A Non-Aadhaar Attendance App for Private Employers

The Aadhaar Enabled Biometric Attendance System was built for government offices and public-sector bodies, and it is the reason many Indian buyers now assume attendance means Aadhaar. For a private company it is a poor fit. You inherit a device estate, an authentication dependency and a category of personal data you had no business reason to hold — all to prove a fact that a login and an approval already prove.

The legal position is narrower than the market assumes. Section 7 of the Aadhaar Act, 2016 ties mandatory Aadhaar authentication to subsidies, benefits and services funded from the Consolidated Fund of India. That is not a private employer paying salaries out of its own revenue. Puttaswamy (2018) struck down Section 57, which had been read as permitting private bodies to demand Aadhaar authentication under a contract. Private-sector Aadhaar use has been treated as voluntary and constrained ever since, which is precisely the ground on which an employee can decline.

The clearest way to describe a non-Aadhaar attendance app is by what is absent. None of the following is captured, stored, hashed or sent anywhere, because none of it is needed to know that someone worked.

• No Aadhaar number, virtual ID or last four digits.
• No Aadhaar authentication, eKYC or AUA/KUA integration.
• No fingerprint, iris or face template — the underlying biometric is never captured, so there is nothing stored and nothing to breach.
• No AEBAS device registration or dependency on UIDAI availability.
• No screenshots, keystroke logging or covert background agent.

Attendance is a claim about a person and a period, and it needs evidence. Workclave's evidence is a session: a block of work started and stopped under a personal login, attached to a project, with a manager's approval recorded against it afterwards. The manager is a named human who knows what work happened that day, and their approval is kept with the session permanently.

That structure is harder to game than a shared badge or a colleague's finger on a reader, because falsifying it requires a manager to approve work they know did not occur, on a record that keeps their name. It is also a richer record: a biometric punch tells you a body touched a scanner at 09:42, while an approved session tells you who worked, for how long, on which client's project, and who signed it off.

It is worth being precise here, because the market copy on this is frequently wrong. The Digital Personal Data Protection Act, 2023 does not create a separate 'sensitive personal data' tier for biometrics — that categorisation came from the SPDI Rules, 2011, which the DPDP framework moves away from. The DPDP Act applies one standard to all personal data: notice, consent or a listed legitimate use, purpose limitation, data minimisation, and deletion when the purpose ends.

Minimisation is what makes the Aadhaar-and-biometrics route hard to defend for attendance. If a login plus a manager approval establishes the same fact, then collecting an Aadhaar number or a fingerprint template collects more personal data than the purpose requires. Not collecting it is the shortest compliance story available: there is no consent artefact to maintain for data you never took, no retention schedule for a template that does not exist, and no breach-notification exposure for a database you never built.

The state Shops and Establishments Acts and the labour codes require an employer to maintain an accurate attendance register and produce it on inspection. They specify the record, not the capture mechanism. No provision names Aadhaar, fingerprints or face recognition as the required method, and an inspector's question is whether the register is accurate and contemporaneous — not which device produced it.

So the practical test is evidentiary rather than technological: can you show who worked, on which day, for how long, and can you show it was not reconstructed after the fact? Timestamped sessions with a recorded approval and a preserved edit history answer that. That is what the register needs to be, whatever collected it.

• Attendance registers required; the capture method left to the employer.
• Timestamped start and end times per session, not a daily guess.
• Manager approval recorded against each session, with the approver named.
• Corrections kept as an audit trail rather than silently overwriting the original.

Because private-sector Aadhaar use is voluntary, a policy built on it has a failure mode built in: a single employee declining leaves you running a parallel manual process for them, which is the process you bought software to eliminate. The same applies to biometric refusal on religious, medical or privacy grounds, and to contractors and consultants whose Aadhaar you have no standing to ask for at all.

A login-based record has no such branch. Permanent staff, contractors, remote engineers and on-site teams all produce the same session record in the same system, and the attendance report has one shape. Removing the opt-out removes the exception handling.

There is no attendance hardware in this model, which changes the arithmetic. No readers per site, no annual maintenance on devices, no replacement units, no field visit when a scanner fails at a branch office. Workclave is free for up to 3 users and ₹199 per user per month after that, with no base platform fee and no seat minimum.

Retiring the readers is not a prerequisite. Plenty of teams keep a gate device for physical access to the building and use Workclave for the attendance and billing record — the part an inspector or a client asks to see. The access-control decision and the attendance-record decision are genuinely separate, and treating them as one is what usually locks a company into biometric attendance it never explicitly chose.