Instagram DM sales: the full process, from the first message to cash collected

What does the Instagram DM sales process look like?

Seven stages make up the Instagram DM sales funnel, from the moment someone raises a hand to the cash your closers collect. The DM work ends when a qualified lead shows up to the call. Your closers take it from there, and their verdict on every call flows back into how the DMs qualify.

Stage	Goal	Number to watch	Target
1. Trigger and outbound	Start a conversation with everyone who raises a hand, plus outbound to people who engage	Speed to lead, on every message	About 2 minutes, day and night
2. Opener	Get a reply	Opener reply rate	30%+ (most coaches come to us at 10 to 15%)
3. Conversation	Qualify, handle objections, deepen the pain, show the outcome is realistic	Booking rate: qualified booked calls per conversation	10%+
4. Booking	Book the right people while they’re interested	Booked calls tied to the conversation that booked them	Every call
5. Show-up	Get the lead onto the call	Show rate	70% to 80%
6. The call	Turn the call into cash	Cash per booked call	Rising as qualification tightens
7. Follow-ups	Bring back leads who went quiet	Ghosted conversations that come back	10 to 20%

The process is the same whether a DM setter runs it or the AI appointment setter my team runs for coaches. What changes is how many conversations it can carry at once, and at what hour.

How does a sales conversation start in your Instagram DMs?

With one of four triggers: a new follower, a comment on a lead magnet, a story reply or an inbound DM. Each one needs an answer within minutes, at any hour. On top of those, your setters start conversations of their own, with outbound to people who engaged but never wrote.

New followers. Everyone who follows you follows for a reason, and most of them will never message you first. On Sam Johnston’s account, about 80% of leads come from new followers. A personal opener goes out a few minutes after the follow, through Meta’s official API. The AI can also look at the follower’s profile first and message only the people who fit your ideal client. Manual messages are a different story: a knee rehab coach told me his account got flagged after his setter sent 50 to 75 manual messages a day to new followers, and now he can’t send more than 10.

Lead-magnet comments. Someone comments a word under your post to get your guide. ManyChat sends it: they tap a button, the guide arrives, and the conversation is open. Our AI runs everything after that. 10 or 15 minutes later it sends a relatable follow-up with a question that’s easy to answer, like “Are you coming back from an ACL injury?” or “How’s training going?”

Story replies. Plan for them. I tell coaches to post a story sequence with a clear call to action twice a week, so story replies come in on a rhythm instead of by accident.

Inbound DMs. People who write to you first, from your bio, an ad or a “DM me” line in your content. I’d still put a lead magnet in most of your content. A “DM me” call to action asks people to stop scrolling and message someone they know will sell them something. A comment for a free guide asks them to grab something they want.

Outbound. All four triggers are inbound: the lead moves first. Most setters also message people who didn’t trigger anything: story viewers, story poll voters and people who like or engage with your posts. Less often than inbound, but routinely, and these people already know your content. On our accounts the AI runs the conversation the moment someone replies. The outbound itself, to story viewers, poll voters and old hot leads, is worked by the one trained person we provide, who also runs the follow-ups in stage 7.

Then speed to lead. As I put it to one coach, “even 10 minutes, most people will forget. They have seen 700 reels in the meantime.” Firms that try to reach a lead within an hour are nearly seven times as likely to qualify it, and in the DMs an hour is already late. Our AI answers in about 2 minutes, with the delay varied on purpose. When a lead sends their whole life story over several messages, it waits until they’re done.

Where it breaks: the trigger fires and nobody answers. At Prime Train, a single lead magnet pulls 2,000 to 3,000 DM replies, more than the setters could ever get through. Today the AI answers every one of them, and Prime Train went from $250K to $350K a month.

What should the first message say?

A personal question about why they followed you, using their name. Never “thanks for the follow”.

For a knee rehab coach that might be “Are you coming back from the injury?” It’s a question they can answer in five words, it’s about them, and it doesn’t look automated.

The number here is opener reply rate: the share of first messages that get an answer. Most coaches come to us at 10 to 15%. The KPI we optimize for is 30%+. Going from 15% to 30% doubles your conversations from the same audience, without one extra follower.

You find the winner by testing. Across our accounts we’ve run more than 500 A/B tests, and the opener is still the first thing I’d test. My team runs 3 to 5 opener tests a month per account: different questions, voice notes, personalized messages and images. The winner usually gets two to three times the reply rate of the usual first message. At Kneehab Academy, the first 20 to 30 openers we tested over four to six months took the reply rate from 15% to 36%, and we’ve run more than 50 tests there by now. A winner doesn’t travel, either. The opener that works for one fitness coach won’t work for the next, because their followers are different people.

Jordan Pham shows what the opener alone is worth. Nobody had ever messaged a new follower on his account. He worried it would hurt his brand, so we started on inbound only, and the AI booked three to four calls a day. Once Jordan was ready, we switched on a personal opener for every new follower, and bookings went to about ten a day. His DM funnel went from $0 to $117,000 in 45 days.

How do you qualify a lead in the DMs?

In a normal conversation, on your own criteria: where they are, what they’ve tried, how long it’s been going on and whether they want help. Then you book the ones who fit and let the tire kickers go politely.

We learn your criteria from your past DMs and from onboarding, so the questions sound like your niche, not like a form. The script gets tested the same way the opener does: my team has run more than 200 different setting scripts side by side, and the one that books more qualified calls is the one that stays. The five-step flow one men’s fitness coach laid out for me (goal, struggle, timeline, deeper pain, do you want help) is in my guide on how to manage setters and closers.

The way a lead answers tells you as much as what they say. One men’s fitness coach’s team sorts leads like this: “Cold leads we describe as taking a long time to respond and responding with very short messages… If they’re hot, they’re sending detailed messages and they’re responding fast.”

When someone doesn’t fit, the AI thanks them and pulls back. Some filters run on their own: calls from countries Kneehab doesn’t serve get cancelled automatically.

Keep money out of the DMs. Prime Train’s account got restricted, and the cause was one qualifying question in the sales script that asked leads about their savings. Once that question came out of the DMs, the restrictions stopped. Income and decision-maker questions go in the booking form.

The number for this stage is booking rate. Everyone who replies counts as a conversation, and booking rate is the share of those conversations that turn into qualified booked calls. The KPI we optimize for is 10%+. I’ve seen anything from 3% to 35%. Jordan’s account books 29.1%, about every third conversation, once every message got an answer within minutes and every new follower got an opener.

How do you handle objections and deepen the pain in the DMs?

Answer the common objections the way you would, keep bringing the lead back to the pain they told you about, and show them that people like them got the result.

Answer them once. When we onboard a coach, my team rescans their past DMs and pulls out the most-asked questions and objections. The coach answers each one once, and the AI uses those answers from then on.

Remember what they said. If someone told you three weeks ago they want to lose weight because of their newborn baby, that’s what the AI points back to when an objection comes. A setter juggling 20 to 50 conversations at once loses that detail. The AI keeps every thread and mirrors the lead’s own language back.

Turn expert questions into a reason to book. A question only you can answer properly is a buying signal. Don’t solve it in the DMs, and don’t leave it hanging either: make it the reason for the call, where it gets a real answer.

Show the outcome is realistic. By now the lead has told you their problem in their own words. The invitation to the call is built on it: we’ve worked with a lot of people who had the same injury, and on a quick call we’ll look at your situation and map out how you get there. On Instagram your follower can find a hundred other coaches in one search, so the call has to sound like help for their exact problem. It’s also why a conversation sells better than a bare link. As I told one coach, “you are the person behind the brand instead of just being a landing page.”

How should you book the call?

Inside the conversation, at the moment the lead says yes: with a booking link that carries their name, or right in the chat.

A personal link per lead ties every booked call to the conversation that produced it, and that’s what makes stage 6 measurable at all. A link that ends with the lead’s own name likely lifts your booking rate too. Booking in the chat works too: “Let’s talk tomorrow, 3 p.m., are you free?” The AI sees the open slots in your calendar and books the time.

Book soon: every day between the booking and the call adds no-shows.

How do you get booked leads to show up?

Confirm in the DMs right away, then remind them on WhatsApp, SMS or iMessage until the call. The KPI we optimize for is a 70% to 80% show rate.

The confirmation goes out the moment the booking comes through, in the same Instagram thread: I’ve just seen your booking, excited for the call, watch this video first. Then the reminders move to the lead’s phone, with testimonials along the way.

Timing matters most. A men’s fitness coach put it plainly on a call with me: “same day is going to be a great show-up rate. The next day, show-up rate is going to drop dramatically. Two days out, they’re probably not showing up.” Volume works against you too: the more calls you book, the lower your show rate usually goes. At Kneehab’s peak, calls were booked up to a week out, and fewer people showed up to them.

I tell every client to have someone call the lead on the morning of the call to remind them. Give that call to someone who sounds good on the phone, because the wrong voice can make it worse.

What happens after the sales call?

Your closers mark every call: qualified or not, closed or not. That verdict goes back into the qualification in the DMs, because the goal is booked calls with people who pay.

Kneehab is the clearest example I have. At the peak the AI booked 1,300 calls in one month, and all five closers were booked out days in advance. More calls had stopped helping. So we added friction to the qualification before the booking link goes out, and this is what it did:

Kneehab went from $45K to $140K a month. That’s why cash per booked call is the number for this stage.

With every call tied to its conversation, you can compare the calls that paid with the ones that didn’t, side by side. As I put it on a call: “what was the conversation with this dude that paid? And what was the conversation with this dude that did not pay?” The differences tell you which question to add and who should never get the link.

That comparison is hard to do inside Instagram. An Instagram DM CRM like DM Tracker, the one we run on every client account, keeps every conversation, pipeline stage and booked call in one place. Connect it to Claude or ChatGPT and you can ask about your conversations, leads and follow-ups in plain words.

How do follow-ups work inside and after Meta’s 24-hour window?

Inside the window, the follow-ups are automated. After it closes, a person nudges, and the AI takes the conversation back the moment the lead replies.

Automated messages run inside Meta’s standard messaging window, which is 24 hours after the lead’s last message for most user actions and can be longer after a click-to-message ad. Inside it, the AI follows up with text, GIFs, images or a voice note in your own voice. At Kneehab, a lead who goes quiet can get a voice note that sounds like Matthew.

After the window, the nudge has to come from a person. The same trained person from stage 1 works a follow-up board, next to the outbound to story viewers, poll voters and old hot leads. That person never has the sales conversation. They nudge, and when the lead answers, the AI picks it up again at any hour. The whole job usually takes an hour or two a day, and one person who’s on point can work through about a thousand follow-ups a day with the board. We provide and train that person as part of the service.

The number here is how many ghosted conversations come back. On our accounts, those nudges bring back 10 to 20% of them. Most coaches skip this stage, and these are leads who already know you.

What changes when the volume spikes?

Your Instagram DM sales process stays the same. Its capacity doesn’t, and that’s where setter teams break.

On 15 December 2025, one of Matthew Maloney’s videos took off and about 6,000 people followed Kneehab Academy in a single day. Every one of them got a personal first message, more than 40 calls were booked that day, and in Matthew’s words: “There wasn’t a single DM that we missed.”

Sam Johnston gained 125,000 followers in April 2026 alone, and every one of them got an opener. In June 2026 the AI ran 20,000 conversations on his account, about 660 a day. Sam’s own estimate of doing that with people: “I’d need 10 different setters. It would be a mess to track.” He went from $27K to $120K+ a month in 90 days, with zero setters and two closers.

When a reel goes viral, the old answer is to hire five more setters, which is always a bad idea. Every new setter is another person to hire, train and manage, and they only reply while they’re on shift. If you’re deciding whether that texting should move from your team to AI, I walk through it in AI setter vs human setter.

Who should run your Instagram DM sales process?

If you’re just starting out and nothing in your DMs is proven yet, run it yourself, or hire an experienced setter. You need to find out which opener, which questions and which kind of lead work for your offer, and a person in your inbox can change course several times a day while you learn. If you want a tool to set up and run yourself, I compare that with done for you in AI setter software vs done for you. And if you personally chat in your DMs and want to keep doing that, keep doing it.

If you hire, know who you’re hiring. A lot of coaches put a cheap VA on the DMs: a chatter who does more or less what you tell them, needs far more training and never reaches the level of a real appointment setter. A professional setter deepens the pain points, creates urgency, handles objections before they come up and qualifies properly. They’re rare, because someone that good at sales psychology usually becomes a closer, where the money is better. An AI setter does the professional’s job with no limit on volume or memory, sticks to every rule and script and replies in about 2 minutes, day and night, once there’s a proven process for it to learn from.

Once your Instagram DM sales funnel is proven and more leads come in than your team can answer well, every stage above gets harder to run by hand. That’s the part my team takes off you. It’s 100% done for you and hands off. My team of 10 salespeople and AI developers runs your AI setter as your appointment setting team, and you never miss a DM or a follow-up again. Your closers stay human, and your job is content and your calls.