You have spent 18 months and millions of dollars preparing foryour ERP go-live. The system is configured. The team is trained. The go-livedate is locked.
Then day one arrives.
A maintenance technician needs a replacement seal for a critical pump. Hesearches the new system. Nothing comes up. He tries different search terms.Still nothing. He calls procurement. They find three records that might be theright part — but none of them have a manufacturer part number, so nobody can besure. A purchase order goes out for all three. Two are wrong. The right onetakes four days to arrive.
This is not a technology problem. The ERP system is working exactly asdesigned. It is a data problem — and it is playing out in facilities acrossevery industry, every time a new ERP system goes live with MRO spare parts datathat was not properly cataloged before go-live.
The Scale of the Problem Is Larger Than Most Teams Realise
More than 60% of industrial ERP material master records contain incomplete orincorrect data at the point of migration. For MRO spare parts — the bearings,seals, gaskets, instruments and critical components that keep industrialequipment running — the number is often higher.
This is not because the people involved are careless. It is because MRO spareparts data is genuinely complex, and because the process of preparing it for anERP system is consistently underestimated in project planning.
Consider what a single spare parts record actually requires to be useful: acorrect and consistent item description, a verified manufacturer name, a validmanufacturer part number, the right unit of measure, the correct classificationcode, links to the relevant equipment tags, and a range of technical attributesspecific to that type of part. Every field matters. A wrong character in a partnumber makes the record unsearchable. An unverified manufacturer name meansprocurement cannot confirm the supplier. A missing equipment link meansmaintenance cannot find the part when raising a work order.
Multiply that across 50,000 spare parts records — a typical number for amid-sized industrial facility — and the scale of the challenge becomesclear.
This is a problem Sharecat Data Services has spent over 30 years helpingindustrial operators solve. In that time, the team has cataloged more than900,000 MRO spare parts items across oil and gas, chemical, pharmaceutical andutilities facilities worldwide — and the same root causes appear, on everyproject, in every industry.
The spare parts data that needs to go into a new ERP system comes from multiplesources at once, each with its own format, its own quality level, and its ownset of problems.
SPIR forms — Spare Parts Interchangeability Records submitted by equipmentsuppliers — are the primary source of recommended spare parts for new orrecently commissioned assets. They are almost always delivered as PDFs. Notstructured data. PDFs. Extracting the relevant information from hundreds of SPIRforms, across dozens of suppliers, each using their own format and terminology,is a significant undertaking that most project teams do not budget time orresources for until it is already too late.
Bill of Materials documentation from engineering contractors has similarproblems. The information is there — locked inside documents that were designedfor engineers to read, not for ERP systems to ingest.
Legacy CMMS data from the system being replaced is often the worst of all.Built up over years of ad-hoc data entry, with no enforced standards and noconsistent naming conventions, it typically contains thousands of duplicaterecords, hundreds of incorrect manufacturer names, and part numbers that wereentered incorrectly years ago and have never been corrected.
The fundamental problem is this: the information exists, but it is trapped.Getting it out, verifying it, normalising it, and loading it into an ERP systemin a format that is actually usable — that is the work that most project teamseither underestimate, rush, or hand to resources who do not have the expertiseto do it correctly.
The consequences of poor MRO spare parts data at ERP go-live follow apredictable pattern.
In the first weeks, the problems are visible but manageable. Parts cannot befound. Maintenance teams raise purchase requisitions manually, outside thesystem. Procurement works around the data gaps by calling suppliers directly.Everyone assumes the data will be cleaned up soon.
It rarely is. Because cleaning up MRO data after go-live — while the system islive, while maintenance teams are depending on it, while procurement isprocessing orders through it — is dramatically harder and more expensive thangetting it right before go-live. The cost of fixing data post-go-live isestimated at five to ten times the cost of getting it right upfront.
Meanwhile, the indirect costs accumulate. Emergency procurement — sourcingparts outside normal channels because the system cannot support standard procurementworkflows — typically costs three to four times the standard procurement price.Duplicate purchasing, driven by the inability to confirm whether a part alreadyexists in stock under a different item number, inflates inventory value andties up working capital. And the loss of confidence among maintenance andprocurement teams — the people who were promised that the new system would maketheir jobs easier — creates an adoption problem that can persist for years.
The facilities that experience these problems did not make bad decisions. Theymade the same decision that most project teams make: they assumed the datawould be handled, and they did not ask hard enough questions about how, bywhom, and to what standard.
The instinctive response to a data problem is to throw resources at it. Hire adata entry team. Get the information into the system. Deal with quality issueslater.
This approach fails, consistently and predictably, for one fundamental reason:MRO spare parts data cannot be entered correctly by people who do notunderstand what they are entering.
A data entry operator who does not know the difference between a model numberand a part number — or who cannot recognise that "Emerson Fisher","Fisher Controls" and "Fisher" are all the samemanufacturer — will create the exact problems they were hired to solve. Wrongmanufacturer names. Transposed part numbers. Inconsistent descriptions. Recordsthat look complete but are functionally useless.
The verification step — checking that a part number is not just present butactually valid for that manufacturer, that a manufacturer name matches theverified supplier record, that the technical attributes are plausible for thattype of equipment — requires subject matter expertise that generic data entryservices simply do not have.
This is why companies that approach MRO spare parts cataloging as a data entryproblem end up with data entry results. And why they find themselves, six monthsafter go-live, facing a data cleansing project that costs more than thecataloging should have cost in the first place.
The companies that go live with MRO spare parts data they can trust share onecharacteristic: they worked with specialists who understood the data, not justthe process of entering it.
Specialist MRO data cataloging is not a faster version of data entry. It is afundamentally different discipline. It requires knowledge of industrialequipment and spare parts, familiarity with supplier naming conventions andpart number formats across hundreds of manufacturers, experience with theclassification standards that apply to your industry, and the tools andreference data to verify information at scale rather than record by record.
The result is data that is accurate on day one — not data that needs to befixed before the system can deliver value.
For an industrial facility going live on a new ERP system, that difference isnot marginal. It is the difference between a go-live that delivers on itspromise and one that becomes a cautionary tale about why ERP projects fail todeliver ROI.
Most project teams discover the answer to this question too late. The time toassess the state of your MRO spare parts data — and to understand what it willtake to get it ready — is not in the final weeks before go-live. It is now.
The questions worth asking today are straightforward:
- Where is your spare parts data coming from, and in what format?
- Who is responsible for extracting, verifying and normalising it?
- Do they have the subject matter expertise to verify manufacturer names andpart numbers — not just enter them?
- What standard will descriptions and classifications follow, and who isenforcing it?
- Has the data been validated against the configuration of your target ERPsystem?
If the answers to any of these questions are unclear, the risk to your go-liveis real.
Sharecat Data Services has specialised in MRO spare parts cataloging forindustrial operators for over 30 years — across oil and gas, chemical,pharmaceutical and utilities facilities, on every major ERP and CMMS platform.The team has cataloged more than 900,000 spare parts items, building the subject matter expertise, reference data and tooling that makes it possible todeliver verified, ERP-ready data at a scale and speed that internal teams andgeneric data entry providers cannot match.
If your ERP go-live is on the horizon, the conversation worth having is notabout whether your data needs attention. It is about how much time you haveleft to fix it.
ISO 27001 Certified
