Strategy · 2026

SAP or IFS? How to choose an ERP integration path you won’t regret

ERP decisions fail in the mirror more often than in the market: most disasters trace back to how companies chose, not what they chose.

The uncomfortable numbers

The landmark McKinsey–Oxford study of large IT projects found they run 45% over budget on average while delivering 56% less value than predicted. Across two decades of research, more than 70% of ERP implementations fail to achieve their original business objectives. The consistency of those numbers across industries and eras points to one conclusion: the root causes are organizational, not technical.

Customization is where budgets die

The working rule is that standard functionality should cover at least 80% of your processes, with customization reserved for genuinely differentiating capabilities. In practice, organizations customize 40–60% of their ERP — usually out of resistance to changing a familiar process, not strategic necessity. That over-engineering is a primary driver of budget overruns, delays, and the long maintenance tail that quietly erodes ROI for a decade.

Why the failures are organizational

The pattern behind the numbers is remarkably stable. Panorama’s industry research puts the top causes of budget overruns at underestimated project staffing (38%), scope expansion (35%) and technical or data issues (34%) — none of which is a property of the software. Twenty years of improving ERP technology has not moved the failure rate, which is the strongest evidence that the root cause never was the technology. Projects fail in the mirror: unclear process ownership, data nobody cleaned, decisions quietly delegated to the vendor.

SAP or IFS — the honest answer

Both are serious backbones; the differences are in fit. SAP’s strength is deep, standardized multi-plant finance and logistics at scale; IFS is particularly strong in asset-heavy, project- and service-driven operations. But the deciding question is not a feature list — it is which system’s standard your organization can adopt with the least customization. Choose the backbone whose standard processes fit your operating model, plan the integration landscape (MES, quality, maintenance) before signing, and hold the customization line at 20%.

The selection discipline, step by step

A decision you won’t regret follows a sequence. First, map your operating model — order-to-cash, plan-to-produce, procure-to-pay — as it actually runs, not as the org chart claims. Second, score each candidate’s standard processes against that map and quantify the customization gap in modules and man-days; the honest number decides, not the demo. Third, draw the integration landscape — MES, quality, maintenance, logistics — before signing, because integration is where “on budget” quietly dies. Fourth, budget total cost of ownership: implementation, training and the ten-year maintenance tail, not the license line.

And hold the 20% customization line in writing. Every exception should have to argue that it is genuinely differentiating — not merely familiar.