How fast should you reply to an Instagram or WhatsApp lead?
Reply speed is the time between a customer's first message and the first real human answer, measured separately for in-hours and after-hours conversations. Treated as one blended number it is a vanity metric, and a Sales QA review of Instagram, WhatsApp, and Facebook conversations shows why: the average merges two businesses, the one that runs at your team's desks and the one that runs after they go home.
How fast should you reply to an Instagram or WhatsApp lead?
Fast enough that the buyer is still in the same session, which in practice means minutes during working hours and a deliberate decision about everything else. No universal number is worth copying, because the right target depends on your category, your price point, and how many people handle DMs.
The harder half of the question is at what hour, and that is where the example report puts the deficit.
Why a single average reply time hides the problem
Because it merges two populations. In the example report on our homepage, one sample account averages 4 hours 10 minutes across all messages, which reads as mediocre but fixable. Split by schedule, the same week tells a different story.
| How the week is measured | Average DM reply time | What it actually tells you |
|---|---|---|
| Within working schedule | 42 minutes | The in-hours team is fine. |
| Outside working schedule | 11 hours 30 minutes | The entire deficit sits in a window nobody owns. |
| All messages blended | 4 hours 10 minutes | Looks like a broad team problem. It is not. |
Those figures come from an example report for a single sample account. They are not research, an industry benchmark, or a customer average.
The blended number sends a manager to the wrong meeting. It suggests retraining people who already reply in 42 minutes, while the threads that are genuinely rotting never come up. Two numbers change the problem from "the team is slow" to "nobody covers the evening."
When do after-hours leads actually arrive?
Later in the evening than most shift patterns cover, and the slowest replies cluster exactly where the volume peaks. In that same sample week, 31% of leads arrived after 20:00.
| Hour lead arrived | Relative lead volume | Reply outcome |
|---|---|---|
| 18:00 | Low | Replied in time |
| 19:00 | Low | Replied in time |
| 20:00 | High | Mixed |
| 21:00 | Peak | Slow or no reply |
| 22:00 | High | Slow or no reply |
| 23:00 | Low | Slow or no reply |
| 00:00 | Lowest | Slow or no reply |
Volume climbs from 20:00 and the outcome degrades with it. At 21:00 the account takes its heaviest hour of inbound leads and produces its worst reply outcome at the same time.
That overlap is the pattern we label "After-hours leads cool down": leads arrive after hours, wait too long for a real reply, and often do not continue. It is structural, not behavioral. No coaching fixes an hour when nobody is working.
Set a reply target instead of copying an industry number
A reply target is a written commitment, for example a first human reply within 30 minutes during working hours and by 9 AM for anything arriving overnight. It beats a benchmark because it is enforceable, and it gives every open conversation an unambiguous status: on time, or late.
- Set it per window, not per day. One number for working hours, one for outside them.
- Apply it to high-intent conversations first. A pricing question from a scored lead and a comment saying "nice" do not deserve the same clock.
- Measure the first human reply. If an automation answered, the clock is still running.
The weekly question then becomes "which conversations missed the target, and what did they share?" In the example report that was 12 late replies among 31 high-intent conversations, and 9 leads still waiting.
How do you cover after hours without hiring a night shift?
Pick one of four options on purpose, instead of letting the evening default to nobody.
- Shift the window. If 21:00 is your peak lead hour, a shift ending at 22:00 covers more buying intent than one starting at 08:00.
- Rotate a light on-call. One person checks high-intent threads twice in the evening.
- Narrow the scope. Answer only pricing, booking, and availability after hours.
- Accept the delay and own the morning. If evenings go unanswered by design, every overnight lead needs a named owner and a reply first thing.
The fourth is a legitimate choice as long as the morning half actually happens. A structured morning brief workflow turns "we will get to it tomorrow" into one person clearing a specific list, and it depends on clean lead handoff, because a thread three people can see is a thread nobody owns.
An auto-reply is not a reply
An auto-reply buys you an acknowledgment and nothing more. It confirms the message arrived, it does not answer the question, and it should never stop the clock on your reply target.
Two conversations from the example report make the point. Nolan Price wrote on WhatsApp that he works nights and wanted a consultation right after the studio opens. He got an auto-reply saying the studio was closed until 7 AM, messaged again on his way home, and got a real answer the next morning, an eight hour wait. Zoe Harris messaged on Instagram, ready to start strength coaching before a work trip. She got a generic auto-reply, then chased with "Checking once more in case there is an opening this week" before a human offered her a slot 4 hours 45 minutes later.
Both threads reached a real answer only because the customer chased, not because the process worked. The buyers who do not chase are the ones you never hear from again. That is its own pattern: "Auto-reply only, no human follow-up."
If you run one, make it specific. "We reply by 9 AM" can be measured. "A team member will reply during studio hours" is a placeholder that lets a warm lead age overnight.
How Reploom measures reply speed
Reploom is AI Sales QA for social conversations. It reviews Instagram, WhatsApp, and Facebook messages, comments, and mentions after they happen, finds the leads a team lost, and delivers a weekly Sales QA report with the original conversations attached as evidence. It is not a chatbot and not an inbox: it does not send, draft, or suggest replies.
For reply speed, that means every conversation gets a 0 to 100 lead score, a buying-intent type, and a one-line reason, so the clock applies to real leads rather than every message. Unanswered leads and overdue follow-ups surface in a live queue measured against the reply target you set, and repeated leaks land in the Monday report grouped into patterns with their source conversations.
For the wider definition, start with what Sales QA for social conversations is. Every number here comes from the example report on the homepage. Reploom, Sales QA for social conversations, measures every conversation against your reply target and reports the gap weekly. To see your own in-hours and after-hours split, book a demo.