HAMMAD YOUSUF

CHANNEL PROSPECTING // FREE LIVE DEMO

DM Analyzer Agent

A busy Instagram/WhatsApp inbox mixes genuine buyers with spam and small talk — sorting it by hand costs the first-reply speed that actually wins deals.

WHAT IS DM ANALYZER AGENT?

An AI DM analyzer agent reads a batch of inbound direct messages and sorts each one into buyer intent, spam, or general/social, flags which ones need a reply within the hour, and produces a triage table — turning a noisy social inbox into a short, ranked to-do list instead of a scroll a founder has to read message by message.

HOW IT WORKS

Reply speed is one of the strongest predictors of whether a social DM converts, and the bottleneck is almost never writing the reply — it's finding the two real inquiries buried in twenty follower-bait comments and 'nice post' messages. This agent applies a three-way classifier (buyer intent / spam / social) tuned on the pattern that a genuine inquiry almost always contains a specific, answerable question — a price, an availability check, a booking request — while spam and social messages don't.

Each classified DM gets an urgency level so the output can be acted on in priority order, and the checklist block turns the two or three that matter into concrete next actions rather than leaving them to be found again on a second read-through of the inbox.

Gemini 2.0 FlashStructured JSON outputIntent classification prompt
system-prompt.md
You are an inbox-intent analyst for a Dubai-based consumer brand's social DMs. Given a batch of inbound DMs, classify each as buyer-intent, spam, or general/social, assign an urgency level, and flag which ones need a reply within the hour. Output ONLY a JSON array of typed blocks matching the OutputBlock union — no prose outside the JSON.

USE CASES

Daily Instagram/WhatsApp inbox triage

A founder running their own DMs gets a sorted list each morning instead of scrolling past follower-bait to find the two real inquiries.

Social media manager handoff

An agency managing a client's DMs forwards only the flagged buyer-intent messages, cutting client-facing noise to what actually needs a decision.

Campaign response monitoring

During a paid social push, DM volume spikes — the agent keeps the buyer-intent signal from getting lost in the surge of comments and reactions.

Multi-channel inbox consolidation

Run Instagram, WhatsApp and Facebook DM exports through the same classifier for one consistent priority list across channels.

RESULTS & BENCHMARKS

Inbox triage time per batch

~15 min manual sort → ~20 sec

projected (modeled)

Autonomous outreach volume (IBRAHIM production)

150 personalised messages/day

real, anonymized

LIVE DEMO

Try it right now

FAQ

Common questions

How does the DM analyzer tell buyer intent from spam?

It looks for a specific, answerable question — a price, availability, or booking request — which genuine inquiries almost always contain and spam/social messages almost never do.

Does it reply to DMs automatically?

No — it triages and flags only. Sending stays a manual, reviewed step, the same honesty rule every agent on this site follows.

Which platforms does this work for?

The logic is platform-agnostic — it works on Instagram, WhatsApp Business, Facebook or TikTok DM exports since it classifies from the message text, not platform-specific metadata.

Hammad Yousuf

Built by Hammad YousufAI Marketing Automation Engineer, 540K+ YouTube subscribers.

See the production missions these agent patterns run in, or hire the whole system.