HR software is where imported products most often fail in Pakistan, and the failure is rarely visible in a demo. Everything looks correct until the first payroll run, when the tax calculation is wrong, EOBI is missing, and the payslip does not match what the accounts department produces by hand.
The local requirements that break imported software
Income tax slabs
Salary tax in Pakistan is calculated on annual projected income against slabs that change with each finance bill, then deducted monthly. Systems built elsewhere use a flat percentage or a different model entirely.
The questions to ask: does the system update slabs when the budget changes, who does that update, and is it included in the subscription. A product where you must configure the slabs yourself each year will eventually be configured wrong.
EOBI and provincial social security
EOBI contributions from both employer and employee, and PESSI, SESSI or the relevant provincial institution depending on where the employee works. Contribution bases and ceilings differ, and a company operating in two provinces has two sets of rules.
Foreign products almost never have these. Ask specifically.
Provident fund and gratuity
Company specific rules, often with vesting periods and different treatment for resignation versus termination. The system needs to hold the rule and calculate the accrual, not just record a deduction.
Payslips and statutory reporting
A payslip employees recognise, plus the returns your finance team files. If the software cannot produce the supporting schedules, someone rebuilds them in Excel every month and the software has saved nothing.
The five modules worth paying for
1. Attendance that matches how you actually work
The most used module and the one most often mis specified. Before evaluating, write down: your shift patterns, whether staff work across locations, how late arrival is treated, whether there is flexible timing, and who approves exceptions.
Biometric devices work well for a single office. For field staff, mobile check in with location is usually better, and it raises a policy question you should settle before buying rather than after: how much location tracking of staff is appropriate, and have you told them.
2. Leave with a real approval chain
Annual, casual, sick, and whatever your policy adds. Balances that accrue correctly, carry forward rules, and an approval route that matches your actual hierarchy rather than a generic manager field.
The measure of success is whether leave requests stop arriving on WhatsApp. If they do not, the system is too slow or too confusing and staff have routed around it.
3. Payroll with the local rules above
Covered. This is where the buying decision should be made.
4. Employee records and documents
CNIC, contract, qualifications, dependants, bank details, and expiry dates for anything that expires. A document repository against the employee rather than a folder on someone’s laptop.
5. Self service
Employees see their payslips, apply for leave, and update their own details. This removes a genuinely large amount of HR administrative load, and it is the module employees notice, which helps adoption of everything else.
What gets bought and rarely used
- Performance appraisal modules. Bought with enthusiasm, used for one cycle, abandoned. The software is almost never why appraisal processes fail.
- Recruitment and applicant tracking, in companies that hire fifteen people a year. A spreadsheet is genuinely adequate at that volume.
- Learning management inside the HR system, which is usually weaker than a dedicated tool. See our note on what LMS platforms actually get used for.
- Engagement surveys and analytics dashboards, unless someone owns acting on the results.
Cloud or on premise
Cloud for almost everyone: no server, accessible across offices, and updates to tax rules arrive automatically, which is the strongest single argument.
On premise still makes sense where salary data policy prohibits external hosting. Understand what you take on: a server, backups someone verifies, and responsibility for applying every rule change yourself.
What to test before buying
- Run a payroll for an employee at three salary levels and check the tax against your accountant’s calculation.
- Show EOBI and provincial social security on the payslip.
- Process a mid month joiner and a mid month leaver.
- Process a final settlement including gratuity and leave encashment.
- Show attendance for an employee who worked a night shift crossing midnight.
- Export the schedules your finance team files.
Point one and point four are where most products fail, and both are avoided in standard demos.
Cost expectations
- Cloud HR and payroll: commonly 200 to 800 PKR per employee per month, with minimums for small companies.
- Biometric hardware: 20,000 to 80,000 PKR per device.
- Implementation and data migration: usually a one off charge, and worth paying for rather than doing yourself.
- Custom build: justified for large organisations with unusual structures, group companies with different entities, or where integration with an existing ERP is the main requirement. See custom versus off the shelf.
Frequently asked questions
Can I use international HR software in Pakistan?
For attendance, leave and records, yes. For payroll, usually not without local customisation, because tax slabs, EOBI and provincial social security are not supported.
What does HR software cost in Pakistan?
Typically 200 to 800 PKR per employee per month for cloud products, plus hardware if you use biometrics and a one off implementation fee.
Do we need biometric attendance?
Useful for a fixed office. For field or hybrid staff, mobile check in is more practical. Decide your policy on location tracking before you deploy it.
How long does implementation take?
Four to ten weeks for a company of fifty to three hundred, most of it employee data cleanup and one parallel payroll run against your existing process.
Ezitech builds and implements HR, payroll and ERP platforms for companies across 10 or more industries. Tell us how many employees and how many locations.