Why ERP and EPM Implementations Fail (And How to Get Ahead of It)
Right now, ERP is one of the biggest conversations in the transformation market. Budgets are being approved, boards are signing off multi-year builds, and finance and technology leaders are finally getting the green light on projects that have been sitting on a roadmap for years.
Which makes it a good time to ask an uncomfortable question: why do so many of these projects still go wrong?
Not because the software is bad. In our experience working alongside finance and technology leaders through these builds, it’s almost never the platform. It’s what happens around it – the team delivering it, the plan behind it, and the groundwork nobody wanted to slow down for.
The data backs this up. Panorama Consulting Group’s research puts the overall ERP failure rate at around 68%, and Gartner has gone further, predicting that by 2027 more than 70% of recently implemented ERP initiatives will fail to fully achieve their original business goals.¹ ² Those aren’t fringe numbers. That’s the majority.
Here’s where it tends to go wrong.
1. The team you’re pitched isn’t the team you get
Every system integrator sells you their best people. It’s the pitch that wins the deal – a room full of senior, credible consultants who clearly know the platform inside out.
Then the contract’s signed, and the reality of resourcing kicks in. Those senior consultants are already committed elsewhere, and the delivery team that actually turns up is a different one entirely – less experienced, less specialised, still capable on paper but not the team you bought.
It’s one of the most common gaps in this market, and one of the hardest to spot until you’re already mid-build.
2. Budgets that were never realistic to begin with
Programme sponsors are frequently asked to deliver against a number that was set before anyone properly scoped the work. Not because the plan was reckless – usually the opposite. It’s a shoestring budget approved with good intentions, and a programme director now trying to make it stretch further than it should.
Underfunding shows up later as scope cuts, corner-cutting, or a system that technically goes live but was never resourced to be built properly.
3. Timelines that were built to look good, not to work
Rushed builds create their own problems. Compress a programme that genuinely needs a year into eight months, and something has to give – usually testing, training, or the parts of the build nobody notices are missing until go-live.
4. Skipping proper discovery
This is the one that quietly undoes everything else. Most ERP failures can be traced back to the very start of the process – the discovery phase that gets rushed or skipped because everyone wants to move on to picking the system.
That includes something as fundamental as choosing the right tier of system for where the business actually is, and where it’s heading. A business scaling toward £500m in the next decade needs a different platform to one that’s staying lean. Without a proper discovery and requirement-gathering phase, that mismatch gets built in from day one, and some research puts the proportion of ERP failures that trace back to this initial phase at over 60%.³
5. Underestimating change management
If we had to pick the one that gets missed most often, it’s this.
There’s a common assumption in this market that anyone who communicates well can do a change role. In practice, strong change and communications specialists are a specific skill set, and they’re the ones who directly influence whether a new system gets adopted or quietly resisted.
Without them, the technology can be exactly right, and the programme can still fail – because the people who are meant to use it never really buy in. Industry analysis backs this up: inadequate change management is consistently cited as the single biggest contributor to ERP failure, involved in more than 4 in 10 failed projects.⁴
What good looks like instead
The businesses that get this right tend to do two things differently.
First, they treat discovery as the project, not a delay to it. Understanding what the business actually needs (now and in three to five years) before committing to a system and a scope.
Second, they think differently about how delivery is resourced. The traditional system integrator model asks you to hand over ownership entirely: agree a number, sign the contract, and trust that the team behind it delivers. You rarely see who’s actually doing the work, or what it’s costing to get there.
There’s a different way to do it – one where the client keeps visibility over exactly who’s working on their programme and how. Rather than a fixed team assigned by availability, resourcing is drawn from a wider bench of vetted consultants matched to what each stage of the build actually needs. It’s a model built around transparency over the life of the programme, not just at the point of sign-off.
The common thread
None of the five points above is really about technology. They’re about the decisions made before a single module goes live – who’s actually resourced to deliver it, whether the budget and timeline reflect reality, and whether the business took the time to understand what it needed before it decided what to buy.
That’s genuinely good news, because it means the failure rate isn’t fixed. Every one of these is avoidable with the right planning and the right people in place early enough to matter. The businesses that get ERP and EPM right aren’t the ones with the biggest budgets or the flashiest systems – they’re the ones that treat the groundwork with the same seriousness as the go-live date.
Where to start
If any of the above sounds familiar, or you’re heading into a build and want to know where the risk actually sits, Cedar’s ERP/EPM benchmarking tool is a good first step. It takes a few minutes to complete and gives you a clear picture of where your organisation currently sits on its ERP or EPM journey – and what to focus on next.
[Link to benchmarking tool]
Sources
- Panorama Consulting Group, 2025 ERP Report
- Gartner research, cited via kreativecoretech.com and concorderp.com ERP statistics roundups, 2026
- Academic research on ERP requirement-gathering and system selection failures, cited via smarttransformationsolutions.com, 2024
- Godlan 2025 ERP failure analysis, cited via concorderp.com, 2026
