Playbooks · 6 min read

Recruiters on Instagram: Never Automate a Rejection

The PostEngage teamEngineering and support ·

You post a role. Backend engineer, three to five years, hybrid, Bengaluru. Within an hour the comments are a wall of "interested", "DM sent sir", "pls check profile", and a dozen people asking whether freshers can apply.

The volume is the reason recruiters reach for automation, and the volume is real. But an inbox of job applicants is not an inbox of shoppers, and the difference shows up in what happens when a reply is wrong. A wrong reply to somebody comparing cafes costs them a coffee. A wrong reply here lands on somebody's week, their rent, and occasionally their decision about whether to resign.

What genuinely automates

The mechanics. Where to apply properly, because a DM is not an application and your ATS exists for a reason. What the role's location and working pattern are, if you have published them. Whether the position is still open — with a caveat we will come back to. How long your process usually takes to respond, stated as your process rather than as a promise about them.

That is fixed, factual, and answerable free by a template at midnight. Get people out of the DM and into the application, politely, and you have already got most of the value available here.

Never automate a rejection

Not softened. Not "we've decided to move forward with other candidates." Not "your profile doesn't match this role."

Three reasons, and the first is enough on its own.

It is somebody's job hunt. A rejection delivered instantly by software, in a chat window, to a person who spent an evening on the application, is the single most alienating thing this industry does, and automating it means doing it at scale. People screenshot these, with your handle visible.

It is also almost always wrong at the moment it is sent: a keyword match has read a sentence, not a CV. The candidate it declines may be perfect for the role you post next Tuesday, and you have just taught them never to write to you again.

And it forecloses. A human rejection can carry something useful — a reason, a redirect, an offer to keep them on file. A template carries none of that and consumes the moment where it could have been said.

Silence is bad. An automated no is worse, because it is a decision nobody made.

Never automate an eligibility answer

The other refusal list, and it arrives constantly:

"Do you sponsor visas?" "Is a six-month gap a problem?" "Will you consider a diploma instead of a degree?" "Is there an age limit?" "Can I apply if I'm on notice period?" "Do you take candidates from a tier-3 college?"

Every one of those is a question about whether a specific person qualifies, and the answer depends on the client's requirement, the role, the market, and sometimes on a conversation the hiring manager has not had yet. Your template does not know any of it.

There is a second layer too. What may lawfully be asked of a candidate and what may be used as a criterion are governed by employment law where the role sits, and that differs by country and often by state.

Put the vocabulary in the negative keyword list: visa, sponsor, gap, eligible, eligibility, age, fresher, notice period, relocat, salary, CTC, package, reject, status, shortlist. Those messages get no automated reply and go to a person.

The leads view listing captured enquiries with the trigger that matched and when each message arrived.
Useful as a list of people who raised their hand for a specific post. Not useful as a candidate database, and it should not be allowed to become one.

The role closes faster than the post disappears

This is the failure mode unique to recruitment content, and everybody hits it once.

A job post keeps working long after the job is gone. The reel is still being served, the comments keep arriving, and your automation is still saying "yes, we're hiring for this — apply here" three weeks after the offer was accepted. You are collecting applications for a role that does not exist, from people who will wait to hear back about it.

The comment window gives you seven days to answer a comment, so the automation cannot chase a very old thread, but the post itself keeps producing new comments and new DMs indefinitely. Nothing here expires an automation — there are no scheduled sends and no end date on a rule.

So: one automation per role, scoped in Advanced to that specific post, and switching it off is a step in your close-the-role checklist alongside notifying the candidates.

The client's name is not yours to give

A lot of recruitment work is confidential — the employer is not named, sometimes because they are replacing someone who does not know yet.

Generated replies are the risk here. An AI writing a fresh reply works from what it has, and a friendly, fluent sentence that names or strongly implies the client is the kind of mistake that ends a contract. If your roles are confidential, the correct build is templates only: fixed wording you wrote, reviewed, and can point at. A credit is spent only when the AI writes a new reply, so templates-only also spends nothing.

Do not take CVs or salary in a DM

Two practical reasons. A resume in an Instagram thread is a document with a phone number, an address and an employment history sitting in a place that is not your ATS and is not covered by your data process. And current-salary and expected-salary numbers given in a chat window get quoted back later in a negotiation neither side remembers accurately.

There is a quieter one as well. Plenty of candidates are writing to you about a move nobody at their current job knows about. Keep the exchange thin, and never let a template ask them to say more in a DM than they need to.

Takeover, and the promise you actually keep

The moment a recruiter replies by hand, automation stands down in that thread permanently, so nothing generated appends itself to something you wrote.

The habit that matters more: if your template says somebody will hear back, somebody has to hear back. There are no scheduled sends here, no follow-up sequences, no reminders — the product will not chase it for you. A template promising contact you do not deliver is the automated version of what candidates already resent. Better to say what is true: applications are reviewed, you will only be contacted if there is a match, here is the link.

Official Graph API only: seven days on a comment, twenty-four hours inside a DM thread, restarted only when they write again. Templated replies are free and unlimited; free tier is 100 credits with no card, packs from ₹499. No CRM or ATS integration — leads leave as CSV.

The other side of this inbox is worth understanding too: what a job seeker can honestly automate is a much shorter list than most posts on the subject admit. Why you cannot message people first is a platform rule, not a plan limit, and what stops a send covers the checks.

One email when we publish.

No drip sequence, no “quick question” follow-up. Unsubscribe is one click and we honour it immediately.

Try it on your own posts

Free forever. Three minutes to set up.

Start free