The 8 modules worth testing
Every vendor's homepage lists the same eight capabilities. The list below isn't the differentiator, how each one behaves under a real payroll run is. For each module, don't ask if the vendor "has" it. Run the specific test.
- Employee records & self-service
- One record from offer letter onward. If a hired candidate has to be manually re-entered as an employee, that's a re-entry cost you'll pay every single month.
- Attendance & leave
- Should feed payroll automatically. If attendance and payroll are separate exports someone reconciles by hand, the "automation" is cosmetic.
- Payroll & statutory compliance
- PF, ESI and TDS are non-negotiable, and they're where "feature complete" demos most often fall apart on real numbers.
- Recruitment & ATS
- Not mandatory for every buyer, but if you're hiring more than two or three roles a month, a disconnected ATS means retyping every hire's details twice.
- Onboarding & exit
- Structured checklists on both ends, not a folder of PDFs someone forwards. Exit matters as much as onboarding: a missed full-and-final settlement is the most common post-exit complaint.
- Performance management
- Goal-setting and reviews should live where managers already work. A separate tool with a separate login is where performance data goes to die.
- HR reporting & analytics
- The real test isn't whether dashboards exist, it's whether the underlying data is already joined. Reports built by stitching three exports together in Excel aren't reporting, they're a workaround.
- Data protection & DPDP compliance
- Every module above touches employee personal data. If the vendor can't explain their consent, retention and erasure flow in plain language, that's the answer.
Questions to ask every vendor
These are the questions that a polished demo doesn't naturally answer, because the honest answer isn't always flattering. Ask them anyway.
- Show me an actual PF/ESI challan generated from a real sample run, not a screenshot from your deck.
- What happens to our data if we cancel mid-contract? In writing, not verbally.
- Can an employee see their own leave balance and payslip without emailing HR?
- Who at your company do we call when payroll fails to run on the 1st, and what's the actual response time?
- What's included in the quoted per-employee price, and what triggers an extra charge at renewal?
- Can we export all our data, in full, in a usable format, on our own, at any time?
Red flags that predict a bad renewal
- ✕Pricing "per seat" that quietly expands what counts as a seat by year two.
- ✕No sandbox with real sample data, only a scripted demo you can't click around in yourself.
- ✕Compliance features marked "coming soon" for anything statutory. This should already exist, not be on a roadmap.
- ✕Self-service that's a read-only PDF export, not an actual portal an employee can act inside.
- ✕Implementation timeline over 60 days for a team under 500 people. That usually means the product needs heavy manual configuration.
The India-specific compliance checklist
Global HRMS platforms retrofitted for India often get this section wrong. Confirm these explicitly, don't assume.
DPDP Act, 2023
- Explicit, revocable consent captured for candidate and employee personal data, not a buried clause in an offer letter.
- Data retained only as long as the stated purpose requires, with a defined erasure process after exit.
- A documented breach-notification process, since the Act puts real obligations on the data fiduciary, not just the vendor.
Statutory payroll
- PF and ESI calculated on current wage ceilings, generated as filing-ready challans, not just numbers on a payslip.
- TDS slabs updated for the current financial year without a support ticket.
- Full-and-final settlement handled as its own workflow, gratuity and leave encashment included, not bolted onto regular payroll.
The new Labour Codes
India's four Labour Codes are being phased in state by state. A vendor that can't describe, specifically, how their wage and working-hours definitions will adapt as each state notifies its rules is telling you compliance is reactive, not built in.
See it on your own org
A 30-minute walkthrough shows what SKILWI surfaces about a team like yours, before you commit to anything.
One system vs. a system plus a decision layer
Most companies don't need to build anything, buying is almost always right at this scale. The real choice is narrower: a single HRMS that records what happened, or that same record-keeping with a layer on top that tells you what to do next, priced in rupees. An HRMS answers "who is on leave." A decision layer answers "should we hire or upskill for this gap, and what does each option cost." They aren't competing purchases, most companies keep the system of record and add the decision layer once the basics are solid.
Common questions
How long should an HRMS evaluation take?
For a company under 500 employees, two to four weeks is enough if you run real data through a sandbox rather than watching slides. Longer than that usually means the vendor list wasn't narrowed early, not that the decision is genuinely hard.
Should recruitment and ATS be part of the same system as payroll?
Not required, but it removes a re-entry step: a hired candidate becomes an employee record without retyping their details. If you hire fewer than two or three people a month, a separate ATS is fine. Above that, the re-entry cost adds up.
What's the single biggest mistake companies make choosing an HRMS in India?
Evaluating on feature checklists instead of a real payroll run. A demo environment with sample data hides exactly the problems, wrong PF ceiling handling, a TDS slab that wasn't updated, a leave policy that doesn't match Indian norms, that only show up once real numbers go through the system.