Why ERP Migrations Fail

  • Smaller Small Medium Big Bigger
  • Default Helvetica Segoe Georgia Times

Sorry, but there is no modernization silver bullet. Migrating does not equal modernization.

Not all ERP migrations fail. But many do. I’d even say the majority do...and that’s what Gartner says too. In their estimation, 75 percent of all ERP projects fail. In a bold prediction a couple years ago, they estimated that 90 percent of all ERP cloud projects, what they call “postmodern application integration strategies,” would fail between 2016 and 2018. More specifically, “Gartner predicts that through 2018, 90 percent of organizations will lack a postmodern application integration strategy and execution ability, resulting in integration disorder, greater complexity, and cost.”

In that press release, Gartner Vice President Carol Hardcastle said, “This new environment promises more business agility, but only if the increased complexity is recognized and addressed per se. Twenty-five or more years after ERP solutions entered the IT applications market, many ERP projects are still deeply compromised in time, cost and, more insidiously, in business outcomes. Today, organizations need to resist the easy temptation to succumb to pressure from business leaders to get started before the organization is really ready, and without a business-agreed ERP strategy. CEO's, CFO's and business leaders must understand what it will take to ensure success.”

Part of what I do on a daily basis is sales-related. I have a hybrid role: Most of my day is spent doing technical work (IBM i upgrades, PTF installation, tech support for managed-services customers, security projects, and whatnot). The other part is spent working with customers or potential customers in my neck of the woods as a salesperson. Every now and again, I will call a company that I know runs IBM i and talk to the IT director only to find that they’re looking at moving to a new ERP and that they’re “going to be off the AS/400 in a year.”

We can imagine how that one plays out, right?

I recently spoke to a customer who has a five-year plan for an ERP migration. The customer understands it’s not a quick fix; it will take a lot of blood, sweat, and tears to move their organization ahead from their status quo. At least they’re being realistic about it.

I’ve been fortunate to be a part of a couple of ERP migrations over the years. One was highly successful, and one was a failure of epic proportions. I’ve been involved in countless data migrations and modernization projects. Some good, some bad.

An ERP migration can fail for many reasons, and perhaps that’s why it’s hard to put a finger on both what to prepare for in advance and (more often) what went wrong when the project is examined in the postmortem.

In no particular order of importance, here are a few anecdotal tips that will help keep any project you have in good order and hopefully result in success.

1. Buy-in is key. Projects that succeed involve everyone who should be involved, from the shop floor staff to the CEO, and that’s in order of importance. The production floor, shipping dock, and warehouse folks need to be asked for their opinion. I’m a firm believer that, in manufacturing, your shippers are the most important personnel in the organization as they’re the last to see the product before the customer does. Their opinions are what counts in terms of technology usability and overall common sense.

2. Excessive oversight can be a productivity killer with a top-heavy project. By “top-heavy,” I mean that it’s largely driven by the top of the organization. I was once on a project team in which I gave some brutal truth to a consultant hired to run it. This consultant wanted to achieve a goal, and my informed opinion, based on current staff workload, was that IT could not deliver in the time frame mandated by the consultant. So he requested hourly in-person updates on progress on a goal that took three weeks to achieve (the original mandate was three days). This type of oversight, designed to light a fire, took about 15 minutes per hour, or two hours per day, away from actual work.

I once had a project manager during a data migration request IT provide a tool that would allow a file-by-file and field-by-field comparison between approximately 2,000 tables, which would allow him to visually verify the migrated data was accurate. The tool would dump hundreds of fields and millions of records to a file from which he’d download batches into Excel for perusal. I remember thinking, “You’ve never done this type of migration before” (he hadn’t), “You must be paid by the hour” (he was), and “What an amazing waste of processor and memory this is going to be” (it most definitely was). In the end, IT spent a week building the thing, and after it was used exactly four times (we made sure to put a counter on it), it was never looked at again.

3. That leads me to the old mantra: Listen to your engineers. If you haven’t read this little rant at http://listentoyourengineers.com/, then it’s a good, entertaining read. Long story short, you have qualified people. They know the business, and they know the culture. They have a vested interest in the success of your company. Trust them to give you accurate information. If you don’t want or like the answer you get, the worst thing to do is ask elsewhere until you get the answer you’re looking for. Sure, get a second opinion...but don’t take a conflicting opinion as gospel. Use it to inform yourself, ask questions, and challenge your people to find solutions that everyone can accept.

ERP: Success or Failure?

Given that ERP migrations are notoriously unsuccessful, perhaps the best question we can ask ourselves is “What will it take to modernize our existing systems and leverage the investments of years past?”

So why do ERP projects fail? In my opinion, they’re still seen as a silver bullet, a Hail Mary—almost as much as viewing modernization as a goal rather than a strategy or mindset that it should be viewed as. It’s still black-and-white, on-and-off thinking.

When cloud computing was just taking off, how many articles told us that we needed to get our apps to the cloud or be left behind? Those articles were designed to shame you into making a hasty move. “What, you still have an IT department? You still pay for staff? You still have servers? Ha! You’re a sucker! We’re all moving ahead, and you’re left behind back in the ’70s!” And it really worked until people started seeing that cloud computing is a deployment tactic, not a strategy. Cloud has its place, and despite years of being marketed as a silver bullet, companies realized they need to rethink how cloud fits them best in a cost/value trade-off…like any other technology, such as ERP.