Praneet Tiwari: Praneet's blogs

1. Focus on business processes and requirements first:

 

It's easy to get dazed with the amazing set of capabilities of an enterprise software. None of that really matters. Concentrate on what you want to do and not on what all can be done. Get you requirements right and then choose the solution rather than choosing an ERP software first and then thinking what now.

2. Do not rush, acknowledge that most software projects have slippage :

In 1999 Hershey was rushing to meet an immovable deadline, since it was trying to get off legacy systems that might not continue to work correctly as of Jan. 1, 2000. The project was planned to go live in April, an inactive month in candy business. Inevitably, there were slippages leading to postponment till October. This led to system going live in the busiest month - October. Certainly a bad time to bring in major changes.

3. Go Piecemeal, a bit at a time:

In an enterprise software it is almost always a bad idea to introduce major changes at one go. Train your staff, introduce change in one department and then move on till the entire organization gets the cake.

4. Plan properly, make sure it's a positive NPV project, include post-implementation performance measurement:

This requires doing more than just developing a high-level business case to get approval from upper management or your board of directors. It also entails establishing key performance measures, setting baselines and targets for those measures, and tracking performance after go-live. This is the only way to truly realize the benefit potential of ERP.

5. Don't load your project with too many goals:

Hershey wanted the ERP solution to meet its Y2K needs as well. From a cost perspective this looks like an effective solution. But to make this solution realize the light of day, one needs to plan well in advance. Overambitious projects have a little chance of success.

6. Commitment from company executives:

Any project without support from it's top-management will fail. Support from a CIO or IT Director is fine, but it's not enough. No matter how well-run a project is, problems arise (such as conflicting business needs), so the CEO and your entire C-level staff needs to be on board to drive some of these.



7. Devote more time up front rather than jumping in straightaway :

An ERP vendor's motive is to close a deal as soon as possible. Yours should be to make sure it gets done right. Too often, companies jump right in to a project without validating the software vendor's understanding of business requirements or their project plan. The more time you spend ensuring these things are done right at the beginning of the project, the less time you'll spend fixing problems later on.



8. Ensure adequate training and change management:

ERP systems involve big change for people, and the system will not do you any good if people do not understand how to use it effectively. Therefore, spending time on money on training, change management, job design, etc. is crucial to any ERP project.



9. Make sure you understand why you're implementing ERP:

This is arguably the most important one. It's easy to see that many big companies are running SAP or Oracle and maybe you should too, but it's harder to consider that maybe you don't need an ERP system at all. Perhaps process improvement, organizational redesign, or targeted best-of-breed technology will meet your business objectives at a lower cost. By clearly understanding your business objectives and what you're trying to accomplish with an ERP system, you will be able to make a more appropriate decision on which route to take, which may or may not involve ERP.

10. Value the Data:

Make sure data provided to the ERP program is accurate. In Hershey's case, there were various unrecorded warehouses and the program had no way to find out the exact inventory position. Data management is one of the critical ingredients of a successful ERP project.