When is it time to replace a legacy ERP system?

In answering this question, probably the first thing to say is that, when organizations leave a decision on implementing a new system until they need it, they have often left it too late. That leads to everything from rushed (i.e. second-rate) implementations to failed implementations.

The reason is that even a pretty straightforward Tier 3 system can take a year to do properly, whilst Tier 1 typically takes about three years and a complex multi-country can take upwards of seven. These figures are based on the implementation of what we can call 'full functionality' ERP systems; i.e. not just accounting systems with some rudimentary buy/assemble/stock/sell capability tagged on.

Why so long? It is certainly true that some companies implement systems much quicker, but it is also true that over 70% of companies consider their implementations to be a disappointment or an outright failure. Although additional factors can be involved, it is probably safe to say there is a connection. For a successful implementation, a number of things have to be done, and done properly, including:

• Performing a high-level system design to specify what the new system has to achieve and deliver
• Reviewing the market to see what is available within the budget
• Short-listing prospective suppliers and obtaining proposals from them
• Reviewing these proposals and obtaining demonstrations of systems and capabilities
• Performing a gap analysis on the preferred system
• Specifying modifications and/or redesigning business processes to cope with identified gaps
• Obtaining final quotation and checking references
• Interviewing and selecting implementation partners

So the decision to replace a legacy system may need to be taken several years before the replacement is required and, for most organizations, that requires a lot of foresight. There can, however, be triggers that can be used to initiate the decision process, even though some of these can be subjective.

Sometimes a date is imposed on the company because the legacy system is being withdrawn at a specified point in the future. Generally, this happens because support for an older system is being withdrawn, and running an unsupported system is viewed as being too risky.

The same can be true of older hardware and operating systems which can have the added complication of requiring support from a diminishing pool of experienced and knowledgeable talent as people with the required skills migrate to newer options.

Another trigger can be forecast regulatory changes, such as the introduction in some countries of VAT (Value Added Tax) to replace simpler sales taxes. The necessary development work to cope with such changes may not be justifiable in older systems.

Harder to pin down may be changes to the business itself; the most obvious of which might be planned or forecast growth that will increase the number of users, and the number of transactions, beyond limits that the existing system imposes or which can be expected to stretch them beyond a point that feels comfortable.

Lastly; there may be new business paradigms on the horizon, such as increased exposure to e-commerce or the opening of subsidiaries in other countries that will need local language versions of the software that are not currently available.

Advice elsewhere on this website includes the recommendation that the core in-house team for an ERP implementation should never be wholly disbanded but should be retained to ensure that system changes (from everything from upgrades to replacement) can be proactive. When that advice is taken, the team can periodically reconvene to review the likelihood and impact of the above triggers so that expensive surprises can be avoided.