A buyer in another time zone sends a WhatsApp inquiry while your team is asleep. Someone, or something, has to answer first. Which one you pick decides whether that inquiry is still warm by morning.
Who should handle first replies on cross-border B2B inquiries: an AI SDR, a WhatsApp chatbot, or a human SDR?
Let an AI SDR own the first reply, on the channel the buyer used and at any hour, using only approved material. Let a human SDR own anything that needs judgment: pricing, custom specs, contracts, complaints, and serious deals. Use a WhatsApp chatbot only for fixed menus, routing, and "we got your message" confirmations.
The three are not rivals for the same job. A chatbot follows a script, a human SDR makes judgment calls, and an AI SDR covers the gap between them: it can hold a real qualifying conversation when no person is online, then hand off with context. The rest of this post shows where each one fits and how to decide.
Why the first reply matters on cross-border inquiries
When a buyer's workday is your night, an inquiry can sit unanswered until someone logs in. By then the buyer may have asked other suppliers the same question, and the person who picks it up has to restart the conversation from zero.
There is also a hard rule on WhatsApp. On the WhatsApp Business Platform, every message or call from a buyer opens or resets a 24-hour customer service window. While it is open, you can reply freely. Once it closes, you can only reach the buyer with pre-approved template messages (Meta: service messages and customer service windows). So an inquiry left for more than 24 hours is not just colder: your options for replying to it have narrowed.
That is why the first reply is worth assigning on purpose instead of leaving it to whoever happens to be online.
AI SDR vs WhatsApp chatbot vs human SDR: comparison table
| WhatsApp chatbot | AI SDR | Human SDR | |
|---|---|---|---|
| How it replies | Scripted menus, keyword rules, fixed answers | Free-text conversation grounded in your approved material | Free-text conversation with full judgment |
| Hours covered | Any hour | Any hour, across time zones | Their working hours |
| Handles unexpected questions | No; falls back to a menu or "please wait" | Yes, within approved material; escalates the rest | Yes |
| Qualifies the lead | Only with fixed form-style questions | Asks qualifying questions in context and scores the lead (BANT-style in PulseAgent) | Yes, with judgment |
| Channels | Usually one (WhatsApp) | WhatsApp, Telegram, and email in PulseAgent | Whatever they use |
| Pricing, contracts, custom specs | Should not touch them | Should not commit; flags or escalates to a human | Owns them |
| Handoff | Usually a tag or a queue, little context | Hands off with conversation history, lead score, and a recommended next step | Is the handoff target |
| What it writes to the CRM | Depends on your integration | Thread, score, and next step on the record | Whatever the rep logs |
| Main risk | Buyers feel stuck in a menu | Answering beyond approved material if rules are loose | Off-hours gaps and inconsistent qualification |
| Best job | Routing, confirmations, fixed FAQs | First reply, qualification, follow-up, off-hours coverage | Judgment, negotiation, closing |
The PulseAgent details in the AI SDR column come from the AI SDR for B2B export page, which describes multi-channel outreach on WhatsApp, Telegram, and email, BANT-style qualification, and escalation of complex negotiations, custom requirements, and high-value deals to your team with full conversation history.
Can a WhatsApp chatbot handle the first reply on its own?
It can confirm receipt and route the buyer. That is useful, and it is better than silence. But a B2B inquiry rarely fits a menu: the buyer asks about a spec, a destination market, or a quantity in their own words. A scripted bot either forces them into a menu or tells them to wait, which is the same wait with an extra step.
Keep a chatbot if all you need is "we received your message, a person will reply during these hours." Move the first reply to an AI SDR or a person when inquiries need follow-up questions to be understood.
What should stay with a human SDR?
Write this list down before you automate anything:
- Custom pricing, discounts, and commercial exceptions
- Contract terms and delivery commitments
- Custom specifications
- Complaints, and any buyer who asks for a person
- High-value or strategic accounts
- Compliance-sensitive messaging
The Apollo alternative or complement checklist has the same rule from the data side: write the human-only cases down before you automate outreach.
On WhatsApp, the handoff should not mean turning automation off for everyone. PulseAgent lets you pause AI on a single conversation, reply yourself from the same number, and resume AI when you're done (WhatsApp SDR human takeover).
Who owns the first reply: a handoff flowchart
Buyer inquiry arrives (WhatsApp, Telegram, email; any time zone)
|
v
[1] Is someone on your team online for this market right now?
|-- yes --> Human SDR replies first, logs the thread in the CRM
| no
v
[2] AI SDR replies on the same channel the buyer used
|-- answers only from approved material (products, specs, FAQs)
|-- asks qualifying questions: quantity, destination, timeline, decision-maker
|-- tells the buyer when a person will follow up
v
[3] Does the buyer ask about pricing, custom specs, contracts,
a complaint, or ask for a person?
|-- yes --> Stop answering, flag "needs human", escalate with context
| no
v
[4] Write thread + score + next step to the CRM record
|
v
[5] Named owner picks up from the CRM record
(on WhatsApp: before the 24-hour customer service window closes)
|
v
[6] Human continues the same thread; the buyer does not repeat anything
A chatbot, if you keep one, sits in front of step 1 only for routing and confirmation.
Which team size justifies an AI SDR?
Headcount is the wrong test. A two-person team that sells into one time zone may never need one; a larger team that sells into five might. Look for these signals instead:
- Hours covered. Map your buyers' working hours against your team's. If a meaningful part of the buyers' day has nobody on your side online, that gap is where the first reply goes missing.
- Inquiries waiting overnight. Check last month's inbound timestamps against first-reply timestamps. If inquiries regularly wait until your next morning, or past the 24-hour WhatsApp window, you have a coverage problem, not a staffing preference.
- Channels. Buyers write on WhatsApp, Telegram, and email, and each channel is watched by a different person or by nobody.
- Languages. Inquiries arrive in languages your on-shift people can't answer well.
- Repeat qualifying work. Your reps spend their first replies asking the same questions (quantity, destination, timeline) before any real selling starts.
- Lost context. The rep who picks up an inquiry has to ask the buyer to repeat what they already said.
If none of these apply, a human SDR plus an auto-reply is probably enough. If several apply, an AI SDR on the first reply is worth a trial, measured with the week-one scorecard.
Where the AI SDR sits next to your other tools
If you already run a data tool and a CRM, the AI SDR belongs in the execution layer between them: the data tool decides who to contact, the CRM stays the system of record, and the AI SDR runs the conversations and writes the outcome back. See Where Clay Ends for the layer map.
Copy this to your AI agent
Paste the block below into ChatGPT, Claude, or your own agent, together with an export of last month's inquiries. It will tell you who should own the first reply in your setup.
You are reviewing how a cross-border B2B sales team handles first replies.
Use only the data I provide; do not guess or invent numbers.
Inputs I will paste:
1. An export of inbound inquiries: channel, buyer country, time received, time of first reply, who replied
2. Our team members, time zones, working hours, and languages
3. Our current tools: auto-reply or chatbot, AI SDR (if any), CRM
4. Our list of topics that must stay human (pricing, contracts, custom specs, complaints)
Please output:
1. A coverage map: buyer working hours vs. our online hours, by market. Mark the gaps.
2. A list of inquiries that waited past our next working morning, and any that waited longer than 24 hours on WhatsApp.
3. For each gap, the recommended first-reply owner: chatbot (routing and confirmation only), AI SDR, or human SDR, with the reason.
4. The handoff rules: what the first reply may say, what must be escalated, and to whom.
5. What must be written to the CRM at handoff so the owner doesn't make the buyer repeat anything.
6. Data that is missing to answer any of the above.
If the data does not support a conclusion, say "not enough data" instead of estimating.
Common questions
Is an AI SDR just a smarter WhatsApp chatbot? No. A chatbot follows a fixed script on one channel. An AI SDR holds a qualifying conversation from your approved material across channels, decides when to escalate, and hands off with context.
Will an AI SDR replace our human SDRs? It takes the first reply and routine qualification, especially off-hours. Pricing, contracts, custom specs, complaints, and serious deals still need a person.
What happens if nobody replies within 24 hours on WhatsApp? The 24-hour customer service window closes, and you can only reach the buyer with pre-approved template messages until they write again.
Can a human take over a WhatsApp conversation from the AI? Yes. In PulseAgent you can pause AI on one conversation, reply from the same number, and resume AI afterwards.
How do we test whether an AI SDR is worth it? Run a one-week trial against a written scorecard, with the first useful result as the gate. See How to Measure the First Useful Result From an AI SDR.
Next step
If several of the signals above apply to your team, compare plans on the PulseAgent pricing page.
By 沐言AI. This post describes a decision method; it does not report results from any specific customer.