Varför tid- och utgiftssystem inte pratar med lönesystemet
87% of organisations run expense and payroll systems without a shared data foundation. Here are the structural reasons why the connection fails, and what it costs.
The data is already in the system. Why is someone still typing it twice?
TL;DRDespite widespread adoption of digital expense tools, 87% of organisations still run payroll and finance systems without a shared data foundation. Manual re-entry persists, error rates hover around 20%, and in many mid-market companies a single finance controller is the only person who understands how money moves between systems. This article breaks down the structural reasons — technical, organisational, and timing-related — why expense and payroll systems consistently fail to connect, and what it costs when they do not.
A manager approves a travel expense report. The employee has been waiting three weeks for reimbursement. The finance team's expense tool says approved. The payroll system has no record of it. Somewhere between approval and payment, the data fell into a gap — and nobody is exactly sure where.
This scenario plays out every month in organisations across Europe, and it is not primarily a technology problem. It is a structural problem. Time tracking, expense management, and payroll each evolved as separate disciplines, built on separate architectures, owned by separate teams, and running on separate schedules. When they need to exchange data, they require a layer of coordination that most organisations have never formally designed.
Here is why the connection breaks — and why it is harder to fix than it looks.
Three clocks, no synchronisation
The most underappreciated root cause of expense-to-payroll failure is not technical. It is temporal. Three independent timers are running simultaneously, and none of them are synchronised.
Clock one: expense incurrence. A business trip happens. Costs are paid. The employee collects receipts — or tries to.
Clock two: expense submission and approval. The employee submits the report. The manager reviews it. A second approval may be triggered above a certain threshold. This process typically takes days to weeks — and 58% of European employees worry that delayed reimbursements will impact their personal finances, according to GBTA research across Nordic and European markets.
Clock three: payroll. In Denmark and across most of Nordic Europe, payroll runs monthly, with a fixed cutoff date. An expense approved one day after that cutoff waits until the following month. Most organisations have never formally communicated this to employees or managers. The submission window required to catch a given payroll run is undocumented policy — learned through missed reimbursements, not onboarding.
The consequence is predictable. 46% of European employees say the expense reclaim process is so difficult they simply avoid submitting smaller amounts altogether, according to that same GBTA survey. From the employee's perspective, the cost-benefit calculation of submitting a €15 taxi receipt does not make sense when the process is unclear and reimbursement unpredictable. From the organisation's perspective, this represents a systematic underreporting of actual business costs — costs that are incurred, absorbed by employees, and never reconciled.
The technical reality: identity crisis at the data layer
When organisations do attempt to connect expense and payroll systems, they encounter a second class of problem: the systems do not speak the same language about the same people.
The expense system knows an employee by one identifier. The payroll system uses another. If those identifiers do not match exactly — because of a system migration, a name change, a rehire, or simply inconsistent setup — records either route silently to the wrong account or fail to transfer entirely. No error message. No alert. The reimbursement simply does not arrive.
Beyond identity, every payroll provider uses a proprietary schema. ADP, Visma DataLøn, Danløn, Lessor, and SD Worx each have different field names, different data structures, and different rules for what constitutes a valid submission. Building a mapping from an expense tool's category structure to a payroll provider's specific earnings and deduction codes requires careful configuration — and it breaks whenever either system updates its schema without notice.
The financial implication is significant. Research from Bindbee puts the cost of building a single advanced payroll integration at $30,000 upfront and $50,000–$150,000 per year to maintain. Integration maintenance consumes up to 30% of engineering capacity at B2B SaaS companies that build their own connectors. For mid-market finance teams without dedicated IT resources, a custom integration is not a project you build once and leave running. It is an ongoing maintenance commitment that competes with every other technology priority.
It is also worth noting what happens when an integration only partially succeeds. In SAP Concur's architecture, if an expense report fails GL/FI posting due to a mapping error, no payroll posting occurs either — the entire chain breaks at the first failure. Finance administrators must manually identify the failed documents and trigger resubmission. The integration's promise of automation creates a new manual remediation workload whenever it breaks.
The ownership gap: when finance and HR run parallel processes
Expense management is a Finance function. Cost control, GL coding, policy enforcement, VAT reclaim — these belong to the finance team. Payroll is an HR function. Compensation, employment tax, e-indkomst reporting — these belong to HR or a payroll bureau.
The problem is that only 13% of organisations have HR and finance on a single native platform, according to Paylocity's 2026 State of Payroll research. The remaining 87% operate parallel processes from different vendor systems, with different data governance policies, different employee records, and no shared owner responsible for the connection between them.
Nobody is accountable for making the handoff work. Finance assumes HR's payroll system will accept the data. HR assumes Finance's expense tool exports it correctly. When a reimbursement does not arrive, both teams examine the other team's system first — and the investigation begins from mismatched assumptions about what was supposed to happen.
For medium-sized organisations — those with 50 to 250 employees — the challenge is compounded by growth. They have accumulated enough complexity to need multi-entity support, layered approval chains, and multi-currency handling, but often lack the IT capacity to build formal integrations. 55% of medium-sized organisations specifically report struggling with manual expense processes, higher than either small or large peers, precisely because they have outgrown manual methods without yet having the infrastructure to replace them.
The shadow spreadsheet: a single point of failure hiding in plain sight
In most mid-market finance teams, there is one person who understands exactly how data moves between the expense tool and the payroll system. They built a macro. Or a monthly export script. Or a set of manual steps performed on the last Thursday before payroll closes.
Wipfli describes these unofficial workarounds as shadow automation — processes that bridge system gaps, influence close and forecasting, and run entirely outside formal controls. The individual who maintains them is rarely identified as a single point of failure, but that is precisely what they represent.
When that person is on leave, the process slows. When they leave the company, it breaks entirely — often mid-payroll-run, under deadline pressure, with no documentation of the underlying logic. The organisation discovers its actual dependency on this person at the worst possible moment.
The ongoing cost of this arrangement compounds the structural risk. Processing a single expense report manually takes an average of 20 minutes and costs $58 (Corpay). Nearly one in five reports contains an error, each requiring an additional $52 and 18 minutes to correct. For a finance team processing 500 reports monthly, error correction alone adds approximately $5,000 per month in staff time — before any structural failure occurs. Add payroll errors, which average $291 per incident to resolve and affect approximately one in five payroll runs, and the cumulative cost of a poorly integrated system becomes significant.
What a functioning connection actually requires
Organisations that have resolved this reliably share a common architecture. They use native integrations between their expense management system and their payroll system — pre-built, maintained connectors that eliminate schema maintenance, enforce the approval gate before any data reaches payroll, and remove the dependency on a single person who understands the data flow. For Nordic companies, this means choosing an expense tool with a direct connector to your specific payroll provider: DataLøn, Danløn, Lessor, or SD Worx.
They also make a deliberate architectural choice about what routes through payroll versus accounts payable. Mileage and per diem allowances are structurally tied to pay periods and appear on the payslip for SKAT reporting — these belong in payroll. Variable, receipt-based expenses — hotels, transport, client meals — move faster and with cleaner tax treatment through direct AP payment. Mixing these paths creates both compliance risk and reconciliation complexity.
Most importantly, there is a named owner. Someone is accountable for the handoff between expense approval and payroll posting. The connection is documented, monitored, and tested before every payroll run — not discovered broken at 4pm on payday.
The three clocks are still running at different speeds. But when the integration is designed correctly, someone has finally synchronised them.
Frequently asked questions
What is the most common reason expense and payroll systems fail to connect?
The single most common failure is an employee ID mismatch between systems. The expense tool assigns one identifier, the payroll system uses another, and without a validated translation table, records either fail silently or route to the wrong employee. This is compounded by earnings code mismatches — where the expense category in your expense tool does not map correctly to the payroll provider's specific deduction or allowance codes.
Why does the payroll cycle timing cause expense reimbursement problems?
Expense management and payroll run on fundamentally different cycles. An expense is incurred on a business trip, submitted days later, approved days after that, and then must wait for the next payroll run — which in Denmark typically runs monthly. If approval falls one day after the payroll cutoff, the employee waits another full month. Most organisations have never formally defined the submission window required to catch a given payroll run, so employees learn the hard way rather than from policy.
Who owns the expense-to-payroll integration — Finance or HR?
That ambiguity is part of the problem. Expense management typically sits with Finance (cost control), while payroll sits with HR (compensation). Research shows only 13% of organisations have both functions on a single native platform. The other 87% rely on some form of manual reconciliation or point integration between separately owned systems. Without a clear integration owner, gaps persist because neither team has the mandate or visibility to close them.
What is shadow automation and why is it a risk for finance teams?
Shadow automation refers to unofficial workarounds — Excel macros, manual export scripts, monthly copy-paste routines — that finance teams build to bridge the gap between their expense tool and payroll system when no formal integration exists. The risk is concentration: often only one person understands the logic. When that person goes on leave or leaves the company, the process breaks with no documentation trail. These shadow systems also bypass formal controls, meaning approval and policy checks can be circumvented without audit visibility.
