Biometric, GPS or NFC: choosing how your staff clock in
Fingerprint readers, phone check-ins and NFC tags each solve a different attendance problem. How to pick the right one for each site, or mix them.
Frank Oppong Konadu5 min read
Ask ten HR managers why their attendance data is unreliable and you’ll hear the same few stories. Someone clocks in for a colleague who is still on the bus. A field officer “arrives” at a client site from their bedroom. A paper register gets filled in on Friday afternoon for the whole week.
The fix is rarely more supervision. It is choosing a clock-in method that suits how and where each group of people actually works. There are three main options today, and most growing organisations end up using more than one.
Biometric readers
A fingerprint or face reader mounted at the entrance. Staff scan in and out, and the device sends each punch to your attendance system.
Where it works best: one fixed location with a lot of people arriving at the same time. Factories, hospitals, schools, warehouses, head offices.
Why people choose it: it all but ends buddy punching, because nobody can lend a colleague their face. Clocking in takes about a second, so queues stay short even at shift change, and staff don’t need a smartphone.
What to watch
- Power and network. A reader that loses power or internet must store punches and send them later. Ask any vendor what happens to punches during an outage before you buy.
- Placement. Put readers where people naturally pass, under cover and away from direct sunlight, which can make face recognition less reliable.
- Enrolment. Someone has to register every new employee on the device. Make it part of onboarding, or the first week’s attendance will be a mess.
GPS check-in on a phone
Employees clock in from an app, and the app records where they are. A good setup only accepts a clock-in within a set distance of an approved site.
Where it works best: people who don’t start the day at one office. Sales reps, field officers, drivers, cleaners and security staff spread across client sites, and teams in several small branches where a reader at each one doesn’t pay for itself.
Why people choose it: there is no hardware to buy or maintain, and you get attendance from places you could never put a reader. It also lets staff do more than clock in, like apply for leave or check their payslip, from the same app.
What to watch
- The radius. Too tight and people standing at the gate get rejected. Too loose and “at the site” includes the café down the road. Around 100 to 200 metres suits most buildings. Large compounds need more.
- Phones and data. Some staff will have older phones, little data or none at all. Plan for them before launch, not after.
- Trust. Explain clearly that location is only checked at the moment of clocking in. People accept GPS attendance far more readily when they know it isn’t tracking them all day.
A simple way to decide
Count where your people are at 8am. If most of them are in one building, start with a reader. If they are spread across many places, start with phone check-in.
NFC tags and readers
Each employee gets a card or tag, and taps it on a reader at the door. You can also flip it round: fix a tag at a location and have the employee tap it with their phone to prove they were there.
Where it works best: sites where you want something quicker and cheaper than biometrics, such as retail branches, event venues and schools. Fixed tags are also good for patrol and inspection rounds, where you need proof that someone visited each checkpoint.
Why people choose it: cards and readers cost less than biometric devices, tapping is fast, and there is no personal data on the card itself.
What to watch: a card can be lent to a friend, just like a password. On its own, NFC doesn’t stop buddy punching. Pair it with occasional spot checks, or keep it for places where that risk is low.
Side by side
| Biometric reader | GPS on a phone | NFC card or tag | |
|---|---|---|---|
| Best for | One busy fixed site | Field and multi-site teams | Low-risk fixed sites, patrols |
| Stops buddy punching | Yes | Mostly, if the radius is tight | No, on its own |
| Hardware cost | Highest | None | Low |
| Needs a smartphone | No | Yes | No (cards) |
| Works without internet | Yes, if the device stores punches | Needs data at clock-in | Yes, if the reader stores taps |
| Personal data involved | Fingerprints or face | Location at clock-in | Card number only |
Most organisations need a mix
A hospital might use face readers at the main entrance, NFC at a small outpatient clinic, and phone check-in for community health staff. A security company might use phone check-in for guards across client sites and a reader at head office.
That is fine, as long as every method feeds one attendance record. The moment fingerprint punches live in one system and phone check-ins in a spreadsheet, you are back to reconciling by hand at month end, and that’s where overtime disputes and payroll errors come from.
Biometrics and the law
Fingerprints and face data are personal data, and in many places they get extra protection. Nigeria’s Data Protection Act 2023 and the EU’s GDPR both treat biometric data as sensitive, and Ghana’s Data Protection Act, 2012 (Act 843) applies to any employee data you collect.
Whatever country you’re in, a few habits keep you on the right side of the rules and of your staff:
- Tell employees, in writing, what you collect, why, and who can see it.
- Collect only what attendance needs. A fingerprint template for clocking in, nothing more.
- Keep it secure, and limit who can export it.
- Delete it when someone leaves.
- Offer another way to clock in for anyone who can’t use the reader.
This section is a starting point, not legal advice. If you are rolling out biometrics across a large workforce, check the requirements with your country’s data protection authority.
How Eban fits in
Eban takes punches from biometric readers, GPS check-ins from the Eban mobile app (only accepted inside each attendance site’s radius) and NFC readers, and puts them all in the same attendance record. Late arrivals, absences and overtime come from one set of data, and they flow straight into leave and payroll without anyone copying numbers between systems.
