Your organisation already runs a disciplined classification standard. It is called SWEC, and it governs which vendors may supply which scopes. The question worth asking is why the material master inside your ERP is held to a lower standard than your vendor list.
If you operate a refinery, a gas terminal, a power station, a wafer fab, or a palm oil complex in Malaysia with more than ten thousand material records, this is written for you. It sets out what is wrong with your material master, what it is costing, and exactly what fixing it involves — including a regional client programme now in its seventh year.
We are Panemu, an Indonesian technology company one hour's time difference from Kuala Lumpur. We build SCS® (Spares Cataloguing System®) and we have catalogued MRO data across oil and gas, power generation, mining, and manufacturing for over fourteen years. This is not a neutral overview.
Same part, three identities
This is what a duplicate actually looks like
Duplicates rarely announce themselves. They hide behind spelling variants, abbreviation habits, mixed English and Bahasa descriptions, and whoever happened to raise the requisition. Here is one physical bearing as it typically appears across three storerooms.
Now multiply that by ten thousand items. The three codes above are three separate reorder points, three safety stocks, and three lines in your spend analysis that should have been one.
The cost of messy data
Your material master is leaking money in four places
You buy what you already own
Same bearing, three descriptions, two cost centres. Procurement duplicates because nobody trusts the lookup.
Stockouts on critical spares
The part exists somewhere in your network. Wrong code, wrong site, wrong taxonomy — so you get downtime instead of savings.
Spend analysis you cannot trust
Bearing spend looks like five suppliers. It is one supplier under five spelling variants. Your negotiating leverage is gone.
Capital locked in dead stock
Slow movers nobody can find, sitting in the warehouse for years under names nobody searches for.
Lost trust in the item master
Technicians type free text into the purchasing system instead of using the catalogue. Every order becomes a manual order.
Mixed-language descriptions
Legacy records written in a blend of English, Bahasa Melayu, and site shorthand. Search only works if you already know how the original author wrote it.
You are not running an inventory problem. You are running a data problem.
Want your real duplication rate? Send us a sample.
Get a free data auditThe fundamentals
What material master data cataloguing actually is
Material master data cataloguing is the structured process of identifying, describing, classifying, and de-duplicating every material record an organisation holds — spare parts, consumables, chemicals, and equipment components — so that each item has one unique, unambiguous identity that procurement, maintenance, warehouse, and finance teams all recognise.
It is not the same as data cleansing. Cleansing corrects what already exists. Cataloguing establishes the standard that defines what "correct" means — and the rules that stop bad records being created tomorrow.
A properly catalogued record answers four questions without ambiguity: what is it, under a controlled noun–modifier name; what are its specifications, as mandatory attributes in a fixed sequence and consistent units; where does it belong, under a classification code such as UNSPSC or NATO/NCS; and is it already here, checked against part number, manufacturer, and specification before the record is created rather than during an audit two years later.
That last question carries most of the value. Retrospective de-duplication is expensive and never complete. Preventing a duplicate at the point of creation costs almost nothing — but it requires a system that enforces it. Spreadsheets cannot.
A precision worth making
SWEC classifies your vendors. It does not classify your materials.
This distinction matters, and it is regularly blurred in conversation. SWEC (Standardised Work and Equipment Categories) is the categorisation framework under PETRONAS licensing and registration. It determines which scopes of work and equipment a vendor is licensed to supply, and vendors must hold at least one SWEC code to transact.
SWEC governs who may supply. It says nothing about the individual material records sitting inside your ERP — how they are named, which attributes they carry, or whether the same valve exists three times under three codes.
Material classification is a separate discipline, using standards such as UNSPSC, NATO/NCS, or eCl@ss. The two are complementary, and the organisations that run both well can trace a line from a licensed vendor category through to a single governed item record. Most cannot, because only one half of that chain was ever given a standard.
Your vendor list has a taxonomy. Your material master deserves one too.
Four sectors, one pathology
The equipment differs. The data problem does not.
Material data fails the same way everywhere: records created individually, by different people, at different times, without an enforced standard. What changes between sectors is which failure hurts most, and when.
Oil, Gas & Petrochemicals
This is where Malaysia's material spend concentrates. PETRONAS reported USD 75.6 billion in revenue and aims to sustain domestic production at roughly two million barrels of oil equivalent per day, with continued investment in deepwater development and enhanced oil recovery set out in its Activity Outlook 2026–2028. Pengerang Integrated Complex alone processes up to 300,000 barrels per day and ranks as the fourth-largest refinery complex in Southeast Asia.
The pressure point is the turnaround. Valves, rotating equipment, and instrumentation spares must be identifiable weeks before a shutdown window opens, and a part that cannot be found becomes a part that gets air-freighted at premium. Cataloguing here should align with ISO 14224 equipment taxonomy and, on capital projects, with CFIHOS handover from EPC contractors.
Sector experience: Arrow Energy, oil & gas, Australia
Power & Utilities
Malaysian generation and grid assets run on outage schedules, and outage schedules run on spare parts availability. Turbine, boiler, generator, and balance-of-plant components carry long lead times, so a missing or misclassified record does not cause a delay of days — it causes a delay of months.
Multi-station operators face a second problem: the same component carries different codes at different plants, so nobody can compare stock or transfer between sites. A centralised material master is what makes fleet-level planning possible at all — and it becomes more urgent as generation mixes shift and new asset classes enter the fleet.
Sector experience: PT Jawa Satu Power, combined cycle, Indonesia
Electronics & Advanced Manufacturing
Penang's electrical and electronics sector contributed around RM 42 billion to state GDP in 2024 across more than a thousand companies, and the state secured RM 4.9 billion in approved manufacturing investment in the first quarter of 2026 alone, with E&E and machinery accounting for roughly 74 percent of it.
Fabs and precision plants run on uptime measured in minutes, not hours. Where material records are unreliable, technicians bypass the catalogue and raise free-text requisitions — which is how a clean material master degrades one order at a time. For groups running several plants, the highest-value work is usually consolidation: merging plant-level material masters so procurement can aggregate volume and negotiate as a single buyer.
Cataloguing methodology applies identically; sector references available on request
Plantation Processing & Heavy Industry
Mills and downstream processing plants are asset-intensive, geographically dispersed, and frequently operate with material masters built site by site over decades. Sterilisers, presses, boilers, and conveyors draw on overlapping spares that are almost never catalogued to a common standard across estates.
This sector is chronically underserved by cataloguing vendors, which means the improvement available from a first structured programme is usually larger here than anywhere else in the portfolio.
Cataloguing methodology applies identically; sector references available on request
Across all four, the remediation is the same: define the standard, enforce it in a system, cleanse against it, and govern what comes next. Only the dictionary and the classification depth change.
Real case study
How PT Merdeka Copper Gold Tbk went from messy item master to embedded data governance
The strongest evidence a cataloguing partner can offer is not the first contract. It is the renewals. This engagement began as a one-time project in 2019 and runs as a daily service today — which makes it the clearest available answer to the question that matters most: does the standard still hold years after the consultants leave?
PT Merdeka Copper Gold Tbk
The challenge · 2019
MDKA was operating at scale across multiple sites, each carrying its own legacy of free-text descriptions, inconsistent codes, and duplicates accumulated over years of operation. A single bearing might appear under three different codes in three storerooms. Procurement could not reliably consolidate spend. Maintenance could not reliably find parts.
At the same time, MDKA was running an ERP consolidation project. Cleansing afterwards would carry the duplicates into the new system. Pausing the ERP project to fix data first would slip the timeline. Two problems, one deadline. They asked Panemu to do both, in parallel.
What we did first
Before touching a single record, we answered the question most cleansing projects skip: what should this data look like? The answer was a NATO NCS standard adapted for mining — MDKA's standard, fit for MDKA's equipment. Everything that followed had a clear definition of "clean" to enforce.
The engagement, year by year
- SCS® implementation and foundation cataloguing27,771 items catalogued between April and October, running alongside the live ERP consolidation.
- Cataloguing extended to subsidiariesA further 21,000 items from August to December, bringing consistent data quality across the group.
- SCS®–Pronto ERP integrationApril to December. Catalogue data now flows directly into the ERP, so new items are governed at the point of creation.
- Server migrationSCS® moved to a new environment for performance and reliability.
- Daily cataloguing as a serviceEvery Create, Change, and Delete request runs through the ANSI governance module. Renewed under back-to-back contracts and ongoing today.
What changed
- Every item now carries a governed golden record — standardised noun–modifier description, verified OEM part number, NCS classification.
- Catalogue data flows directly into Pronto ERP through a live integration, governed at the point of creation rather than cleaned up afterwards.
- Cataloguing runs daily as a service, not as a periodic clean-up project — so the backlog never rebuilds.
- The data still meets the standard set in 2019, more than five years later.
The lesson: a custom standard is the foundation. Most cleansing projects skip straight to fixing records. Without a defined standard — your standard, fit for your industry — you are only relocating the mess. MDKA started with the right question: what should our data look like? Once that was answered, everything after it had a clear definition of "clean" to enforce.
This is a mining engagement in Indonesia, and we will not present it as evidence that we know your refinery. What it does demonstrate transfers directly: that we can build a standard from scratch for a specific industry, enforce it in a system, integrate it with an ERP, and keep it holding for more than five years. It also happens to sit in the same regional equipment ecosystem as Malaysian operations — the same OEMs, the same distributors, the same mixed-language description habits.
Running multi-site operations in Malaysia? We will run the same analysis on your data.
Book a consultationRegional fit
One hour apart. Same working day.
Offshore data work usually means overnight turnaround and a coordination penalty. Between Yogyakarta and Kuala Lumpur the time difference is one hour, which means questions raised in the morning get answered the same morning — not the next day.
Language your data is actually written in
Legacy material descriptions in Malaysian operations are rarely pure English. Our cataloguers read Bahasa without translation loss, which matters when the original record says “galas bebola” rather than “ball bearing”.
Shared equipment ecosystem
Malaysian and Indonesian operations buy from largely the same OEMs and regional distributors. Part number conventions, catalogue sources, and supplier documentation are already familiar territory.
ASEAN commercial practice
Similar contracting norms, similar procurement processes, and no meaningful cultural translation layer between your team and ours.
Cost structure that survives scale
Programmes of 100,000 records and above remain viable at a cost European-based providers cannot match — without reducing the quality of the cataloguing work itself.
Our approach
Software alone will not fix your data. Expertise will.
Automated tools can flag patterns, but they cannot reliably tell a hydraulic seal from a pneumatic one, interpret an ambiguous manufacturer reference, or apply a classification standard the way an experienced cataloguer does. The difference shows up months later, when you are correcting output that looked clean.
- Fast approximations you will spend months correcting
- Struggle with ambiguous or mixed-language source descriptions
- No accountable judgement behind a classification call
- Clean-looking output that masks underlying gaps
- Specialist cataloguers who understand your equipment
- Software that enforces your naming and classification rules
- Standards applied with accountable, reviewable judgement
- A catalogue you can actually trust — and audit
How we deliver
Five phases. Two of them happen after go-live.
Most cleansing vendors stop at cutover weekend. Migration is one event; keeping data clean is daily work. Each phase produces a defined output you inspect before authorising the next.
-
Data audit
We profile a representative sample from your ERP, spreadsheets, and legacy systems, measuring duplication rate, attribute completeness, and classification consistency against international standards.
Output: free data audit report with prioritised improvement areas · 5 working days
-
Standard & governance design
Naming conventions, the noun–modifier dictionary, classification scheme, attribute templates, and approval workflows are agreed with your data owners, then configured in SCS®. This is your standard, not ours.
Output: approved cataloguing standard and configured SCS® environment · 3–4 weeks
-
Cleanse & enrich
Specialist cataloguers process records in controlled batches. Every record passes duplicate screening on part number, manufacturer, and specification. Missing attributes are researched against OEM documentation.
Output: golden records with standardised descriptions and complete attributes · volume-dependent
-
Classify & migrate
Each item is classified to the agreed scheme, then delivered as ready-to-load files for SAP S/4HANA, Oracle, IBM Maximo, or Pronto — pre-validated so your team imports without reformatting and cutover stays on schedule.
Output: ERP-ready load files with field mapping · runs with phase 3
-
Govern with ANSI
After go-live, every Create, Change, and Delete request passes through the ANSI approval workflow. Nothing enters the master without clearing every check, and every action is logged against a named user.
Output: sustained data quality with a complete audit trail · ongoing
Cleansing is a project. Standardisation is a system. Governance is the discipline that compounds.
Standards
We work to your standard, not ours
There is no universally correct classification standard. Vendors who insist otherwise are describing their own tooling constraints, not your requirement. The right choice depends on your ERP configuration, your reporting obligations, and what your counterparties already use.
- UNSPSC — United Nations Standard Products and Services Code
- Four-level commodity taxonomy: Segment, Family, Class, Commodity. The usual choice where spend analysis and procurement reporting drive the decision. Natively supported in SAP and Maximo material master fields.
- NATO / NCS — NATO Codification System
- Item-level supply codification. Relevant where defence, aviation, or government supply chains are involved, and wherever a single item identity must be recognised across separate organisations. This is the base standard we adapted for Merdeka Copper Gold's mining operations.
- eCl@ss — classification with attribute dictionary
- Classification combined with a standardised attribute dictionary. Common in supply chains anchored to European manufacturers, and increasingly relevant in advanced manufacturing.
- ISO 8000 — master data quality
- The quality standard for the data itself. Where UNSPSC states what an item is, ISO 8000 states whether the record describing it is fit for purpose. This is the vocabulary your auditors use.
- ISO 14224 — reliability and maintenance data
- The reliability and maintenance data taxonomy for oil and gas. The bridge between material data and equipment failure data.
- CFIHOS — IOGP capital facilities handover
- The specification for information handover on capital projects. Where contractually specified, project material data must reconcile with your operational material master.
- Client-defined schemes — hybrid and legacy structures
- Most operators run a hybrid: one scheme for reporting, an internal commodity code for procurement, and a legacy structure that cannot be retired because too many things point at it. SCS® holds multiple classification schemes against a single item record, so you are not forced into a migration nobody asked for.
What we bring
A specialist team — not just a tool.
Bundled service + SCS®
We develop the standard and implement it through our own cataloguing and data governance platform. One contract, one accountability — nobody to point at when something goes wrong.
ERP/EAM-ready exports
Pre-validated load templates for SAP S/4HANA, Oracle, IBM Maximo, and Pronto. Your team imports without reformatting.
International standards-based
Classification built on UNSPSC, NATO NCS, eCl@ss, and ISO 8000 adoption — defensible to your auditors and your ERP team alike.
Capability stays with you
Training and governance transfer are part of delivery, not an upsell. The objective is that you need us less over time, not more.
Already trusted by
Leading asset-intensive companies
Start with the audit. Decide afterwards.
We will tell you what your data actually looks like before you commit to anything. If the answer is that you do not need us, we will say so.
Request your free data audit Five working days. Mutual NDA on request.Your options
Build in-house, contract cataloguers, or partner
| Build in-house | Contract cataloguers | Panemu | |
|---|---|---|---|
| Time to first output | 6–12 months | 4–8 weeks | 3–5 weeks |
| Standard development | Yours to build | Rarely included | Built for your industry |
| Duplicate prevention | Manual | Manual | Enforced at entry |
| Audit trail | Usually absent | Varies | Complete, via ANSI |
| Cost at 100K+ items | High fixed cost | Scales linearly | Scales sub-linearly |
| Knowledge retention | Retained internally | Leaves with the vendor | Transferred by design |
| Post-go-live governance | Yours to build | Not included | Included and handed over |
Building in-house is right if you have sustained volume, existing governance maturity, and a genuine intention to keep the capability permanently. Contract cataloguers are right for a short, bounded backlog with a clear end date. If you need throughput and the system that keeps the result clean afterwards, that is where we are the better answer.
The part most vendors skip
Why clean data gets dirty again — fast
A pattern we see repeatedly: an organisation invests in remediation, achieves a clean catalogue, declares success, then returns three years later with the same problem and a smaller appetite for solving it.
The cause is always the same. The project corrected existing records without controlling how new records are created. Without a workflow governing new entries, every request becomes free text again — and within roughly ten months the duplicates, misclassifications, and non-standard entries are back. All that migration preparation turns out to have been a one-time fix.
Three things stop the drift, and all three must be present. Enforcement at the point of creation, meaning a duplicate screen that runs when a requisitioner types a part number rather than during an annual audit; a documented policy is not enough, because policies become optional under time pressure and systems do not. Named ownership, meaning a specific person accountable for the standard with authority to reject non-conforming records. And documented rules that survive personnel change, because a standard living in one experienced cataloguer's head has a retirement date.
We do not just help you migrate clean. We help you stay clean.
Questions we get asked
Frequently asked questions
Is SWEC a material classification standard?
No, and the distinction matters. SWEC classifies vendors and the scopes they are licensed to supply under PETRONAS licensing and registration. It does not classify the individual material records inside your ERP. Material classification uses standards such as UNSPSC, NATO/NCS, or eCl@ss. The two are complementary: SWEC governs who may supply, material classification governs what you hold.
What is the difference between material cataloguing and data cleansing?
Data cleansing corrects what is already there — typos, blank fields, obvious duplicates. Cataloguing establishes the standard that determines what "correct" means: the naming convention, the attribute set, the classification scheme, and the rules for creating new records. Cleansing without cataloguing produces tidy data that drifts again within two years.
Should we use UNSPSC or NATO NSN?
UNSPSC if the primary driver is spend analysis, procurement reporting, and ERP-native classification — it is simpler to maintain and natively supported in SAP and Maximo. NATO/NCS if item-level identity must be recognised across separate organisations, particularly in defence, aviation, or government supply chains. Both can be held against the same item record in SCS®, so this is rarely an either/or decision.
Can we cleanse data while our ERP migration is already running?
Yes, and it is often the only sensible sequence. Merdeka Copper Gold faced exactly this in 2019: a messy material master and an ERP consolidation on the same deadline. Cleansing afterwards would have carried duplicates into the new system; pausing the ERP project would have slipped the timeline. We ran both in parallel and catalogued 27,771 items between April and October that year.
Do we have to replace our ERP?
No. Cataloguing operates on the data, not the platform. SCS® sits alongside your ERP as the governed environment where records are created and controlled, then delivers validated load files into SAP S/4HANA, Oracle, IBM Maximo, or Pronto.
How do you engage commercially with Malaysian operators?
Engagement structure depends on your procurement requirements. Where a locally incorporated and licensed counterparty is required — as it is for PETRONAS licensing and registration, which requires local incorporation and a minimum paid-up capital — we work through a Malaysian partner arrangement. Where you contract directly with international suppliers, we contract directly. Tell us which applies to you and we will confirm the structure before any proposal.
How long does a programme take?
A free data audit takes five working days. Standard and governance design takes three to four weeks. Cleansing and classification scale with record count and enrichment depth — as a reference point, a 27,771-item foundation batch was delivered in roughly six months alongside a live ERP consolidation. Output is delivered in batches, so value arrives before completion.
What does it cost?
Pricing is per record and depends on required enrichment depth — a standardisation pass costs materially less than full attribute research against OEM documentation. The free audit establishes which of your records need which treatment, so the quotation reflects your actual data rather than an industry average.
Can you work with our existing naming convention?
Yes. SCS® is configured to your dictionary, your attribute templates, and your classification scheme. If you do not have one, we develop it with you — as we did for Merdeka, where the deliverable was a NATO NCS standard adapted specifically for mining equipment. We do not require you to adopt our taxonomy.
Can your cataloguers handle records written in Bahasa Melayu?
Yes. Our cataloguers work in Bahasa and English without a translation layer, which matters because legacy material descriptions in Malaysian operations are frequently a mixture of both, plus site-specific shorthand. Interpreting the original intent correctly is the difference between a consolidation and a mistake.
How is our data protected?
A mutual NDA is available on request and executed before any transfer. Work is performed inside SCS® under role-based access, with duties segregated across warehouse, planning, and procurement roles, and every action logged in the ANSI audit trail rather than distributed to individual workstations.
Can we license SCS® without the service?
Yes. Organisations building internal capability license the platform, receive training, and run cataloguing with their own team, with Panemu providing the standard, configuration, and ongoing support.
See your duplication rate in five working days
Send us a sample of your material master with variation across categories and sites. We will return a free data audit report covering:
- Your measured duplicate ratio
- Attribute completeness across the sample
- Classification consistency against international standards
- The top improvement areas, ranked by operational impact
- A realistic scope and effort estimate for remediation
Your organisation already runs a disciplined classification standard. It is called SWEC, and it governs which vendors may supply which scopes. The question worth asking is why the material master inside your ERP is held to a lower standard than your vendor list.
If you operate a refinery, a gas terminal, a power station, a wafer fab, or a palm oil complex in Malaysia with more than ten thousand material records, this is written for you. It sets out what is wrong with your material master, what it is costing, and exactly what fixing it involves — including a regional client programme now in its seventh year.
We are Panemu, an Indonesian technology company one hour's time difference from Kuala Lumpur. We build SCS® (Spares Cataloguing System®) and we have catalogued MRO data across oil and gas, power generation, mining, and manufacturing for over fourteen years. This is not a neutral overview.
Same part, three identities
This is what a duplicate actually looks like
Duplicates rarely announce themselves. They hide behind spelling variants, abbreviation habits, mixed English and Bahasa descriptions, and whoever happened to raise the requisition. Here is one physical bearing as it typically appears across three storerooms.
Now multiply that by ten thousand items. The three codes above are three separate reorder points, three safety stocks, and three lines in your spend analysis that should have been one.
The cost of messy data
Your material master is leaking money in four places
You buy what you already own
Same bearing, three descriptions, two cost centres. Procurement duplicates because nobody trusts the lookup.
Stockouts on critical spares
The part exists somewhere in your network. Wrong code, wrong site, wrong taxonomy — so you get downtime instead of savings.
Spend analysis you cannot trust
Bearing spend looks like five suppliers. It is one supplier under five spelling variants. Your negotiating leverage is gone.
Capital locked in dead stock
Slow movers nobody can find, sitting in the warehouse for years under names nobody searches for.
Lost trust in the item master
Technicians type free text into the purchasing system instead of using the catalogue. Every order becomes a manual order.
Mixed-language descriptions
Legacy records written in a blend of English, Bahasa Melayu, and site shorthand. Search only works if you already know how the original author wrote it.
You are not running an inventory problem. You are running a data problem.
Want your real duplication rate? Send us a sample.
Get a free data auditThe fundamentals
What material master data cataloguing actually is
Material master data cataloguing is the structured process of identifying, describing, classifying, and de-duplicating every material record an organisation holds — spare parts, consumables, chemicals, and equipment components — so that each item has one unique, unambiguous identity that procurement, maintenance, warehouse, and finance teams all recognise.
It is not the same as data cleansing. Cleansing corrects what already exists. Cataloguing establishes the standard that defines what "correct" means — and the rules that stop bad records being created tomorrow.
A properly catalogued record answers four questions without ambiguity: what is it, under a controlled noun–modifier name; what are its specifications, as mandatory attributes in a fixed sequence and consistent units; where does it belong, under a classification code such as UNSPSC or NATO/NCS; and is it already here, checked against part number, manufacturer, and specification before the record is created rather than during an audit two years later.
That last question carries most of the value. Retrospective de-duplication is expensive and never complete. Preventing a duplicate at the point of creation costs almost nothing — but it requires a system that enforces it. Spreadsheets cannot.
A precision worth making
SWEC classifies your vendors. It does not classify your materials.
This distinction matters, and it is regularly blurred in conversation. SWEC (Standardised Work and Equipment Categories) is the categorisation framework under PETRONAS licensing and registration. It determines which scopes of work and equipment a vendor is licensed to supply, and vendors must hold at least one SWEC code to transact.
SWEC governs who may supply. It says nothing about the individual material records sitting inside your ERP — how they are named, which attributes they carry, or whether the same valve exists three times under three codes.
Material classification is a separate discipline, using standards such as UNSPSC, NATO/NCS, or eCl@ss. The two are complementary, and the organisations that run both well can trace a line from a licensed vendor category through to a single governed item record. Most cannot, because only one half of that chain was ever given a standard.
Your vendor list has a taxonomy. Your material master deserves one too.
Four sectors, one pathology
The equipment differs. The data problem does not.
Material data fails the same way everywhere: records created individually, by different people, at different times, without an enforced standard. What changes between sectors is which failure hurts most, and when.
Oil, Gas & Petrochemicals
This is where Malaysia's material spend concentrates. PETRONAS reported USD 75.6 billion in revenue and aims to sustain domestic production at roughly two million barrels of oil equivalent per day, with continued investment in deepwater development and enhanced oil recovery set out in its Activity Outlook 2026–2028. Pengerang Integrated Complex alone processes up to 300,000 barrels per day and ranks as the fourth-largest refinery complex in Southeast Asia.
The pressure point is the turnaround. Valves, rotating equipment, and instrumentation spares must be identifiable weeks before a shutdown window opens, and a part that cannot be found becomes a part that gets air-freighted at premium. Cataloguing here should align with ISO 14224 equipment taxonomy and, on capital projects, with CFIHOS handover from EPC contractors.
Sector experience: Arrow Energy, oil & gas, Australia
Power & Utilities
Malaysian generation and grid assets run on outage schedules, and outage schedules run on spare parts availability. Turbine, boiler, generator, and balance-of-plant components carry long lead times, so a missing or misclassified record does not cause a delay of days — it causes a delay of months.
Multi-station operators face a second problem: the same component carries different codes at different plants, so nobody can compare stock or transfer between sites. A centralised material master is what makes fleet-level planning possible at all — and it becomes more urgent as generation mixes shift and new asset classes enter the fleet.
Sector experience: PT Jawa Satu Power, combined cycle, Indonesia
Electronics & Advanced Manufacturing
Penang's electrical and electronics sector contributed around RM 42 billion to state GDP in 2024 across more than a thousand companies, and the state secured RM 4.9 billion in approved manufacturing investment in the first quarter of 2026 alone, with E&E and machinery accounting for roughly 74 percent of it.
Fabs and precision plants run on uptime measured in minutes, not hours. Where material records are unreliable, technicians bypass the catalogue and raise free-text requisitions — which is how a clean material master degrades one order at a time. For groups running several plants, the highest-value work is usually consolidation: merging plant-level material masters so procurement can aggregate volume and negotiate as a single buyer.
Cataloguing methodology applies identically; sector references available on request
Plantation Processing & Heavy Industry
Mills and downstream processing plants are asset-intensive, geographically dispersed, and frequently operate with material masters built site by site over decades. Sterilisers, presses, boilers, and conveyors draw on overlapping spares that are almost never catalogued to a common standard across estates.
This sector is chronically underserved by cataloguing vendors, which means the improvement available from a first structured programme is usually larger here than anywhere else in the portfolio.
Cataloguing methodology applies identically; sector references available on request
Across all four, the remediation is the same: define the standard, enforce it in a system, cleanse against it, and govern what comes next. Only the dictionary and the classification depth change.
Real case study
How PT Merdeka Copper Gold Tbk went from messy item master to embedded data governance
The strongest evidence a cataloguing partner can offer is not the first contract. It is the renewals. This engagement began as a one-time project in 2019 and runs as a daily service today — which makes it the clearest available answer to the question that matters most: does the standard still hold years after the consultants leave?
PT Merdeka Copper Gold Tbk
The challenge · 2019
MDKA was operating at scale across multiple sites, each carrying its own legacy of free-text descriptions, inconsistent codes, and duplicates accumulated over years of operation. A single bearing might appear under three different codes in three storerooms. Procurement could not reliably consolidate spend. Maintenance could not reliably find parts.
At the same time, MDKA was running an ERP consolidation project. Cleansing afterwards would carry the duplicates into the new system. Pausing the ERP project to fix data first would slip the timeline. Two problems, one deadline. They asked Panemu to do both, in parallel.
What we did first
Before touching a single record, we answered the question most cleansing projects skip: what should this data look like? The answer was a NATO NCS standard adapted for mining — MDKA's standard, fit for MDKA's equipment. Everything that followed had a clear definition of "clean" to enforce.
The engagement, year by year
- SCS® implementation and foundation cataloguing27,771 items catalogued between April and October, running alongside the live ERP consolidation.
- Cataloguing extended to subsidiariesA further 21,000 items from August to December, bringing consistent data quality across the group.
- SCS®–Pronto ERP integrationApril to December. Catalogue data now flows directly into the ERP, so new items are governed at the point of creation.
- Server migrationSCS® moved to a new environment for performance and reliability.
- Daily cataloguing as a serviceEvery Create, Change, and Delete request runs through the ANSI governance module. Renewed under back-to-back contracts and ongoing today.
What changed
- Every item now carries a governed golden record — standardised noun–modifier description, verified OEM part number, NCS classification.
- Catalogue data flows directly into Pronto ERP through a live integration, governed at the point of creation rather than cleaned up afterwards.
- Cataloguing runs daily as a service, not as a periodic clean-up project — so the backlog never rebuilds.
- The data still meets the standard set in 2019, more than five years later.
The lesson: a custom standard is the foundation. Most cleansing projects skip straight to fixing records. Without a defined standard — your standard, fit for your industry — you are only relocating the mess. MDKA started with the right question: what should our data look like? Once that was answered, everything after it had a clear definition of "clean" to enforce.
This is a mining engagement in Indonesia, and we will not present it as evidence that we know your refinery. What it does demonstrate transfers directly: that we can build a standard from scratch for a specific industry, enforce it in a system, integrate it with an ERP, and keep it holding for more than five years. It also happens to sit in the same regional equipment ecosystem as Malaysian operations — the same OEMs, the same distributors, the same mixed-language description habits.
Running multi-site operations in Malaysia? We will run the same analysis on your data.
Book a consultationRegional fit
One hour apart. Same working day.
Offshore data work usually means overnight turnaround and a coordination penalty. Between Yogyakarta and Kuala Lumpur the time difference is one hour, which means questions raised in the morning get answered the same morning — not the next day.
Language your data is actually written in
Legacy material descriptions in Malaysian operations are rarely pure English. Our cataloguers read Bahasa without translation loss, which matters when the original record says “galas bebola” rather than “ball bearing”.
Shared equipment ecosystem
Malaysian and Indonesian operations buy from largely the same OEMs and regional distributors. Part number conventions, catalogue sources, and supplier documentation are already familiar territory.
ASEAN commercial practice
Similar contracting norms, similar procurement processes, and no meaningful cultural translation layer between your team and ours.
Cost structure that survives scale
Programmes of 100,000 records and above remain viable at a cost European-based providers cannot match — without reducing the quality of the cataloguing work itself.
Our approach
Software alone will not fix your data. Expertise will.
Automated tools can flag patterns, but they cannot reliably tell a hydraulic seal from a pneumatic one, interpret an ambiguous manufacturer reference, or apply a classification standard the way an experienced cataloguer does. The difference shows up months later, when you are correcting output that looked clean.
- Fast approximations you will spend months correcting
- Struggle with ambiguous or mixed-language source descriptions
- No accountable judgement behind a classification call
- Clean-looking output that masks underlying gaps
- Specialist cataloguers who understand your equipment
- Software that enforces your naming and classification rules
- Standards applied with accountable, reviewable judgement
- A catalogue you can actually trust — and audit
How we deliver
Five phases. Two of them happen after go-live.
Most cleansing vendors stop at cutover weekend. Migration is one event; keeping data clean is daily work. Each phase produces a defined output you inspect before authorising the next.
-
Data audit
We profile a representative sample from your ERP, spreadsheets, and legacy systems, measuring duplication rate, attribute completeness, and classification consistency against international standards.
Output: free data audit report with prioritised improvement areas · 5 working days
-
Standard & governance design
Naming conventions, the noun–modifier dictionary, classification scheme, attribute templates, and approval workflows are agreed with your data owners, then configured in SCS®. This is your standard, not ours.
Output: approved cataloguing standard and configured SCS® environment · 3–4 weeks
-
Cleanse & enrich
Specialist cataloguers process records in controlled batches. Every record passes duplicate screening on part number, manufacturer, and specification. Missing attributes are researched against OEM documentation.
Output: golden records with standardised descriptions and complete attributes · volume-dependent
-
Classify & migrate
Each item is classified to the agreed scheme, then delivered as ready-to-load files for SAP S/4HANA, Oracle, IBM Maximo, or Pronto — pre-validated so your team imports without reformatting and cutover stays on schedule.
Output: ERP-ready load files with field mapping · runs with phase 3
-
Govern with ANSI
After go-live, every Create, Change, and Delete request passes through the ANSI approval workflow. Nothing enters the master without clearing every check, and every action is logged against a named user.
Output: sustained data quality with a complete audit trail · ongoing
Cleansing is a project. Standardisation is a system. Governance is the discipline that compounds.
Standards
We work to your standard, not ours
There is no universally correct classification standard. Vendors who insist otherwise are describing their own tooling constraints, not your requirement. The right choice depends on your ERP configuration, your reporting obligations, and what your counterparties already use.
- UNSPSC — United Nations Standard Products and Services Code
- Four-level commodity taxonomy: Segment, Family, Class, Commodity. The usual choice where spend analysis and procurement reporting drive the decision. Natively supported in SAP and Maximo material master fields.
- NATO / NCS — NATO Codification System
- Item-level supply codification. Relevant where defence, aviation, or government supply chains are involved, and wherever a single item identity must be recognised across separate organisations. This is the base standard we adapted for Merdeka Copper Gold's mining operations.
- eCl@ss — classification with attribute dictionary
- Classification combined with a standardised attribute dictionary. Common in supply chains anchored to European manufacturers, and increasingly relevant in advanced manufacturing.
- ISO 8000 — master data quality
- The quality standard for the data itself. Where UNSPSC states what an item is, ISO 8000 states whether the record describing it is fit for purpose. This is the vocabulary your auditors use.
- ISO 14224 — reliability and maintenance data
- The reliability and maintenance data taxonomy for oil and gas. The bridge between material data and equipment failure data.
- CFIHOS — IOGP capital facilities handover
- The specification for information handover on capital projects. Where contractually specified, project material data must reconcile with your operational material master.
- Client-defined schemes — hybrid and legacy structures
- Most operators run a hybrid: one scheme for reporting, an internal commodity code for procurement, and a legacy structure that cannot be retired because too many things point at it. SCS® holds multiple classification schemes against a single item record, so you are not forced into a migration nobody asked for.
What we bring
A specialist team — not just a tool.
Bundled service + SCS®
We develop the standard and implement it through our own cataloguing and data governance platform. One contract, one accountability — nobody to point at when something goes wrong.
ERP/EAM-ready exports
Pre-validated load templates for SAP S/4HANA, Oracle, IBM Maximo, and Pronto. Your team imports without reformatting.
International standards-based
Classification built on UNSPSC, NATO NCS, eCl@ss, and ISO 8000 adoption — defensible to your auditors and your ERP team alike.
Capability stays with you
Training and governance transfer are part of delivery, not an upsell. The objective is that you need us less over time, not more.
Already trusted by
Leading asset-intensive companies
Start with the audit. Decide afterwards.
We will tell you what your data actually looks like before you commit to anything. If the answer is that you do not need us, we will say so.
Request your free data audit Five working days. Mutual NDA on request.Your options
Build in-house, contract cataloguers, or partner
| Build in-house | Contract cataloguers | Panemu | |
|---|---|---|---|
| Time to first output | 6–12 months | 4–8 weeks | 3–5 weeks |
| Standard development | Yours to build | Rarely included | Built for your industry |
| Duplicate prevention | Manual | Manual | Enforced at entry |
| Audit trail | Usually absent | Varies | Complete, via ANSI |
| Cost at 100K+ items | High fixed cost | Scales linearly | Scales sub-linearly |
| Knowledge retention | Retained internally | Leaves with the vendor | Transferred by design |
| Post-go-live governance | Yours to build | Not included | Included and handed over |
Building in-house is right if you have sustained volume, existing governance maturity, and a genuine intention to keep the capability permanently. Contract cataloguers are right for a short, bounded backlog with a clear end date. If you need throughput and the system that keeps the result clean afterwards, that is where we are the better answer.
The part most vendors skip
Why clean data gets dirty again — fast
A pattern we see repeatedly: an organisation invests in remediation, achieves a clean catalogue, declares success, then returns three years later with the same problem and a smaller appetite for solving it.
The cause is always the same. The project corrected existing records without controlling how new records are created. Without a workflow governing new entries, every request becomes free text again — and within roughly ten months the duplicates, misclassifications, and non-standard entries are back. All that migration preparation turns out to have been a one-time fix.
Three things stop the drift, and all three must be present. Enforcement at the point of creation, meaning a duplicate screen that runs when a requisitioner types a part number rather than during an annual audit; a documented policy is not enough, because policies become optional under time pressure and systems do not. Named ownership, meaning a specific person accountable for the standard with authority to reject non-conforming records. And documented rules that survive personnel change, because a standard living in one experienced cataloguer's head has a retirement date.
We do not just help you migrate clean. We help you stay clean.
Questions we get asked
Frequently asked questions
Is SWEC a material classification standard?
No, and the distinction matters. SWEC classifies vendors and the scopes they are licensed to supply under PETRONAS licensing and registration. It does not classify the individual material records inside your ERP. Material classification uses standards such as UNSPSC, NATO/NCS, or eCl@ss. The two are complementary: SWEC governs who may supply, material classification governs what you hold.
What is the difference between material cataloguing and data cleansing?
Data cleansing corrects what is already there — typos, blank fields, obvious duplicates. Cataloguing establishes the standard that determines what "correct" means: the naming convention, the attribute set, the classification scheme, and the rules for creating new records. Cleansing without cataloguing produces tidy data that drifts again within two years.
Should we use UNSPSC or NATO NSN?
UNSPSC if the primary driver is spend analysis, procurement reporting, and ERP-native classification — it is simpler to maintain and natively supported in SAP and Maximo. NATO/NCS if item-level identity must be recognised across separate organisations, particularly in defence, aviation, or government supply chains. Both can be held against the same item record in SCS®, so this is rarely an either/or decision.
Can we cleanse data while our ERP migration is already running?
Yes, and it is often the only sensible sequence. Merdeka Copper Gold faced exactly this in 2019: a messy material master and an ERP consolidation on the same deadline. Cleansing afterwards would have carried duplicates into the new system; pausing the ERP project would have slipped the timeline. We ran both in parallel and catalogued 27,771 items between April and October that year.
Do we have to replace our ERP?
No. Cataloguing operates on the data, not the platform. SCS® sits alongside your ERP as the governed environment where records are created and controlled, then delivers validated load files into SAP S/4HANA, Oracle, IBM Maximo, or Pronto.
How do you engage commercially with Malaysian operators?
Engagement structure depends on your procurement requirements. Where a locally incorporated and licensed counterparty is required — as it is for PETRONAS licensing and registration, which requires local incorporation and a minimum paid-up capital — we work through a Malaysian partner arrangement. Where you contract directly with international suppliers, we contract directly. Tell us which applies to you and we will confirm the structure before any proposal.
How long does a programme take?
A free data audit takes five working days. Standard and governance design takes three to four weeks. Cleansing and classification scale with record count and enrichment depth — as a reference point, a 27,771-item foundation batch was delivered in roughly six months alongside a live ERP consolidation. Output is delivered in batches, so value arrives before completion.
What does it cost?
Pricing is per record and depends on required enrichment depth — a standardisation pass costs materially less than full attribute research against OEM documentation. The free audit establishes which of your records need which treatment, so the quotation reflects your actual data rather than an industry average.
Can you work with our existing naming convention?
Yes. SCS® is configured to your dictionary, your attribute templates, and your classification scheme. If you do not have one, we develop it with you — as we did for Merdeka, where the deliverable was a NATO NCS standard adapted specifically for mining equipment. We do not require you to adopt our taxonomy.
Can your cataloguers handle records written in Bahasa Melayu?
Yes. Our cataloguers work in Bahasa and English without a translation layer, which matters because legacy material descriptions in Malaysian operations are frequently a mixture of both, plus site-specific shorthand. Interpreting the original intent correctly is the difference between a consolidation and a mistake.
How is our data protected?
A mutual NDA is available on request and executed before any transfer. Work is performed inside SCS® under role-based access, with duties segregated across warehouse, planning, and procurement roles, and every action logged in the ANSI audit trail rather than distributed to individual workstations.
Can we license SCS® without the service?
Yes. Organisations building internal capability license the platform, receive training, and run cataloguing with their own team, with Panemu providing the standard, configuration, and ongoing support.
See your duplication rate in five working days
Send us a sample of your material master with variation across categories and sites. We will return a free data audit report covering:
- Your measured duplicate ratio
- Attribute completeness across the sample
- Classification consistency against international standards
- The top improvement areas, ranked by operational impact
- A realistic scope and effort estimate for remediation