not recommended at this time for what would be considered common business processes that other organizations would share. It may be possible to leverage existing business processes from vendors and there is no point in reinventing the wheel.

The As-Is business process can be used as a starting point in obtaining benefits. The business process document should contain the existing problems and their impact. If, for example, a problem causes a clerk to waste 30 minutes per day and there are 10 clerks doing similar work, then it’s easy to extrapolate what the savings would be.

Non Value-Added Activities

Another way to determine the potential savings in improving operational efficiency is to conduct a survey on the time employees spend on all non-value-added activities such as re-keying information or reconciling information between systems. The survey should be confidential and conducted by an external party to reduce the risk that employees will avoid telling the truth.

But be careful with the potential savings identified by eliminating non value-added activities. It can be a big morale problem if employees think their jobs are at risk. Often, employees are not laid off, but the company can grow without adding new people. As well, rather than spending time on non value-added activity, employees can be involved in more useful activity such as analysis or talking to customers.

Some benefits are easy to quantify such as eliminating the costs to support the legacy system. Other benefits are very difficult to quantify such as improved decision making. But it is possible to quantify these intangible benefits. For example, it might take two employees to develop custom reports for decision makers. With some business intelligence tools, the time spent to create some reports may be eliminated.

Business Case

When all the costs and benefits have been identified and validated, a cash flow statement such as the following can be prepared. ROI (Return On Investment) is the key metric decision makers want to see. When computing ROI over long periods of time, it is recommended to calculate the (NPV) Net Present Value of the inflows and outflows. The inflows include savings from various departments due to eliminating inefficient business processes. The outflows contain paying relief staff hired or subcontracted to support SME (Subject Matter Experts) that are too busy with the implementation to perform their normal activities.

A business case does not need fancy graphs or complicated algorithms. Excel includes a formula for NPV and ROI is a simple calculation = (Inflow-Outflow)/Outflow. Most importantly make sure that all the numbers are substantiated. And don’t be afraid to show a negative ROI. Sometimes the intangible benefits will suffice. Sometimes the business case is that there is no choice if the legacy system can no longer be supported.