An HRIS Runs Your Policies. It Doesn't Write Them.
Most HR system rollouts that stall in the Philippines stall on missing data and unwritten rules, not on the software. What that looks like in practice, and what to settle before you choose a system.
The software is rarely what stalls
When an HR and payroll rollout drags, the first suspect is usually the software. In our experience it is rarely the cause. Setting up a company in a system takes days. What takes weeks is everything the system needs from the company: employee records that are complete, and rules that someone has actually decided and written down.
An HRIS does not invent your policies. It applies them, the same way, to every employee, every cut-off. If a rule only exists as "how payroll has always done it," the system cannot copy it. It needs the rule spelled out.
That is the uncomfortable part. A new system does not fix an unclear policy. It exposes it, usually on the first payroll.
Two kinds of gaps
Almost every delay we see falls into one of two groups.
Data that is not available. Birth dates missing from half the masterlist. TIN or SSS numbers kept in a different file per branch. Project-based contracts with no end date. Leave balances nobody can trace back. None of this is a software problem, but the system cannot be set up without it.
Rules that are not written down. The rule exists, in the sense that payroll has been running for years. But it lives in a spreadsheet formula, or in one person's judgement, and it has never been put into words. When someone asks "what exactly happens if...", nobody is sure.
The second group is harder, because it looks like the first group's problem until the numbers come out wrong.
What unwritten rules look like
These come from real implementations, with names removed.
The grace period that only applied one way. A company had a 15-minute grace period. Arriving 20 minutes late was charged as 20 minutes, not 5: once you pass the grace, it no longer counts. Everyone agreed on that. Nobody had decided whether the same applied to leaving early. The old payroll forgave the first 15 minutes of undertime. The payroll officer expected the full amount. Both were following "the policy."
Seconds on the clock. An employee clocked out at 5:13:54 PM on a shift that ended at 5:30. Is that 16 minutes of undertime or 17? It sounds trivial until it is multiplied across hundreds of employees and every cut-off, and the payroll officer's manual count does not match the system's.
Allowances that may or may not be part of basic pay. Rice, clothing and medical allowances were paid every cut-off. Do they count when computing the daily rate for absences? For overtime? For late deductions? The company's own spreadsheet used one answer for absences and another for tardiness. Nobody had chosen that on purpose.
A half day nobody had defined. Is an employee who arrives at noon half present and half absent? Absent for the morning and late for the afternoon? Present with four hours of tardiness? Each answer is defensible. Each gives a different payslip. The register and the deduction have to agree, and that only happens when the rule is decided once.
Contracts with no end dates. More than a hundred project-based contracts were loaded without end dates, because the end dates lived in a separate list. Contract-end reminders never fired. The data existed. It just was not where the system could use it.
The daily rate divisor. Is a monthly salary divided by 261, 313 or 365 days to get the daily rate? The answer changes every absence deduction and every overtime computation. In many companies it was set years ago by whoever built the spreadsheet, and nobody remembers why.
Mid-year transfers. An employee moved between two companies in the same group in June. For the year-end tax and BIR Form 2316, the new employer needs what the old one paid and withheld. If nobody collects it, the year-end tax is computed on half a year.
None of these are exotic. They are ordinary questions that a manual process can avoid answering for years, because a person quietly handles each case.
Why a system makes it visible
A person running payroll in a spreadsheet can make a judgement call: this employee was late for a good reason, that allowance probably should not count. Each call is reasonable. Over time, the calls become the policy, but they are never written down, and they are not always consistent.
A system cannot make judgement calls. It needs a value for the grace period, a yes or no for whether it applies to undertime, a list of which allowances count as basic pay. When those answers are missing, set-up stops. When they are guessed, the first payroll does not match the last one, and the new system gets the blame.
This is why the first parallel payroll run is so useful. Every difference between the old payroll and the new one is a question about a rule. Usually it is not a bug. It is a decision nobody made.
What to settle before you choose a system
You do not need perfect data to start. You do need to know where the gaps are. Before you compare HR systems, settle these:
- The masterlist. Birth date, hire date and the four government numbers for every employee, in one file.
- Salaries and allowances. Each employee's current rate, and for each allowance: is it part of basic pay, is it taxable, is it de minimis?
- Attendance rules. Grace period, whether it applies to undertime, rounding, and when lateness becomes a half day or an absence.
- Leave. Types, yearly credits, accrual, carry-over, and a signed-off balance per employee as of go-live.
- Payroll rules. Cut-offs and pay days, which cut-off each government contribution comes from, and who is entitled to overtime and holiday pay.
- Year-to-date figures. If you go live after January: pay and tax so far, previous-employer figures for this year's hires, and running loan balances.
- Approvers. Who approves leave, overtime and DTR corrections in each department.
For each rule, the test is simple: could someone new to the company read it and compute the same number you would?
Our free HR System Readiness Check walks through these areas in five short steps and lists what to prepare for each gap. It works whichever system you end up choosing.
Where software does help
Once the rules are clear, a system earns its keep. It applies the same rule to every employee every cut-off. It keeps a record of who changed what. It produces the government reports from the same numbers as the payslips. It stops the quiet drift that happens when each case is handled by hand.
But the order matters. Decide the rules, write them down, collect the data, then choose the system that runs them. The other way around, you end up writing your policies in the middle of your first payroll, under a deadline, with employees waiting for their pay.
For a week-by-week list of what to gather, see the HRIS implementation checklist.