Screen Naukri Applicants Automatically: 800 Applies to 20 Interviews
How to screen Naukri applicants automatically: what the Response Manager filters catch, why resume scores fail volume roles, and a knockout-then-call workflow.
To screen Naukri applicants automatically, check job requirements first, confirm consent for the screening call, then run structured interviews and review the evidence behind the shortlist. You can use Naukri's screening products or move applicants into another system through a supported integration, an approved export, or your own apply link.
Picture a telecaller role posted on Monday. By Thursday there are 800 applications and the hiring manager wants 20 interviews. Sorting the inbox helps, but someone still needs to establish who can do the work. The workflow below uses those numbers as a worked example, not a customer result or a promise to finish in one day.
Naukri already offers more than resume filters. Its AI REX guide describes automated outreach, including calls and WhatsApp, screening questions and interview scheduling. Evaluate that option alongside a separate screening pipeline. The distinction that matters is what evidence each system collects and whether it fits your hiring process.
What the Naukri response inbox actually gives you
Start with the tools already in your account. Naukri documents profile matching, questionnaires and response-management actions; the exact options depend on your subscription. These can narrow an inbox before a recruiter opens individual profiles.
| What Naukri gives you | What it answers | What it cannot answer |
|---|---|---|
| Relevance Score out of 10 | Match across skills, designation, experience, salary, location and education | How the candidate performs a live task |
| Profile filters | Which applications match selected fields | Current facts the profile does not capture |
| Questionnaires attached to the job, usable as filters | A yes/no or multiple-choice answer typed at apply | Whether it is true, or what they would say if you followed up |
| Sort by freshness or relevance | Which profiles to read first | Who should be rejected without further evidence |
| Folders (shortlist, on hold, reject), comments, forward to a hiring manager | Where each apply is in your process | Nothing about the candidate; this is bookkeeping |
Sources: Naukri's Relevance Score explanation and Response Manager overview. These describe inbox features; AI REX is a separate part of the evaluation.
A profile match is useful evidence about stated experience. For a telecaller, delivery partner or field sales executive, you also need evidence of how the person handles the work. A questionnaire adds answers; a conversation lets you follow them up.
Where resume scores fail for volume roles
Three failure modes, and one piece of arithmetic.
Profile detail can become an accidental gate. A two-line profile typed on a phone gives a matching system less to work with than a carefully written CV. That does not establish a difference in selling ability. This is a reason to check low-match applicants against the actual job requirements, not evidence that every high-scoring profile is misleading.
Profile data can be out of date. A stated notice period, location or salary expectation needs confirmation for this vacancy. Ask about the actual shift, worksite and joining window before spending time on a longer interview.
The written profile can test the wrong language skill. If the work is customer support in Kannada and Hindi, judge a task in those languages. English CV polish should not become an extra requirement unless written English is part of the job.
And even when the score is right, it does not solve the volume. A filter that keeps a quarter of 800 applies leaves 200 profiles, and 200 profiles at three minutes each is ten hours of reading that ends with a list of people you have still not spoken to. That is the bottleneck in the bulk hiring process, and sorting does not move it.
Screen Naukri applicants automatically: the four-step workflow
Illustrative counts, not customer results or a completion deadline. The 440 exclusions need review; the 90 incomplete calls need a next step. Consent is checked before scheduling.
The counts are a worked example, not a benchmark. Completion, eligibility and shortlist rates will differ by role, city and calling window.
Step 1: Get the list out of the inbox
Two situations. Applicants who complete your own apply form are already in your pipeline. Where your Naukri posting supports an external apply destination, use that route; adding a link to the description alone does not redirect everyone who clicks Naukri's Apply button.
For existing responses, confirm the supported export or ATS handoff with your account manager. Naukri's recruiter FAQ documents folder and download actions, but that does not establish the spreadsheet columns, quotas or access charges on your plan. Test a small export first. Map its fields to the import template, retain the source and consent record, and check duplicates by phone number before importing.
Step 2: Apply the hard knockouts
A knockout is a deterministic rule about the job, not a judgment about the person. Write the set down before you look at a single profile, and apply it to every row the same way.
| Knockout | Ask it as | Declines when | Keep the override for |
|---|---|---|---|
| Location and commute | "Which area do you live in, and how long is the daily commute to the site?" | Outside the radius you set (say 60 minutes door to door) | A candidate relocating with a confirmed date |
| Notice period | "What is the earliest date you could start, allowing for your notice period?" | Later than the joining window | A rare skill, or a role where you will buy out notice |
| Shift | "This is a rotational shift with night weeks and weekend work. Can you work that schedule?" | A confirmed no; clarify uncertainty first | Roles with a genuine day-shift variant |
| Salary band | "The fixed salary range is X to Y, with the variable component explained separately. Does that fit your expectation?" | Expected fixed above the band ceiling | A strong profile within a margin the hiring manager has pre-approved |
| Language | "Which languages can you handle a customer call in?" | None of the languages the role needs | Roles where the language can be trained in the first month |
| Role-specific requirements | The licence, equipment or qualification the work actually needs | A verified essential requirement is missing | Equivalent qualifications or employer-provided equipment, where acceptable |
Review the exceptions before closing an application. A missing or ambiguous answer needs clarification; it is not a confirmed failure. Keep an override with a recorded reason, and tell candidates when a decision has been made. If you imported contact details without screening answers, collect those answers first; a CSV row cannot establish eligibility by itself.
In the worked example, five checks exclude 440 of 800: location 180, notice 90, shift 70, salary band 60, language 40. These are sequential counts, with each person counted once. That leaves 360 eligible. The team reviews the exclusions and handles outcome communication separately.
Step 3: Talk to everyone who clears them, in parallel
Use a twelve-minute structured call with the same core questions and scoring anchors, in the language the work requires. Ask for the candidate's last target and its unit, give a customer objection to handle, or ask what they did when the cash did not tally. Follow up on vague answers. The phone-screening question guide shows how to turn a general question into an answer you can check.
Calling all 360 eligible candidates for twelve minutes would take 72 hours of conversation. Concurrent calls reduce elapsed time, but candidate availability, retries, capacity limits and the calling window still matter. In this example, three of four complete the screen: 270 completed calls, or 54 hours of conversation. The remaining 90 need a retry, a reschedule or a recorded withdrawal; they are not failed interviews.
Step 4: Interview the ranked top
Start with the highest-ranked reports, listen to the evidence behind the ratings, and review borderline cases before selecting 20 interviews. Decide what happens to the other 250 completed screens and the 90 incomplete ones. A ranked list reduces review work; it does not remove the need to check scoring errors or keep candidates informed.
Which applications deserve a call at all
For this workflow, use verified job requirements to decide eligibility. A relevance score can help order the queue, but a low match alone is not evidence that someone cannot do a frontline role.
| The application | Call? | Why |
|---|---|---|
| Passes every knockout, High Match | Yes | The easy case. |
| Passes every knockout, Low Match | Yes | A sparse profile can still belong to someone who meets the job requirements. |
| Fails a hard knockout: cannot reach the site or work the shift | No, once confirmed | Clarify relocation or shift alternatives before closing the application. |
| Fails a soft knockout by a little: 45 days' notice against a 30-day window | Hold | Include if the eligible pool runs short; a human clicks the override, not the rule. |
| Salary expectation above the band | No, unless pre-approved | Get the hiring manager's margin in writing before the screen starts, then apply it to everyone. |
| Skipped an optional questionnaire, or gave an unreadable free-text answer | Clarify | Missing information is not a failed requirement; collect what the decision needs. |
| A Resdex profile that never applied | Not yet | Establish interest and record agreement to this screening call before scheduling it. |
For every "yes", confirm consent and explain the AI interview and recording before scheduling. For every "no", keep the requirement and evidence that supported the decision.
The export and import mechanics, honestly
There are three ways applicants get out of Naukri and into a pipeline that can call them. Which one you use depends on what you already run.
Where the posting supports it, route applicants to your form. People who complete that form enter your pipeline directly.
- Works when
- Your posting supports an external application destination or a permitted link to your form.
- Watch for
- A link in the description does not move existing applications or guarantee that every applicant uses your form.
- Consent
- Explain the screening call, AI interviewer and recording when collecting agreement on the form.
Use a supported export, map the available fields to your screening system's import template, and test a small batch.
- Works when
- The applies are already sitting in the inbox, and you do not run an ATS.
- Watch for
- Column names, download limits and whether a download counts as a CV access differ by plan. Confirm with your Naukri account manager rather than with a blog post, including this one.
- Consent
- Importing data creates no new consent. Preserve existing evidence and collect any missing agreement before scheduling an AI call.
Use a supported Naukri integration with your applicant tracking system. Greenhouse documents one such integration.
- Works when
- You already use a supported ATS and have the required Naukri access.
- Watch for
- Greenhouse's help page says a paid Naukri subscription and an additional integration module are required. Confirm pricing and field mappings with Naukri.
- Then
- Verify how applicants move from the ATS to your screening system. A Naukri-to-ATS connection does not automatically connect a third product.
The Greenhouse integration guide documents the ATS route and its prerequisites. Export columns, caps, access charges and current commercial terms still need confirmation for your account. An undocumented scraper is not a substitute for that handoff.
What 800 applies cost to clear by hand
Whatever you choose, someone pays for the hours. Here is the worked example with times attached: three minutes to read a profile, twelve to screen a candidate by phone, ten to review a ranked report.
Illustrative task arithmetic, not customer data or equal-coverage alternatives. The 72 hours assumes all 360 complete a call; the example's 270 completions take 54 hours. Reviewing 20 reports excludes setup, retries, exception review, outcome messages and final interviews. AI calls incur calling costs.
These tasks produce different evidence, so the bars are not a measured savings claim. Put your team's completion rate, retry effort and review time into the cost-per-hire calculator before setting a deadline.
How EasyInterview clears a Naukri inbox
EasyInterview provides an apply form, AI phone screens and ranked reports for teams handling applicants from several channels.
Share the branded apply link through supported job-board placements, your careers page or a QR poster. Existing applicants can enter through CSV import. EasyInterview has no live Naukri API connection; the apply link and CSV import are the available routes, and native ATS integrations remain on the roadmap.
On the apply form, configure requirements such as location, shift, joining date and salary band. Review exclusions and use the "include anyway" override where appropriate. Imported contact rows do not supply those form answers or automatically run the same knockout checks, so verify eligibility before scheduling them. For imports, the recruiter records an attestation that candidates agreed to contact about this role, including a recorded AI phone interview; the import does not send a consent request for you.
Your team selects eligible candidates with consent on file and schedules screens inside the workspace calling window. The phone interview supports eleven languages: English, Hindi, Bengali, Tamil, Telugu, Kannada, Malayalam, Marathi, Gujarati, Punjabi and Odia. Set the question bank for the role, use follow-ups to clarify answers, and review the recording and transcript alongside the scored report. Concurrent calls reduce the need for a recruiter to conduct every first conversation, subject to capacity and candidate availability.
The current workflow uses the apply page and phone call, with optional outcome emails when your team hands off decisions for applicants with an email address. WhatsApp slot-picking, automated consent messages and one-tap messaging reschedules are not part of this version. Your team handles any additional candidate communication.
Evaluate the workflow with a pilot on one live vacancy alongside your existing screen. Compare completion, evidence quality, recruiter time and interview outcomes; the 800-to-20 funnel above is not a published customer result. Use the comparison page to orient your evaluation, then ask for a demo of the current phone-screen workflow.
Frequently asked questions
Does Naukri have an ATS integration?
Yes. Greenhouse documents a Naukri integration requiring a paid Naukri subscription and an additional integration module. Confirm your ATS, field mappings and commercial terms with Naukri. This does not mean EasyInterview has a direct Naukri integration.
Can I export Naukri applicants to Excel?
Naukri documents download actions, but confirm whether your account exports the applicant fields you need in a spreadsheet. Resdex database downloads and responses to a posted job are different contexts. Test the format, columns, limits and any CV-access charges before relying on a daily export.
Should I use the Relevance Score to reject candidates?
Do not use it as the only rejection reason. It measures profile-to-job match; it does not replace confirmation of job requirements or a role-specific task. Review the evidence and allow corrections to missing or outdated information.
How many Naukri applicants should I actually call?
Plan a consistent screen for eligible candidates who agree to it. If capacity is limited, use an explicit, job-relevant prioritisation rule and track who remains unscreened. Concurrent calls can expand coverage, but non-completion needs a follow-up policy of its own.
Does importing a CSV collect consent for an AI phone screen?
No. Importing records does not create agreement to a recorded AI interview. Check what each candidate was told and agreed to, retain that evidence, and collect missing agreement before scheduling. In EasyInterview, the recruiter attests to existing consent on import; there is no automatic outreach flow to collect it.
Clear the inbox by talking to it
An inbox becomes useful when each application has a clear next step. Confirm the job requirements, establish consent, run a consistent screen, and inspect the evidence behind the shortlist. Keep incomplete calls and exceptions visible. That process works whether the screening happens through Naukri's own products or a separate pipeline.
Book a demo to see EasyInterview take one Naukri req from the apply link to a ranked, evidence-scored shortlist.
Rohit Pandit
Founder, EasyInterview · easyinterview.life
I build EasyInterview, an AI hiring pipeline for high-volume roles in India. These are field notes on screening at volume, hiring signal, and interview integrity, written from the work rather than from a keyword list.
Connect on LinkedIn
