Attendance without a biometric machine: the options for contract staff at client sites
You can take reliable attendance without biometrics by having the site supervisor tap each staff member's face photo on a phone, at the site, at the start of the shift. The supervisor sees the person and their photo together, so the mark lands on the right person, with no fingerprint machine, no face scanning and no biometric data to store.
Part of our guide: Contract staff attendance: the complete guide for labour contractors
By HaazriPro Team · Last updated
Why do contractors look for attendance without biometrics?
Most attendance products sold to businesses use biometrics: a fingerprint machine on the wall, or an app that scans a face and matches it against stored data. In an office with one entrance, that can work well. At contract sites it often does not, for practical reasons:
- One device per site. An agency with thirty small client sites needs thirty machines, each with power, mounting and upkeep.
- Hard conditions. Construction and housekeeping staff often have worn, dusty or wet fingers. Night posts and basements have poor light for cameras.
- Moving staff. When staff rotate between sites, every device needs every person's enrolment.
- Sensitive data. Fingerprints and face templates are biometric data. Holding them for hundreds of contract staff is a responsibility many small agencies would rather not take on.
- Cost. Hardware, installation and replacements add up, and the client rarely pays for them.
The real goal is not "biometrics". It is making sure the mark is on the right person, so nobody is paid or billed for a day they were not there. There is more than one way to get there. This guide is part of our wider look at contract staff attendance.
How do the main methods compare?
Already pricing machines? The biometric machine vs phone app comparison covers hardware cost, setup time and what happens when a fingerprint does not read.
| Method | How identity is checked | Needs at the site | Offline | Data held |
|---|---|---|---|---|
| Fingerprint machine | Finger matched against an enrolled print | A device, power, mounting | Depends on the device | Fingerprint templates |
| Face recognition app | Camera image matched by software | A phone or tablet with a good camera and light | Depends on the product | Face templates |
| GPS check-in | A phone reports its location | A smartphone for every staff member | Usually needs network | Location history |
| Face photo grid | The supervisor sees the person and taps their photo | The supervisor's phone | Yes | One photo per staff member |
What does each method actually prove?
Fingerprint machines
A fingerprint scan proves that an enrolled finger touched that machine. It is strong evidence when it works. The weak points are the conditions around it: failed reads that need a manual override, machines that stop working and are not fixed for a week, and sites too small to justify a device. Every manual override is a gap in the record.
Face recognition
Face recognition software compares a camera image with stored face data and decides whether they match. It removes the device on the wall, but depends on light, camera quality and how the software was trained. It also means building and holding a biometric template for every staff member, which brings data protection duties. If you choose it, ask the vendor how the data is stored, who can see it, and how it is deleted when someone leaves.
GPS check-in
GPS apps record where a phone was. They prove the phone was at the site, not who was holding it. A staff member can hand their phone to a colleague. GPS also needs every staff member to have a working smartphone with data, which many contract staff do not.
Face photo marking
In HaazriPro, every staff member's photo is taken once, when they join. At the site, the supervisor's phone shows a grid of faces for everyone assigned to that site and shift. The supervisor looks at the person in front of them and taps their photo to mark present, late, half-day, leave or absent. Nothing is scanned or matched: the photo is there so the supervisor picks the right person instead of the right-sounding name. The proof is the supervisor's mark, made at the site, against a face they can see.
Is face photo marking good enough to stop proxy attendance?
Proxy attendance needs someone to mark a person who is not there. With a name register, that is easy: one tick on one row. With a face grid, the supervisor has to deliberately tap the face of someone who is not standing in front of them. That is no longer a mistake; it is a decision, and it is recorded. Combined with a time stamp on every mark, the owner can see when each day was marked and by whom.
It is not a lock that no one can open. A supervisor who wants to cheat still can, just as someone can override a fingerprint machine. What it removes is the far more common problem: honest mistakes between similar names, and friendly marks made without thinking. Our guide on why name-based registers cause pay disputes shows how much those cost.
HaazriPro works this out every month from your attendance, for every staff member and site. Start 30-day free trial
Worked example: equipping 25 small sites
An agency runs 25 client sites with 2 to 12 staff each, about 180 staff in all, and one supervisor for every few sites.
| Question | Fingerprint machines | Face photo grid |
|---|---|---|
| Devices needed | 25, one per site | None extra: supervisors use their phones |
| Enrolment | Each staff member at each site they may work | One photo per staff member, once |
| Site with no power or network | Depends on the device; often paper backup | Works offline, syncs later |
| Staff transferred mid-month | Re-enrol or update each device | Change the site assignment |
| Biometric data held | Fingerprints of 180 people | None |
For an agency like this, the cost of biometrics is not mainly the price of the machines; it is the upkeep and the gaps when a device is down. A method that runs on the supervisor's phone moves with the supervisor.
Which method suits which site?
- A large factory or office with one entrance and stable staff: a fingerprint machine or face recognition can work well, if the client provides power and space.
- Many small sites, rotating staff, a supervisor visiting each site: face photo marking on the supervisor's phone fits best.
- Sites with no network: choose a method that saves marks locally and syncs later; see our guide on offline attendance.
- Staff spread across many client locations: the method must follow site assignments; see multi-site attendance.
What about the records the law asks for?
Whatever method you use, the output must be a proper attendance record. For central-sphere establishments, the 2026 central rules under the Labour Codes require an attendance register-cum-muster roll (Form XIV), which may be kept electronically. State-sphere establishments follow their state's rules. A method that stores marks per staff member, per day and per site can produce that record. If you are still on paper, start with a clean monthly attendance sheet.
Sources
- OSH and Working Conditions (Central) Rules, 2026, G.S.R. 345(E): registers, Form XIV (eGazette)
- PIB, 21 Nov 2025: the four Labour Codes made effective
This guide explains the rules in plain language. It is not legal or tax advice. Laws and rates change, so check your data protection and register obligations with your lawyer before you act on it.
Attendance by face photo: no machine, no biometric data
Supervisors mark attendance by tapping each staff member's face photo, even offline. The salary sheet and the client bill are ready at month end.
Start 30-day free trial