The office spreadsheet says there are five filters on a van. The technician says two have already been used. Another job needs three tomorrow, and someone has ordered a box because nobody can confirm what is available.
The purchase may be sensible. It may also be unnecessary. Finding out means checking a message, two spreadsheets and a supplier email.
That is why inventory stacks sprawl when replenishment is still manual. Each gap in the stock record creates another place to track the answer. Software as a service (SaaS) sprawl can follow when the business buys separate tools for scanning, requests, approvals and reporting before deciding how the underlying records should work together.
Why inventory stacks sprawl when replenishment is still manual
Consider a fictional maintenance contractor with one storeroom, three service vans and Xero for accounting. It tracks parts in spreadsheets. There is no inventory-platform subscription to cancel.
The office maintains the master list. Each technician has a van sheet. Parts requests arrive through messages, purchase decisions sit in email, and supplier deliveries are checked against a printed order. Someone reconciles the copies when there is time.
Each record has a reasonable purpose. The problem is that several people can change their version of the same quantity without seeing the others' work. A van count becomes stale after the next job. A purchase marked “ordered” disappears from the shortage discussion even though nothing has arrived. A reservation exists in somebody's memory.
Now the business considers a barcode app. Then an approval tool. Then a connector to get the approved request into accounting. Each addition can improve one task while leaving the office to assemble the complete stock picture.
SaaS sprawl can grow from that fragmented responsibility. Tool sprawl begins before any new subscription is purchased. More applications can make the work faster locally while preserving the same uncertainty between steps.
The spreadsheet starting point is supported by a public practitioner account. A service manager described roughly 20 technicians carrying fire-alarm parts, monthly spreadsheet inventories and manual office review. Their requirements included mobile and desktop access, van locations and barcode scanning. It is one person's reported experience, rather than evidence of how common the problem is. Service-van inventory discussion.
Stock on hand is only one part of the answer
The question “Do we have three filters?” needs more than a quantity. The office needs to know whether the parts are usable, committed, accessible and expected to arrive before the job.
| Record or event | What it tells you | What it does not establish |
|---|---|---|
| Physical stock | Parts counted or recorded at a location | That every part is usable or available for another job |
| Reservation | Stock committed to a particular job | That the technician has collected or consumed it |
| Transfer dispatch | Parts have left the source location | That the destination has accepted the full quantity |
| Approved purchase | Someone authorised a defined purchase | That an order reached the supplier or goods arrived |
| Goods receipt | Specified quantities arrived and were accepted | That the whole order arrived or the supplier bill is correct |
| Supplier bill or payment | A financial obligation or payment is recorded | That stock should be received again |
Workflow ownership means maintaining these distinctions as the work proceeds. A dispatch and a receipt belong to the same transfer, but they are not the same event. A purchase request can be approved while its supplier order remains unsent.
Software visibility improves when those relationships are explicit. The office can see a part in transit and the person responsible for resolving a shortage. A dashboard that displays one green “complete” state can hide both.
Check the software's native operation before adding more tools
Manual replenishment does not prove that packaged inventory software lacks the required capabilities. Unleashed's published offering includes partial receiving, landed costs and inventory planning. Odoo's replenishment documentation explains forecast-based replenishment and planning horizons.
Those are reasons to demonstrate suitable products before assuming a custom build is necessary. For an existing user, investigate configuration, purchased scope and adoption before commissioning a second approval or stock application.
For the fictional contractor, the decision is buying a suitable product or owning a bounded stock operation. It does not need to rent a large platform before it is allowed to consider the second path.
The inventory management software comparison separates simple supplies tracking, purchasing, manufacturing, warehouse fulfilment and retail. A service van can be a stock location without needing all the processes of an ecommerce warehouse.
Too many software tools is a useful warning, but the count alone is not the diagnosis. Xero and a stock application can perform distinct jobs. Two independently editable stock totals for the same van create a different problem.
Follow one shortage through a SaaS stack audit
Take a recent part shortage and reconstruct the work. Where was the need first recorded? Who checked available stock? Did they see existing reservations and incoming purchases? Who decided to transfer or buy? What showed that the decision had actually been carried out?
Mark every point where someone copied a fact, hunted for a status or asked another person to reconcile two versions. Also mark the work that should remain: counting parts, approving expenditure, inspecting damaged goods and resolving a real discrepancy.
That distinction matters when measuring software waste. Recording a consumed part is necessary. Re-entering the same movement from a message into a van sheet and then into the office workbook may be avoidable. A physical count still has value even after the software is reliable.
SaaS spend management should start with this operating evidence. Otherwise, the business can cut a visible subscription while leaving the expensive manual reconciliation intact—or buy a new subscription that merely moves it elsewhere.
An illustrative cost of keeping stock copies aligned
This example uses Australian dollars (A$) and assumed activity levels. It is not measured customer performance, a wage benchmark or a forecast of savings.
Assume three vans each produce five stock entries per working day over 20 days: 300 entries a month. The office spends one additional minute per entry copying information already recorded elsewhere. Separately, it spends 15 minutes each working day aligning the van and master sheets, plus 15 minutes on each of 12 enquiries about purchase or receipt status. Value that time at A$70 an hour.
| Avoidable activity in the illustration | Monthly time | Monthly value | One year | Three years | Five years |
|---|---|---|---|---|---|
| Re-entering 300 existing records | 5 hours | A$350 | A$4,200 | A$12,600 | A$21,000 |
| Aligning the separate stock sheets | 5 hours | A$350 | A$4,200 | A$12,600 | A$21,000 |
| Chasing already-recorded order or receipt status | 3 hours | A$210 | A$2,520 | A$7,560 | A$12,600 |
| Total modelled administration | 13 hours | A$910 | A$10,920 | A$32,760 | A$54,600 |
The activities must be measured separately to avoid counting the same hour twice. Workload and hourly value remain constant; the model excludes tax, inflation, necessary stock control, stock losses and any new software costs.
There are no assumed inventory subscription savings. The contractor starts with spreadsheets, and its accounting service remains. Removing half the modelled administration would release 6.5 hours a month, valued at A$455. That is staff capacity, not an automatic reduction in payroll or cash expenditure.
For a business that already pays for overlapping trackers, connectors or unused software licences, add only charges that can actually be cancelled under its contract. Recurring software costs do not fall merely because fewer people open an application.
A SaaS total cost of ownership comparison must include the usable product package, implementation, migration, training and support. For an owned system, include development, hosting, security updates, backups, recovery and integration maintenance. Compare those costs against a conservative estimate of avoidable work, rather than treating the five-year illustration as money available to spend today.
Software consolidation needs a complete operating boundary
A barcode screen and a purchase-approval form would leave much of the contractor's work unresolved. Drivers and back-office staff need a coherent application they can use throughout the stock cycle.
For this proposed operation, that means a shared parts catalogue and location balances, job reservations, consumption and returns, transfers, replenishment decisions, approved purchases, partial receipts, stock counts and a movement history. The office needs queues for discrepancies and outstanding work. Technicians need fast mobile entry, clear feedback and a way to correct mistakes.
Poor reception adds requirements. If selected entries can be queued offline, people need to distinguish pending work from accepted stock changes. Conflicts must reach someone who can resolve them. A device reconnecting must not replay a consumption entry twice.
These requirements can be delivered by a suitable packaged product or assessed as an owned application. Software consolidation succeeds when the old duplicate registers can retire because their responsibilities have a dependable home.
Keep Xero's financial role explicit
The proposed stock application can own physical quantities, reservations and movements while Xero retains the financial books. The office could review supplier invoice information against purchases and receipts before a controlled handoff creates a draft bill for financial review.
That is a proposed integration boundary, not a claim that the application or Xero connection has been built. It needs confirmed field mappings, permissions, duplicate prevention and recovery when a request succeeds but its acknowledgement is lost.
The design also needs an agreed treatment for valuation and cost of goods. Sending a bill successfully does not settle those accounting questions. A payment-status update must never create a second receipt for goods that already arrived.
Keeping this boundary explicit gives each system a reason to exist. The stock application explains the parts; the financial system explains the financial records and decisions within its scope.
AI can help build the application without becoming the stock authority
Artificial intelligence (AI) can help turn interviews and spreadsheet examples into software, propose item mappings and explain exceptions. Existing vendors also provide planning and AI assistance; Unleashed advertises AI inventory tools. AI alone is therefore a weak reason to replace a working product.
For the contractor, the useful question is whether a tailored application can support the whole operation at an acceptable ongoing cost. Live AI inside the finished system is optional. Deterministic stock arithmetic, permissions and reliable recovery are not.
If AI suggests a purchase, show the evidence: usable stock, reservations, incoming orders, supplier pack sizes and lead time. Let the authorised person decide. An assistant should not infer that an emailed invoice proves delivery or silently accept an ambiguous part-code match.
Retire the spreadsheets through a reconciled cutover
Before moving authority, agree the catalogue, units and pack conversions. Count the storeroom and vans, identify damaged and reserved stock, and reconcile open purchases and transfers. Preserve the original sheets and record who accepted the opening balances.
Rehearse the change with incomplete work: a transfer in transit, an order partly received and a technician entry waiting to synchronise. Test export, permissions, backup restoration and the accounting handoff. Assign an operator to resolve exceptions and arrange ongoing technical support.
At cutover, stop the old sheets accepting new independent movements. Keep them as an archive. Rollback must account for parts consumed or delivered after launch; restoring an earlier balance does not reverse a physical event.
Start the buying or building decision with the shortage nobody can explain quickly. Ask each proposed solution to carry it from reservation through transfer or purchase, receipt, consumption and correction. Choose the option that makes the daily work dependable without adding another competing record.
The Sortly competitors guide and inFlow alternatives develop the smaller-operation buying choices. The stock replenishment workflow explores the proposed operating sequence. Where spreadsheets, several roles, record ownership and migration need investigation, explore Deep Discovery, available through a limited, account-enabled rollout.
