Your ERP Is Live. Which Master Data Errors Quietly Distort Every Plan?
Small master-data errors in lead times, calendars, routings, supplier terms and material attributes can quietly distort garment plans long after the ERP project is declared live.
The short answer
Master data can distort a garment plan long before anybody notices a transaction error. If supplier lead times, factory calendars, operation routes, material attributes, capacity assumptions or packing parameters are wrong, the ERP can calculate a perfectly consistent answer to the wrong problem.
After go-live, the most valuable master-data work is not cleaning every field equally. It is identifying the small group of assumptions that change planning, purchasing, costing, production and delivery decisions, assigning clear business ownership, and monitoring when those assumptions drift away from reality.
This is one of the operating questions behind ERP Is Live. How Do You Turn Garment Data Into Better Daily Decisions?: whether the master data is good enough for the plan to deserve trust.
Master-data errors are dangerous because the system can still look perfectly normal
Transactional errors are obvious because somebody cannot complete a task or a document does not balance. Master-data errors often remain hidden because the system still works.
A purchase suggestion is created. A capacity plan is generated. A costing result is calculated. A delivery date appears. The output looks authoritative because the system processed it correctly.
The problem is upstream. A supplier lead time still says 21 days when the actual pattern has become 35. A factory calendar ignores a local holiday. A routing assumes an operation runs at a standard that the current product family never achieves. A trim minimum order quantity is outdated. A material unit conversion is wrong.
Those errors do not crash the ERP. They quietly distort the decisions built on top of it.
Some master data has far more operational leverage than the rest.
Not every master-data field deserves the same governance effort.
Prioritize data that changes one or more of these outcomes:
- promised date;
- material readiness;
- capacity requirement;
- procurement urgency;
- cost or margin;
- quality requirement;
- packing/shipping readiness.
For a garment business, high-leverage master data typically includes style and material attributes, supplier parameters, lead times, color-size structures, operation sequences, work-center capacities, efficiency assumptions, factory calendars, packaging rules and logistics parameters.
The exact list depends on the business model, but the principle is consistent: govern the assumptions that propagate widely.
Lead time should reflect how the supplier or process actually performs
Lead time is one of the most influential planning inputs and one of the easiest to leave stale.
A supplier may have quoted a standard lead time years ago. Actual performance may now vary by material family, color, season, order quantity or production location. If the master value never changes, planning will repeatedly create purchase urgency too late.
A better approach compares master lead time with recent actual milestones. The objective is not to change the master after every late delivery. It is to identify persistent deviation that justifies a controlled update.
The owner should know what event starts and ends the lead time. "PO to arrival" is different from "supplier confirmation to factory receipt." Clear event definitions matter.
A wrong factory calendar distorts every plan built on top of it.
A missing holiday or incorrect shift calendar can alter available capacity, planned completion and delivery calculations.
In multi-factory groups this becomes more complex because calendars differ by country, site, department and sometimes production area. Local teams need authority to maintain real calendars, while the group needs visibility that the calendars are current.
Calendar governance is operational work. It should not depend on an annual IT cleanup.
Routing detail should match the decisions the business needs to make.
A routing can be too simple or too detailed.
If operations are too generic, capacity and costing may not reflect real constraints. If every micro-step is modeled, maintenance becomes expensive and users may stop trusting the route because it is never current.
The correct level of detail is the level required to support planning, execution and cost decisions reliably.
Review routes when product construction changes, new equipment is introduced, work is subcontracted or actual performance repeatedly diverges from the standard.
Material attributes become risky when factories interpret them differently
Garment materials carry attributes that affect sourcing, consumption, quality and production.
Width, color, composition, finish, shrinkage, inspection class, unit of measure and packaging can all matter. If similar materials are described differently, users create duplicates. If critical attributes are omitted, teams re-enter context in free text.
Master data needs an owner because reality keeps changing.
Every high-leverage data domain should have:
- an owner;
- a definition;
- a source of truth;
- an update trigger;
- a review frequency;
- an exception path.
For example, sourcing may own supplier lead times. Planning may own factory calendars and work centers. Industrial engineering may own routing standards. Merchandising/product teams may own style attributes. Quality may own inspection classifications.
The owner is not expected to edit every record personally. The owner is accountable for the rules.
Operational exceptions reveal weak master data better than abstract completeness scores.
Avoid measuring quality only by completion percentage.
A field can be 100% populated and still be wrong.
Better indicators include:
- number of planning exceptions caused by wrong lead time;
- repeated manual date overrides;
- duplicate material creation;
- capacity rework caused by route errors;
- purchase amendments caused by wrong MOQ/pack assumptions;
- orders replanned because calendars were wrong.
Those measures connect data governance to operating value.
The best correction happens close to the business event that proves the data wrong
Central governance is useful for standards. Local operations are often the first to see drift.
A factory should be able to flag that a routing standard is unrealistic. Purchasing should be able to flag that a supplier lead time has changed. Warehouse should be able to flag unit-of-measure confusion.
The governance model needs a controlled path to review and correct those issues quickly. Otherwise users learn to work around the master rather than improve it.
Where Epicor Venus gives garment master data operational consequence
Epicor provides the enterprise-grade ERP foundation for supplier, inventory, purchasing, finance and resource data. Epicor Venus adds the garment-specific style, BOM, routing, calendar, material and production context that determines whether planning assumptions reflect the way the factory actually operates.
ACC's role is to help manufacturers decide which master data has real leverage, who owns it and which operating event should trigger correction. Epicor Venus provides the industry structure; Epicor provides the enterprise foundation that depends on those definitions being trustworthy. Explore Epicor Venus for garment and apparel manufacturing.
The strongest review follows the data that repeatedly causes planning exceptions.
Once a month, review the highest-impact exceptions from planning, purchasing, production and costing. Ask which ones were caused by bad assumptions rather than one-off events.
Then choose a small number of master-data corrections with clear owners. Track whether the same exception recurs after the change.
This creates a feedback loop between actual operations and system assumptions.
Change points are where master data needs the most discipline
Master data quality improves when updates are tied to business events. A new fabric source should trigger a review of lead time, width, usable yield and inspection assumptions. A new factory should trigger calendar, capacity, efficiency and routing checks. A changed buyer packing rule should trigger the relevant order and packaging attributes. Waiting for a quarterly data-cleaning exercise allows weak assumptions to affect many orders before anyone notices.
The approval model should match the risk. Low-impact descriptive fields can remain close to the operational user. Fields that drive planning, costing, procurement or compliance need stronger ownership because a single change can influence many downstream transactions.
It is also useful to track “manual override frequency” by master-data field. If planners repeatedly override a lead time, efficiency assumption or material parameter, the system is telling you that the master is no longer credible. Repeated overrides should create a review task rather than becoming accepted background noise.
Finally, make effective dates explicit. A new standard should not silently rewrite the logic behind orders that were planned under the old assumption. Where the business process requires it, preserve which version applied when the order was created or released. That makes later variance analysis much more trustworthy.
Frequently asked questions
Should every master-data change require central approval?
No. Governance should match risk. High-impact or group-standard definitions may need stronger control; routine local updates can use delegated ownership with audit history.
Is master-data quality mainly an implementation problem?
No. Data changes as suppliers, factories, products and processes change. Governance must continue after go-live.
How often should lead times be reviewed?
Review frequency should match volatility and business impact. The important point is to compare assumptions with actual performance rather than leave values unchanged indefinitely.
Can AI clean master data automatically?
AI may help identify anomalies or duplicates, but business owners still need to decide what is operationally correct.
User workarounds are evidence that the master no longer deserves trust
Do not force compliance without diagnosis. Ask users to show three recent cases where the master produced a bad decision. Classify whether the issue came from an outdated value, unclear definition, missing local parameter or a process that never updates the field. Then fix the maintenance mechanism, not only the three records.
A trustworthy master earns adoption because users see that corrections persist and improve planning. A master that is repeatedly overridden teaches users that the spreadsheet is safer.
The real test is whether plans become more reliable after the correction
A master-data review becomes credible when the team can trace one planning exception back to the field that distorted it. Pick a recent late material, impossible capacity plan or wrong routing assumption and follow the evidence backward until the master-data cause becomes visible.
After correction, the same planning scenario should become more believable. If users still need manual overrides for the same reason, the master has not really been fixed.