When your operation was small, the system you built yourself fit like a glove. A few clients, a single vault, some spreadsheets, and custom code built to handle precisely what the business needed at the time. You built it because no off-the-shelf tool understood allocated holdings, chain of custody, or serialized bars the way your business does. It was precise, it was yours, and it worked. Until it didn’t.
For a depository, vault operator, bullion bank, or refinery, inventory is not a number on a shelf. It is someone else’s wealth, held under your name, backed by your word. The vaults in London alone safeguarded roughly 9,500 tonnes of gold in mid-2026, worth around $1.2 trillion, which works out to about three-quarters of a million individual bars, each with its own weight, purity, and owner. At that scale, “the spreadsheet is probably right” is not something you can say to a client, an auditor, or an insurer.
A homegrown precious metals inventory system is worth replacing once growth creates more manual work rather than easing it, once you can no longer fully rely on reconciliation and reporting, or once keeping the system running eats up engineering hours that should be spent building the business instead. The system rarely fails in one dramatic moment. It just gets harder to trust as you grow.
Below are the warning signs, a quick self-check, and a safe way out.
Why precious metals companies built their own systems in the first place
Most proprietary vault systems were not built out of ambition but necessity. Off-the-shelf warehouse software was designed to move pallets and cartons, not to track ownership, custody, and financial accountability at once. It had no real concept of allocated, segregated, and pooled holdings, serialized bars, assay records, dual-control approvals, or client-specific custody rules. Building your own was often the only way to run the business correctly.
That was a reasonable decision, and it made a lot of operators into leaders. The question today is not whether it was right, but if it still serves the next decade of growth.
Warning sign 1: Reconciliation has become a manual grind
The first crack usually shows up in reconciliation. Precious metals reconciliation means keeping physical inventory, ownership records, financial systems, and client accounts in agreement, continuously. When your team produces that agreement by exporting, merging, and hand-checking, the system has stopped doing the work and started making it.
The gap between what is true and what your system says
Managing a vault on manual records is like navigating with last week’s map while the tide moves every hour. By the time a reconciliation is finished, another movement has already happened somewhere else. In a custody business, that gap between reality and the record is where reconciliation problems and client disputes tend to surface.
Indigo Precious Metals, a Singapore trading firm, described hitting this point as it grew: internally managed spreadsheets and manual workflows could not keep pace with rising volumes, so the firm moved to automate its data flows and reduce manual reconciliation.
Every manual touch is a chance for a costly error
Spreadsheets feel controllable because you built them, but they carry more risk than most teams assume. In a 2026 survey of more than 1,000 US operations professionals, 22 percent said they ran into spreadsheet errors every single day, and teams lost an average of 3.6 hours a week just fixing them. And catching those errors is harder than it looks: separate 2025 research found that careful manual review typically catches only about half of the mistakes present. On a bar list, where one transposed serial number or a misplaced decimal on a weight is a custody discrepancy rather than a typo, that margin matters.
Warning sign 2: Every new client or vault means custom work
Growth is supposed to be good news. On a homegrown system, it can be the moment things slow down. A system designed around a single facility rarely expects more than one, and a platform built for a small client roster tends to bake their specific rules into the code.
Onboarding slows down instead of speeding up
When each client wants a different custody model, its own segregation rules, and its own reporting and billing, and those rules live in a few developers’ heads, adding a client turns into a development project. A depository taking on a new bullion-bank account, or a vault operator onboarding a fund that requires strictly segregated storage, should be able to configure those rules, not rebuild the system for each one. On a homegrown platform, every new account competes for the same scarce engineering time, so onboarding slows exactly when the business is trying to grow.
The footprint grows, the leverage doesn’t
Open a second vault, then a third, and staff start reconciling across sites by hand and re-keying the same movement in more than one place. The workload grows with the footprint instead of staying flat, which is the opposite of scale.
Warning sign 3: Your audit trail can’t stand up to scrutiny
An audit trail is not a log you assemble after the fact. It is a continuous, tamper-resistant record of who did what, when, and to which asset. A homegrown system that reconstructs history from spreadsheets and memory struggles to produce that on demand, and those requests usually come from auditors, insurers, and clients.
Expectations have also risen. Clients, auditors, and insurers increasingly want continuous, verifiable custody records and, in many cases, proof of reserves rather than a periodic manual count. In a business where inventory is a financial asset, ownership records and chain of custody are the product as much as the metal is.
If proving clean custody and good-delivery lineage for a single bar takes hours of digging, that is a gap worth closing. Precious metals dealers also operate under anti-money-laundering obligations in the US, so an audit trail you cannot fully trust is a gap in your compliance posture.
Warning sign 4: You can’t see your true position in real time
Picture the request every depository has to be ready for: a bullion-bank client asks for a current statement of exactly what you hold in their name, by bar and by vault. A long list of rows is not a live position. If “what do we hold, for whom, where, right now?” cannot be answered on demand,the system is archiving the past instead of showing you the present.
Modern operations need a live view: total holdings by client, by vault, by metal, and by ownership model, available the moment someone asks. That same real-time position is what makes continuous client reporting possible, what supports proof-of-reserves expectations, and what lets you produce an audit-ready report on request rather than on a deadline. When the picture is always current, you spend less time reacting and more time deciding.
Warning sign 5: The whole system rests on a few people
If one or two developers left tomorrow, could anyone else safely change, secure, or fully explain your inventory system? If the honest answer is no, you have a concentration of risk in a few people rather than a durable platform.
Homegrown systems tend to concentrate knowledge in a small number of heads, and often on a small number of machines. Enterprise operations usually need documented processes, redundancy, and continuity that does not depend on any single person staying. This is a common pattern: in a 2025 survey of more than 500 US IT professionals, 68 percent said their legacy systems were maintained by already-stretched internal teams.
The hidden cost of keeping proprietary software alive
The larger cost of a homegrown system is often not the code you already paid for. It is the ongoing effort to keep it running. Gartner has estimated that around 70 percent of IT spending goes toward keeping existing systems running rather than building anything new.
That is the quiet trap. A system that once gave you an edge slowly becomes the thing your most skilled engineers maintain instead of improve. The real cost of running vault software rarely stops at the license fee or the payroll behind it. It is the growth you set aside while your team keeps the old system running.
And the costs rarely stay inside IT. When the system can’t keep up, onboarding a new client takes longer, audits and insurance reviews take more effort to satisfy, and the confidence clients place in you, the real reason they store with you rather than someone else, gets harder to reinforce. Left unaddressed, the system quietly caps how much new business you can safely take on.
When to replace it: a quick scale-readiness self-check
If you answer “yes” to three or more, it is worth planning a replacement.
- Does reconciling across vaults, clients, or systems still take manual effort every day?
- Would you struggle to prove a clean chain of custody for a single bar within minutes?
- Does onboarding a new client or vault require custom code or developer time?
- Would you lose critical know-how if one or two people left?
- Do your engineers spend more time maintaining the old system than improving anything?
- Is your exact, current position across every client and vault hard to see on demand?
And waiting rarely makes the move easier. The longer a homegrown system stays in place, the more workarounds and dependencies build uparound it, which is one reason so many replacements get delayed. In the 2025 Saritasa survey, half of respondents said the main reason they had not modernized was simply that the current system still worked, while fear of disrupting operations and data-migration concerns were among the next most-cited blockers.
Build vs. buy: rebuild it, or replace it with a WMS?
The build vs. buy question comes down to one thing: is your inventory system a source of competitive advantage, or a cost of doing business you keep re-paying? If it genuinely differentiates you, keep investing. If it is table stakes that drains your team, buying is often the more practical choice.
Let’s be fair, because sometimes building is still right. If your operation depends on a workflow no vendor supports, you have the engineering depth to secure and maintain it for years, and that system is a real reason clients choose you, a proprietary build can be worth it. Pretending every operator should buy would be dishonest.
For many, though, the math has shifted. Modern precious metals inventory management software now handles the ownership models, custody workflows, and integrations that once demanded custom code, while giving you the security, scalability, and ongoing development a homegrown system struggles to sustain. You get the specialization without carrying the entire maintenance burden alone.
What to look for in modern precious metals inventory software
Not every platform earns the name. When you evaluate one, treat these as non-negotiables.
- True ownership models: native support for allocated, segregated, and pooled holdings, not a workaround.
- Asset-level identity: tracking at the SKU, bar, or lot level with serial, weight, purity, and assay detail.
- Configurability: the ability to add clients, custody rules, and billing through configuration, not new code.
- Reconciliation and real-time visibility: one current picture across every client and vault.
- Control and security: dual-control workflows, role-based permissions, and audit-ready trails that support your compliance efforts rather than replace your responsibility for them.
- Integrations: clean connections to ERP, trading, and finance systems so data stops living in silos.
- Scale and continuity: a cloud-capable, supported platform that does not depend on one or two people to survive.
How to switch without breaking chain of custody
You can replace the system without breaking chain of custody by preserving history, mapping ownership precisely, and proving the new records before you rely on them. Migration is the real concern, and a fair one. Plan for those directly, and the odds move your way.
The sequence that protects custody looks like this:
- Clean and map your data first. Standardize units, confirm serials and weights, and map every allocated, segregated, and pooled holding before anything moves, so ownership is never ambiguous mid-migration.
- Carry the full history forward. Bring chain-of-custody and audit history into the new system so nothing about an asset’s past is lost.
- Run in parallel and reconcile to zero. Operate old and new side by side until the records match perfectly.
- Verify with a physical audit. Pull a sample of real bars from the vault and confirm they match the new digital record exactly. There’s no faster way to build confidence in a custody system than checking it against physical metal in your hand.
Done in this order, a migration protects custody instead of threatening it. Ownership never goes dark, the audit trail stays intact, and the physical audit at the end gives you, your clients, and your auditors a clean, verifiable starting point on the new system.
Get the 4-step migration playbook as a printable worksheet.
What a modern precious metals inventory platform looks like
A modern platform does the thing a homegrown system was always trying to do, only without the overhead. It holds every ownership model, preserves chain of custody across every movement, helps you reconcile physical and financial records, and adapts to each client’s rules through configuration instead of new code.
Datex Footprint WMS was shaped by decades of work alongside vault operators, depositories, and high-value logistics providers, which is why it is built to carry that complexity rather than fight it. No band-aid fixes, and no maintenance responsibility that permanently sits with your internal staff.
The bottom line: you are not managing inventory, you are safeguarding trust
Strip away the software question and here is what remains: a precious metals operation does not really manage inventory. It safeguards other people’s wealth, and its own reputation for proving that wealth is exactly where it should be. Ownership records, chain of custody, and a defensible audit trail are the product. The metal is what they point to.
A homegrown system can carry that for a while. The signs that it no longer can are the ones above: reconciliation by hand, onboarding that needs code, audit trails you reconstruct, a position you cannot see in real time, and a platform only two people understand.
When you notice them, the sequence is simple. Map where the manual work and reconciliation gaps live. Attach a real number to the engineering time the legacy system consumes, beyond whatever it costs to license. Then pressure-test a modern platform against your most demanding custody workflow, not your simplest one.
To see how a platform designed for high-value assets manages custody, ownership, and reconciliation, take a look at how Footprint WMS tracks allocated, segregated, and pooled holdings at the bar level, preserves chain of custody across every movement, and keeps an audit-ready trail.
Before you talk to anyone, score your current setup and map a safe migration.
Frequently asked questions
When should a precious metals company replace a homegrown inventory system?
Replace it when growth adds manual work instead of absorbing it, when reconciliation and reporting can no longer be fully trusted, or when maintaining the software consumes engineering time you need elsewhere. Triggers like a new vault, a new client custody model, a required integration, or an audit near-miss suggest the system has reached its ceiling.
Why do spreadsheets fail for bullion or vault inventory?
Spreadsheets lack the controls, permissions, and continuous audit trail a custody operation needs, and errors are common. Recent surveys report that many operations teams hit spreadsheet errors daily and that careful manual review catches only about half of them. On a bar list, one wrong serial or weight is a custody discrepancy, so that error rate is a real risk.
Should precious metals companies build or buy inventory software?
Build if the system genuinely differentiates you and you have the engineering depth to secure and maintain it for years. Buy if it is table stakes that drains your team. Many operators now find that configurable precious metals inventory management software handles their custody and ownership complexity without the ongoing maintenance burden of a homegrown build.
What is the total cost of ownership of a proprietary vault system?
It is far more than the original build. It includes ongoing engineering time to maintain and secure the system, the technical debt that accumulates as the business grows, and the opportunity cost of work your team cannot ship because it is keeping the old system running. Industry research consistently shows most IT effort goes to maintenance rather than new development.
How do you replace a precious metals inventory system without breaking chain of custody?
Clean and map your data first, carry the full chain-of-custody and audit history into the new system, run old and new in parallel until records reconcile to zero, then verify with a physical spot audit against real bars. Planning for data quality, integration, and change management up front keeps the migration on schedule.
What should a precious metals WMS support that a homegrown system doesn’t?
Built-in support for allocated, segregated, and pooled holdings; tracking down to the SKU, bar, and lot level with serial numbers, weight, purity, and assay data; dual-control approval steps; permissions based on role; live visibility across every vault; and straightforward connections to ERP and trading systems; and audit-ready trails that support compliance efforts. Just as important, it should absorb new clients and rules through configuration rather than custom code.
