ERP replacement is rarely triggered by a failure. The system still runs, still produces invoices, still closes the month. What has happened instead is that the organisation has quietly built a layer of human effort around it, and the cost of that layer has grown past the cost of replacing the software — but because it is distributed across people's ordinary work, nobody adds it up.
The clearest signal is the shadow spreadsheet. If the numbers people actually use for decisions live outside the ERP — a production plan in Excel, a commission calculation someone maintains, a sales report assembled by hand each month — then the system is not the system of record, it is a place where data is entered. That is worth measuring rather than tolerating: count the hours per month spent producing information the software was purchased to produce.
The second is that changes have become impossible. A business asks for a modification and is told it cannot be done, or can be done at a price and timeline that makes it not worth asking. Frequently this is because the system was customised so deeply years ago that it can no longer be upgraded, so it sits on a version that is out of support and cannot receive the change. Once you are on an unsupported version with unsupportable customisations, the decision has effectively been made; the only question is when.
The third is compliance drag. Indian statutory requirements move — e-invoicing thresholds, e-way bill rules, the Invoice Management System, GST format revisions — and a supported system receives those as updates. An unsupported one receives them as projects, or as manual workarounds, and each new requirement is another reason someone maintains a spreadsheet. If your finance team is meeting statutory obligations despite the ERP rather than through it, the software is now a liability rather than an asset.
Two further signals worth noticing. Integration refusal: if connecting the ERP to anything modern — a marketplace, a logistics partner, a payment gateway, a business intelligence tool — is quoted as a major project every time, the system's architecture is the constraint on everything else you want to do. And knowledge concentration: if one person understands how it works and the vendor who implemented it no longer exists or no longer supports that version, your continuity risk is not theoretical.
None of this argues for replacing quickly. ERP replacement is expensive, disruptive and frequently disappointing, and a system that is merely old and unloved is not a reason. The threshold worth applying is arithmetic: add up the annual cost of the workarounds, the manual reporting, the compliance patching and the changes you cannot make, and compare it with the cost of replacement amortised over five years. Organisations that do this honestly usually find the answer is clear in one direction or the other — and that they have been avoiding the calculation precisely because they suspected which way it would go.
