KOHb - Getty Images

How to migrate to a new ERP system

ERP migrations can go off track fast. Strong governance and disciplined scope control can help CIOs improve the odds of success.

Executive summary

Gartner experts suggest the following ERP migration tips:

  • Establish strong governance early, including clear decision rights and escalation paths in the first 90 days.
  • Define the scope tightly and be ready to say "no" to requests that don't align with business goals.
  • Prioritize change management and data readiness, as both are major drivers of delays and cost overruns.
  • Anchor the business case in measurable outcomes.

Few IT projects test CIOs as much as ERP migrations do.

ERP migrations rank among the most complex and costly initiatives CIOs lead. They affect core business processes, require significant data transformation and demand tight coordination between IT, business units and system integrators (SIs). Success often depends less on the technology itself and more on early decisions around scope, data readiness, change management and how well organizations define measurable business value. CIOs must also balance competing pressures and find ways to say "no" without losing stakeholder buy-in.

In the following interview, Gartner partners Jackie Swanson and Andrew Hess share how CIOs can stay within budget, avoid common pitfalls and structure ERP programs to support business goals and enterprise-wide adoption.

Editor's note: The following transcript was edited for length and clarity.

What is your role at Gartner, and how does it relate to ERP migrations?

Jackie Swanson: We're partners in Gartner Consulting. I sit more on the industry side, and Andrew sits more on the technology side. We help clients with governance and program management during projects to ensure implementation goes smoothly.

Andrew Hess: I'm a senior expert partner in Gartner. I focus primarily on applications in ERP modernization, including strategy, selection and program setup. I have over 30 years of ERP experience. For about half of it, I was an SI, and the other half, I held industry roles before joining Gartner. So, I've seen both sides of it. What we're talking about is what I do daily with clients.

The first 90 days of an ERP migration are critical. What actions during that window make the biggest difference when it comes to keeping the project on budget?

Swanson: In the first 90 days, how you set up the project plan, guardrails and steering committee -- how you're making decisions and all the upfront work that it takes to organize the actual project -- is the best way to set you up for success. 

Hess: Project governance is the first thing CIOs should think about, just as Jackie described. As organizations are busy negotiating with a software vendor or the SI, they should already be setting up their program governance to know decision rights, escalation paths -- things like that.

Have you seen changes in how organizations manage ERP implementations over the years?

Swanson: I used to work for PwC, and in the past 10 years, I've seen a tremendous rise in people hiring third parties to play the role of moderator between the SI and the client to make sure that everything is on track. This is because so many past projects have been derailed. It's helpful to have a strong program management team in place.

It's easy for scope creep to occur as you go down these big ERP implementations, because it's exciting.
Jackie SwansonManaging partner, Gartner Consulting

What else can CIOs do to stay within budget on their ERP migrations?

Swanson: When you're looking at how to maintain the budget, scope is key. It's easy for scope creep to occur as you go down these big ERP implementations, because it's exciting. You have an opportunity to redesign processes and implement the best ways of working, independent of the system. You can go down the rabbit hole of redesigning the process and then implement the system over that process or implement new modules that you hadn't put in the original budget.

However, you must keep a close eye on that scope. In some cases, it makes sense to expand the scope or redesign the process. But if you do that, there are implications for the timeline and budget, and you want to ensure that the triangle -- scope, timeline and budget -- makes the most sense for the project.

Many ERP projects get bogged down in unnecessary customization. Can you share an example of this?

Swanson: One of the arts of implementation planning is deciding how much of the customization path you'll embrace. It's highly dependent on the culture, budget and currency of your operation. Sometimes, highly customized is the way to go because you're doing something so unique that an out-of-the-box setup won't cut it for a certain function.

But most of the time, clients want to stick to the 80/20 rule. They want 80% out of the box because it's embedded in best practices, and then 20% can be configured -- not necessarily customized -- but configured.

Hess: I'll give an example. This is one that I'm working with right now. It's an industrial company, and we were early in the strategy phase. We were asking leaders what they want to get out of the implementation. This one person involved in sales had built an elaborate spreadsheet to help with forecasting. His idea of success would be that his spreadsheet still works in the future. All I could think was, 'We're trying to get you out of the spreadsheet.'

You don't want to be anchored to how you do things now. I was involved with ERP back in the 90s, and that is what we did. In hindsight, it was like 'Let's configure the new system to do things exactly the way you do them now.' That often leads to suboptimal implementation.

ERP vendors have done a lot of R&D to deliver what they view as best practices, and by and large, they are. So, to Jackie's point, if you can take 80% of it out of the box, you really save yourself a lot of headaches during implementation and in ongoing support and maintenance.

Swanson: The rise of AI has also caused everybody to reevaluate their processes. It's a great opportunity when implementing an ERP to look at how you want to work in the future. What are the core tasks that we can push to AI? What will let us be more strategic?

What metrics should CIOs define before starting an ERP migration? 

Swanson: There are some obvious metrics that we've all been taught, such as 'Is it delivered on time?' and 'Is it delivered on budget?' But we also push clients to track adoption. If you have great change management, you want to know, 'Are people embracing the system?' and 'Does it benefit their daily work life?'

Hess: Think about the business case, too. It's not attractive if your business case is 'we have an old system, and we need a new one.' For example, I work closely with manufacturers, so I focus on things with very clear, measurable value. One of my clients has too much inventory. If there's a way to reduce inventory by, say, 5% -- something that doesn't seem too far-fetched -- when you do the math, that starts to be $100 million or more.

We tell organizations to define clear business cases. There are defined costs. For instance, you know what the software, SI and internal labor will cost. But what are the benefits? The business case can be efficiency-enabled. Often, it's easier if you have hard dollars. Examples could be 'We're going international, and our current system won't let us do that.' What is international growth going to be? Or your inventory reduction? If you're doing order fulfillment and only hitting 80% of it, are you increasing the number of perfect orders on time?

Make sure that the project team is focused on and understands the business case so they can represent the company's interests as they work with the SI.

What's a common mistake you see CIOs make early in migrations that leads to cost overruns?

Hess: We led a discussion at a symposium last fall, and I asked the question, 'Going back, what would you do differently?' A CIO who I'm working with now said he would have focused much more on change management and data -- and much earlier. He did focus on them early, but he would have done more and probably underestimated the level of effort required.

If you take shortcuts around change management, you're undermining everything else you're doing from a build standpoint. Data is probably the one from a timeline standpoint. You're starting data even before you start the implementation. Go look at your customer master and identify duplicates.

The last thing I'll add is scope. The default answer to new scope requests should be 'no.' If you don't control the scope, then you're absolutely going to have cost and timeline overruns.

By saying, 'No, we're just going to do what we have today,' you risk losing some buy-in.
Andrew HessSenior partner, Gartner Consulting

How can CIOs balance pressure from business leaders who want to expand the scope?

Hess: It's difficult. Business partners often see the ERP migration as a great opportunity to introduce things they really want -- and understandably so. By saying, 'No, we're just going to do what we have today,' you risk losing some buy-in.

It comes back to the business case -- what's the goal of this? If the goal is operational efficiency, simplification of the IT landscape or something similar, then many of these needs can be worked through early in the process.

It's important to have open discussions upfront about the capabilities the organization actually needs. Ideally, those requests are aligned with the company's broader strategic direction. You also must find a way to deliver some of them -- especially when they can be provided out of the box. I call those 'freebies.' For example, a better UI. If you're coming off a 30-year-old system, I can guarantee whatever you're putting in now is going to have a better UI. That alone can make people more comfortable with the ERP and reduce resistance.

Swanson: We're also strong advocates for a structured system selection process. It's easy for companies to view that as an unnecessary cost or something they can do themselves, but there's a big benefit to doing it properly. The system is a huge investment, so if you narrow down the capabilities you need, the functions you want and the modules that fit those upfront, then you're saving yourself agony later.

In that initial stage, when you're deciding which of these capabilities and functions you want -- and negotiating for them -- you have a much stronger power to decide what you're going to need, versus when you're in the throes of it, and you're like, 'I just want to add all the functionality, because it looks so cool.'

Is there anything else on this topic CIOs should consider?

Hess: ERP implementations function best when they're not seen as just an IT project. You need active business support and engagement at the leadership level and within the teams. If you have harmony between your business users and IT, you gain a lot of power because they complement one another.

Tim Murphy is an award-winning reporter covering IT strategy for Informa TechTarget.

Dig Deeper on Digital transformation