← Journal

Why ERP programmes fail to embed

And what you can actually do about it - before the programme even starts

ERP & Transformation · 18 June 2026 · 4 min read

Why ERP programmes fail to embed

Here's a stat that should stop any executive in their tracks: between 55% and 75% of ERP implementations experience cost overruns, schedule delays, or fail to deliver what they promised.

That's not a niche problem. That's the norm. And yet organisations keep launching these programmes in the same way, making the same decisions - or avoiding the same decisions - and wondering why the system they spent millions on is sitting half-used two years later.

The frustrating truth is that most ERP failures aren't technology failures. The software works.

The risk is almost always what happens around the software: unclear goals, messy data, people who weren't brought along for the ride, and decisions that got deferred until they became problems.

The good news? Almost all of it is preventable - if you're willing to make some calls before the programme kicks off.

The six ways ERP programmes come unstuck

It's rarely one thing that kills an ERP programme. It's a slow accumulation of avoidable problems, each one manageable on its own, but collectively fatal.

The usual suspects are well-documented: scope that creeps because nobody defined what the programme wasn't going to do; a 'big bang' go-live where everything switches on at once and any problem becomes everyone's problem simultaneously; change management not treated as a proper work stream; data that turns out to be far messier than anyone admitted; an executive sponsor who only showed up when he/she decided; and a methodology that doesn't actually suit the type of work being done.

The pattern across all of these failures is the same: problems that started in the earliest phases of the programme compound over time until they're too expensive or too late to fix. Which is exactly why the solution has to start before the programme formally begins.

The decisions you need to make before day one

Define what success looks like - before the vendor is selected, the consultants arrive, or the plan is built. And alongside those KPIs, produce an explicit list of what the programme will not deliver. Scope creep doesn't sneak in through the back door; it walks in through the front because nobody thought to lock it.

The next non-negotiable is a 'Vanilla First' commitment. Every ERP programme starts with this noble intent – but then quickly generates a queue of customisation requests from people who want the new system to work exactly like the old one. Resist it. Every customisation adds cost, time, and future upgrade headaches. Process change is almost always faster and cheaper than custom development - and it actually improves how the business operates.

Change management isn’t an initiative you bolt on - it's how you ensure the business actually uses what you've built. People who've been involved from the start are stakeholders. People who receive a training invitation six weeks before go-live are obstacles.

Then there's data - start cleaning your data before the programme begins. The data mess doesn't disappear if you ignore it; it just gets discovered at the worst possible moment.

Appoint your Change Management Lead early - even before the programme start - so that they have the opportunity to work with the business to define that golden thread that will make engagement smoother and quicker.

Structures that prevent future failures

Getting the pre-programme decisions right creates the conditions for success. But a few durable structures will protect those conditions through the messy reality of delivery.

A Steering Committee with real authority is not the same as a steering committee that receives status updates. The difference is the power to descope, delay, or throw more resource at the programme when needed. ERP programmes constantly generate decisions that cross functional lines - finance versus operations, IT versus the business - and without a body that can resolve those conflicts quickly, they fester.

Adopt a phased go-live. Start with a Minimum Viable ERP - core financials, one business unit, one geography - and learn before rolling out more broadly. The first wave is never perfect; far better to discover that when it affects 20% of the business than 100%.

Plan your hypercare period as carefully as you plan your go-live. Too many organisations demobilise the programme team on a calendar milestone rather than when the business is genuinely stable. Agree upfront on what 'stable' means - and don't exit hypercare until you hit them. Under-resourced post-go-live support is one of the most reliable ways to turn a difficult implementation into one that doesn’t actually reap the benefits of the new ERP solution.

And when it's all done? Measure whether it worked. Go back to those KPIs you defined before the programme started and check at 3,6,12 months. It sounds obvious. Almost nobody does it. Without that measurement loop, you can't learn, you can't improve, and you can't make the case that the investment was worth it.

Successful ERP implementation is a business transformation, not a technology project. The organisations that get it right treat it that way from the very first conversation - long before a single line of code is configured, and long before anyone books the go-live party.