Exhibit A
What makes people reply?
- Useful replies
- Opt-out replies
Grouped · % of delivered (illustrative)
Highest bar: Useful replies in Buttons + personal detail at 14.6 % of delivered (illustrative).
Hover, focus or tap a column for exact values; press it to pin.
| Category | Useful replies | Opt-out replies |
|---|---|---|
| Announcement, no ask | 1.8 | 0.7 |
| Question in the body | 6.5 | 0.6 |
| Two quick-reply buttons | 11.2 | 0.5 |
| Buttons + personal detail | 14.6 | 0.4 |
What is the WhatsApp reply rate?
Reply rate is the share of recipients of a business-initiated WhatsApp message who send something back: typed text, a voice note, an image, or a tap on a quick-reply button, which arrives as an inbound message of type button for template buttons or interactive for reply buttons and lists. It is usually unique repliers divided by delivered recipients, measured within a fixed window after delivery such as 24 or 72 hours.
Attribution is mostly mechanical. When a customer taps one of your buttons, or uses WhatsApp's reply gesture on your message, the inbound webhook carries a context object whose id is the wamid of the message being answered — a contextual reply. Typed replies without that context are matched to the most recent campaign the person received within the window. Counting both gives the reply rate; counting only context replies gives a stricter, lower figure.
Reply rate measures action, one step beyond Read Rate. It matters operationally too: each reply starts a 24-hour customer service window, the exchange becomes Two-Way Messaging, and a bot or an agent has to answer. Under the pre-July 2025 model such a reply started a User-Initiated Conversation billed as a unit; today the window decides what you may send, and in India free-form answers inside it have been charged at the service rate (₹0.145) since 1 October 2026.
Key takeaways
- Reply rate is the share of recipients of a business-initiated message who reply, by text, voice, media or a button tap.
- Attribute replies to a campaign with context.id, the wamid of the message being answered, plus a fixed time window.
- Every reply opens a 24-hour customer service window, so replies are both a result and a promise to answer.
- Count opt-outs and complaints separately from useful replies, because a STOP is a reply too.
- Quick-reply buttons and one clear question in the body are the most reliable ways to raise replies.
The WhatsApp reply rate in code
-- Unique repliers within 48 hours of delivery, with opt-outs counted separately.WITH sends AS (SELECT wamid, campaign_id, wa_id, delivered_atFROM whatsapp_messagesWHERE direction = 'outbound' AND campaign_id IS NOT NULL AND delivered_at IS NOT NULL),replies AS (SELECT s.campaign_id, s.wa_id,BOOL_OR(i.body ILIKE 'stop%') AS opted_outFROM sends sJOIN whatsapp_inbound iON i.wa_id = s.wa_idAND i.received_at BETWEEN s.delivered_at AND s.delivered_at + INTERVAL '48 hours'AND (i.context_id = s.wamid OR i.context_id IS NULL) -- taps and swipe-replies carry context.idGROUP BY s.campaign_id, s.wa_id)SELECT s.campaign_id,COUNT(DISTINCT s.wa_id) AS delivered_recipients,COUNT(DISTINCT r.wa_id) AS repliers,ROUND(100.0 * COUNT(DISTINCT r.wa_id) / COUNT(DISTINCT s.wa_id), 1) AS reply_rate,ROUND(100.0 * COUNT(DISTINCT r.wa_id) FILTER (WHERE r.opted_out)/ COUNT(DISTINCT s.wa_id), 2) AS optout_rateFROM sends sLEFT JOIN replies r ON r.campaign_id = s.campaign_id AND r.wa_id = s.wa_idGROUP BY s.campaign_id;-- If campaigns overlap, credit context-less replies only to the latest send.
context.id gives exact attribution for taps; the time window catches typed replies.
How does the WhatsApp reply rate work?
Fix the denominator and window
Use delivered recipients as the base, and count replies within a set window — 24 hours for time-sensitive offers, 72 hours for slower decisions.
Attribute each reply
Match inbound messages to a campaign by
context.idfirst, then credit typed replies without context to the latest campaign the person received inside the window.Classify the replies
Split button taps by payload, and tag typed replies as positive, question, complaint or opt-out. Report the useful reply rate and the opt-out rate separately.
Answer inside the window
Route replies to a bot or agent immediately. A reply left unanswered for 24 hours closes the window and wastes the engagement the campaign paid for.
Iterate on the drivers
Test quick-reply buttons against a typed question, the wording of the ask and the send time, with A/B tests on a slice of the audience.
Why does the WhatsApp reply rate matter?
Reply rate is the clearest sign a message was relevant enough to act on, and replies are worth more on WhatsApp than on most channels: they open a window for free-form help, let a bot or agent move straight to an order or booking, and positive engagement supports the quality signals Meta watches. A low reply rate on a message that asks a question means the question, the timing or the audience is wrong — and a campaign that succeeds at drawing replies must be staffed for the conversations it creates.
Example: Replies a lab can actually handle
A Bengaluru diagnostics lab sends a marketing template offering a discounted annual health package to 12,000 past patients, with two quick-reply buttons: "Book a slot" and "Remind me next month". Within 48 hours 14% of recipients reply — mostly button taps, which arrive with context.id pointing at the campaign message — and 0.6% send STOP. Because each reply opens a 24-hour window, the lab staffs its booking bot and two agents for the morning after the send and books most taps directly in the chat. An earlier version without buttons, which asked people to "reply to book", drew under a third as many replies.
Common mistakes with the WhatsApp reply rate
Counting STOP as success
Opt-outs and complaints are replies too. Report them separately, or a campaign that annoys people can look like a hit.
Leaving replies unanswered
A campaign that asks a question nobody is staffed to answer wastes the window and frustrates customers. Automate or staff before the send.
Double-counting overlapping campaigns
A typed reply after two sends can be credited to both. Prefer
context.id, then credit only the latest campaign inside the window.Comparing different windows
A 72-hour reply rate is not comparable with a 6-hour one. Fix the window per campaign type and keep it.
Asking for replies you cannot use
"Reply YES" without a plan for YES creates work, not value. Make every button lead somewhere useful.
Reply Rate FAQs
What is a good WhatsApp reply rate?+
Do quick-reply button taps count as replies?+
What is the difference between reply rate and read rate?+
Do customer replies cost money on WhatsApp?+
Best practices for the WhatsApp reply rate
- End with one clear question, or offer two or three quick-reply buttons.
- Use button payloads that encode the campaign and the choice, such as
BOOK_SLOT. - Track time to first answer on replies, not just the reply rate.
- Compare reply rate by segment to find audiences that have stopped engaging.
* Per-message rates for India, ex-GST: marketing, utility and authentication effective 1 July 2026; service messages ₹0.145 since 1 October 2026 (they were ₹0 until 30 September 2026). Volume commitments earn discounts — final rate is confirmed on your account; applicable GST extra. Rates for other countries differ (see the international rate table on /pricing).