The horror stories of bad ERP implementations have been well documented. Annual surveys indicate that new ERP’s never quite measure up to expectations. Between 50-60% of companies do not realize the benefits expected. A similar percentage experience material disruptions to their business upon implementation. Close to 20% are deemed project failures.
The following three cases represent only some of the dysfunctions I have witnessed from ERP mistakes.
- Company A – A large multinational distributor with warehouses on 4 continents (operations in over 30 countries) instituted a new ERP. One feature of the new system was that it would use algorithms to automatically generate POs for replenishment orders. The company had a long lead-time in its supply chain (90-120 days), and after inventory ballooned by over 20%, it realized the algorithms had issues, yet more inventory was still on the way. Due to the unique nature of the company’s business, inventory turns for its more than 10,000 SKUs were relatively slow for a distributor (4-5x on average), so this excess inventory would not bleed off quickly. This increase in inventory, combined with escalating term loan payments, created a liquidity crunch. The company had to resort to manually reviewing each PO while it was trying to debug the algorithms. This particular problem took over 6 months to get a handle on, and even longer to completely solve.
- Company B – A small metal-bending company had installed a new ERP within the last year, hoping it would yield all sorts of benefits. The company had been operating at a loss, so paying for the ERP further eroded the company’s liquidity position (the owner was not collecting a paycheck at the time). Due to the lack of funds, they did not invest enough in training, so employees did not know how to properly operate the new system and therefore became less productive. This conversion was like pouring gasoline on a fire. The company spent money, yielded no efficiency improvements and caused its workforce to become less productive. In this instance, burning the cash in a bonfire would not have set the company back as far.
- Company C – A $400 million manufacturer with facilities in 3 continents (7 countries) had installed a financial reporting system within the last two years. This system had a forecasting module, so nothing was done in Excel (not necessarily bad). As the company was starting to underperform, we were reforecasting the income statement. This exercise involved making the assumptions and then waiting for the company’s personnel to do their magic. This process was taking entirely too long (hours) to get a revised forecast. Finally we found out what was required. The system was configured so that a new forecast was based on variances from the original budget, which had to be manually keyed to arrive at the new forecasted levels. So, for a 5% sales reduction, someone had to take a calculator to determine the adjustment, and then manually punch it in. Needless to say, as time was of the essence, we aborted the system and moved everything back to Excel.
Here are some things to consider if you are entertaining an ERP conversion:
- Engage your current ERP provider – It may be worthwhile to have a re-assessment with your current ERP provider. By having them review how the company is using the present system, you can get a better idea if the problems are related to the system, or how the business is interacting with the system.
- Create a comprehensive requirements definition – Make sure that each area of the business has established a thorough list of what functionality is needed in an ERP. Care must be taken to ensure that all current functionality is included and that it doesn’t turn into a massive wish list (wants vs. needs).
- Minimize Customization – One of the first things done after selecting a new software is to get it to “fit” inside the company’s way of doing things. The thought process being that the company is unique and the software isn’t designed for the company’s specific needs. While some customization is valid, many times, the customization is a work-around for the company’s bad processes. Through customization, which requires a lot of coding and testing (i.e. lots of $$), the company then immortalizes its bad way of doing things into a program that was perfectly functional out of the box. Instead of using the word customization, try thinking of it as bastardization instead.
- Be realistic – Any good plan needs to be grounded in reality. Talking to many different software providers, consultants, and current customers, is probably the best way to get informed and to ask the right questions. Try to learn from others’ mistakes so you can have a good plan when it comes to implementation costs, human resources (internal and external) and timing.