A single retail or BPO campaign in India can pull tens of thousands of applications through one requisition, and the software meant to organise that flood often just relocates the chaos into a database. Recruiters still chase candidates on WhatsApp, track pipelines on parallel Excel sheets, and lose days to interview coordination that no one owns. The applicant tracking system (ATS) was supposed to fix this. Whether it does depends far less on the length of its feature list than on a handful of things that rarely make it into a sales deck.
The Naukri JobSpeak Index reached 3,233 points in February 2026, up 12% year-on-year, with fresher hiring in the 0 to 3 year band growing 17%. Entry-level and campus volume is where ATS decisions get stress-tested, since application counts run highest and recruiter patience runs thinnest there. Cognizant CEO Ravi Kumar S told the company’s Q1 2025 earnings call that it would hire 20,000 freshers in the year to reshape its delivery pyramid. At that scale, the wrong tool doesn’t slow a team down. It quietly breaks it.
Why Scale Breaks Most ATS Deployments
Scale is the point at which an ATS either earns its cost or exposes itself as expensive shelfware. High-volume hiring compresses every weak link, and a tool that looked fine in a demo with 40 test applications behaves differently when a campus drive dumps 4,000 into the queue overnight. The failure is rarely dramatic. It shows up as recruiters reverting to spreadsheets, duplicate records, and a pipeline no leader can see.
Nishi Ojha, Senior Manager – Human Resources at Tripare AI, puts the core problem in plain operational terms, and the observation is worth sitting with before evaluating any platform.
“One of the biggest mistakes is assuming that buying an ATS automatically solves a hiring problem. It doesn’t,” Ojha emphasises. “If your hiring process is unclear, the ATS will simply make that unclear process digital. I’ve seen teams still depend on Excel and WhatsApp even after having an ATS because basic things like ownership and turnaround time were never fixed.”
That framing reorders the whole evaluation. The software is a multiplier, and it multiplies whatever process it inherits. A hiring workflow with undefined ownership and no turnaround-time discipline doesn’t get fixed by a licence purchase. It gets digitised, chaos and all. Indian IT services firms and global capability centres running perpetual bulk requisitions feel this first, because their volume leaves no room for a process to quietly self-correct. Cognizant’s plan to onboard 20,000 freshers in a single year is one requisition pipeline among many at that volume.
The Priorities That Actually Matter
The features that survive contact with real volume are the ones that reduce friction rather than add capability. Feature-count comparisons reward the wrong instinct, since the platform with the most modules is often the one recruiters quietly abandon. Knowing what an applicant tracking system actually does at its core keeps the evaluation anchored to workflow rather than module count. Dr Rajani Tewari, Chief People Officer (CHRO) and Executive Committee Member at GreenCell Mobility, narrows the field to what holds up under load.
“My three priorities would be scalability, intelligent automation and candidate experience,” explains Dr Tewari. “The ATS should manage large volumes without merely digitising recruiter workload, integrate seamlessly across the hiring ecosystem, and make the candidate journey simple and responsive. For me, the best ATS is not the one with the most features; it is the one with the least friction.”
Each of those priorities deserves its own scrutiny, because each fails in a different way at scale.
Scalability Without Recruiter Overload
Scalability means the system absorbs volume without transferring the burden onto the recruiter. A tool can technically “handle” 10,000 applications while still forcing a recruiter to manually tag, route, and status-update every one, which is scale on paper and gridlock in practice. The difference between managing volume and digitising recruiter workload separates a genuine scale tool from a data-entry trap.
This is where automation and human judgment have to be balanced rather than maximised. Bulk hiring needs automation to survive, yet full automation from day one strips out the human contact that keeps candidates engaged.
Intelligent Automation With A Human Floor
Intelligent automation earns its place when it removes repetitive work without removing the candidate’s sense that a person is involved. Resume parsing, duplicate detection, interview scheduling, and status nudges consume recruiter hours and gain most from automation. The judgment calls- who advances and who doesn’t- are where a human floor has to stay.
Ojha draws that boundary directly: “I also don’t believe in automating everything from day one. High-volume hiring needs automation, but the candidate should still feel that there is a human being behind the process.”
The warning matters more as AI-led screening spreads. A resume parser, a chatbot screener, a ranking model, and a shortlist rule can each look harmless in isolation while together deciding who reaches a human.
Peer-reviewed research on AI-enabled recruitment documents how algorithmic bias produces discriminatory hiring outcomes across gender and other traits when models learn from skewed historical data. India has no dedicated AI-hiring fairness law yet, which puts the burden of running a bias audit squarely on the employer rather than the vendor.
Candidate Experience As An Operational Metric
Candidate experience stops being a soft concern at high volume and becomes a throughput problem. When a drive processes thousands of applicants, a clunky application form or a black-hole silence after screening translates directly into drop-off, and drop-off at the top of a bulk funnel is expensive to replace.
The cost of getting this wrong compounds with every ghosted candidate who tells others. Bulk hiring runs on referral and reputation, especially in tier-two cities where a company’s name travels fast through campus networks, and candidate ghosting runs in both directions when the funnel feels impersonal. A broken experience forces a team to keep refilling the same funnel.
How To Test An ATS Before You Buy It
The evaluation method matters as much as the criteria, because a polished demo is designed to hide exactly the friction that shows up in production. Vendors control the demo environment, the data set, and the pace, none of which resemble a live bulk drive.
However, Ojha rejects the demo-only approach outright. “I personally don’t like evaluating a tool only through a product demo. Take one of your real high-volume roles and run it through the system. That tells you very quickly whether the tool will actually help the recruiters or just look good in a presentation. I also believe the people who will use the tool every day should have a say in the decision,” she explained.
Two practical tests fall out of that. A live requisition routed through the system, rather than a sandbox demo, reveals where recruiters stall. The buying decision also belongs partly to the recruiters who will live in the tool daily, because a platform that impresses a procurement committee but frustrates its daily users gets abandoned regardless of what the contract says.
The stakes are financial as well as operational, since cost-per-hire in India climbs every time a poorly chosen tool sends a team back to manual workarounds.
Multi-location hiring adds a further test that single-site evaluations miss. A company running roles across Noida, Bengaluru, and Pune needs one consolidated view, or it slides back into the fragmentation Nishi Ojha describes below.
“For me, the biggest thing is visibility. You don’t want one recruiter maintaining an Excel sheet for Noida and another tracking candidates on WhatsApp for Bangalore. I’ve seen how quickly this becomes messy when the team starts scaling, and multiple roles are running at the same time. An ATS should give the team one clear view of what is happening across locations and recruiters. More importantly, it should make the recruiter’s day easier instead of becoming another system they have to keep updating.“
The table below maps each priority to the failure it prevents and the concrete test that surfaces it.
| Priority | What It Prevents | How To Test It |
| Scalability | Recruiters buried in manual tagging and routing at volume | Run a live 1,000+ applicant role, measure recruiter clicks per candidate |
| Intelligent automation | Repetitive work eating recruiter hours; over-automation losing candidates | Check which tasks auto-complete and where a human still decides |
| Candidate experience | Top-of-funnel drop-off and reputational damage | Apply as a candidate; time the form and the post-screen response |
| Cross-location visibility | Fragmentation into parallel Excel and WhatsApp trackers | Confirm one consolidated pipeline view across cities and recruiters |
| Integration | Data silos between sourcing, assessment, and onboarding | Trace one candidate end-to-end through connected systems |
The Compliance Layer Indian Employers Can’t Skip
Candidate data carries legal weight in India that many ATS evaluations treat as an afterthought, and that gap is now a live risk. The Digital Personal Data Protection Rules, 2025 were notified on 13 November 2025 and operationalise the DPDP Act, 2023, bringing candidate data squarely into a consent, purpose-limitation, and retention framework.
An employer processing applicant data is a Data Fiduciary; every candidate is a Data Principal with enforceable rights. The Schedule to the Act sets the top penalty for a security-safeguard failure at ₹250 crore, applied per instance.
Dr Tewari frames the underlying principle in a way that reorients how HR should think about the resumes sitting in an ATS.
“Candidate data is borrowed data, not organisational property, and that distinction is important. Companies must be clear about consent, access, purpose, retention and third-party sharing, while ensuring compliance with India’s evolving data-protection framework,” Dr Tewaristates.
“As AI-led screening increases, organisations must also actively guard against algorithmic bias. Technology can enable the decision, but accountability must remain human.”
Borrowed data is the sharpest available summary of the obligation. An ATS holds resumes, contact details, assessment scores, and often recorded interviews, all of which are personal data a candidate has lent for a defined purpose rather than handed over for indefinite reuse. That reframing forces concrete questions during vendor selection: how long the platform retains rejected-candidate records, whether consent is captured at the point of application, what the tool shares with third-party assessment or background-verification integrations, and who inside the organisation can access what.
The AI dimension Dr Tewari names now has government backing. MeitY’s India AI Governance Guidelines, released on 5 November 2025, set out seven guiding principles that include Fairness and Equity and Understandable by Design.
The guidelines are advisory rather than binding, but they sit on top of the DPDP framework and signal where enforcement is heading. An ATS that ranks candidates with an opaque model and can’t explain a rejection is a governance liability waiting to surface.
In The End…
Both contributors reject feature-maximalism, and that rejection is the most useful filter an Indian HR team can carry into a vendor conversation. Dr Rajani Tewari calls the best ATS the one with the least friction; Nishi Ojha calls it the one that takes repetitive work off recruiters, gives leaders visibility, and still lets the process feel human. One conclusion, two angles.
With AI making its way into ATS as well as many other aspects of hiring, a good ATS is also one that can be relied upon to be unbiased. Especially in bulk hiring, where both quantity and quality matter, ease of operation, a lack of friction, as well as compliance are important factors to keep in mind when selecting the right tool for you.
