The Real Reason Most ERP Modernization Projects Go Over Budget — And How to Fix It Before You Start

PwC and Palantir's newly expanded alliance specifically targets a pattern that shows up in ERP transformation projects across industries: implementations that go over budget or fail outright not because the chosen software was wrong, but because the underlying data wasn't clean, consistent or well-understood when the project began. This is a well-documented risk, and it's worth understanding why it happens and what actually addresses it, regardless of which vendor or systems integrator you work with.

Why Data Quality — Not Software Selection — Is the Real Risk

An ERP system is fundamentally a structured way of organizing and enforcing rules around your organization's data — inventory, financials, customer records, supplier relationships. When that underlying data is inconsistent, duplicated, or riddled with legacy workarounds accumulated over years of using a prior system, the new ERP doesn't fix those problems automatically — it often surfaces them all at once, mid-implementation, at the worst possible time to discover them. Migrating clean, well-understood data into a new system is a fundamentally different (and far more predictable) project than migrating data whose actual state is unknown until the migration itself reveals it.

How to Address Data Quality Before You Start, Not During

• Run a data quality assessment as its own distinct project phase before committing to an implementation timeline — treating data cleanup as something that happens 'during' migration is the single most common source of budget overruns
• Identify and document every legacy workaround and manual process currently compensating for gaps in your existing system — these are exactly the hidden dependencies that surface unexpectedly during migration
• Assign clear ownership for data quality decisions before implementation begins — someone needs the authority to decide how ambiguous or conflicting legacy records get resolved, rather than that decision falling to whoever is doing the migration work under time pressure
• Use automated data profiling tools to quantify the actual scope of data quality issues (duplicate records, missing fields, inconsistent formats) rather than relying on anecdotal awareness of 'some data problems'
• Build a realistic timeline buffer specifically for data remediation, separate from the core implementation timeline, since the scope of data issues is rarely fully known until you start actually looking

The Practical Takeaway

Before committing to an ERP modernization timeline and budget, invest in a genuine data quality assessment as a distinct, earlier phase — not a checkbox item folded into the broader implementation plan. The organizations that stay on budget are consistently the ones that know the real state of their data before migration begins, not the ones that discover it during the process.