Glossary: Wando and WhatsApp Business terms

If you're just starting out, these are the terms you'll hear most often. Keep them handy as a reference.

• Template (HSM): a message pre-approved by Meta to start a conversation with a customer.
• phone_number_id: the identifier of the WhatsApp number within Meta.
• Webhook: the address where Meta sends incoming messages so Wando can receive them.
• Conversation: the thread of messages with the same contact.
• Contact: a customer's record (phone, name, email).
• AI suggestion: an automatically proposed reply that you approve or edit.
• Auto-reply: an automatic message for absence or outside business hours.

What is the difference between a template (HSM) and an auto-reply?

They sound similar but they're used at opposite moments of the conversation. The template (HSM) is used to START the contact: when you write first and the customer hasn't messaged you yet. Meta reviews it and approves it before you can use it, and the approval period is up to 48 hours. If the template isn't approved, it won't go out.

The auto-reply, on the other hand, is triggered when the customer has already messaged you and no one is available: due to schedule, absence or outside your business hours. It doesn't need Meta's approval because it goes within the conversation window the customer opened.

In practice: template for campaigns and first contact, auto-reply so no one is left hanging when the team isn't there. Mixing them up is the most common mistake for those just starting.

What does it mean that the AI 'learns' from the business's real replies?

It's not magic or a separately trained model. It's simple: when the team approves or edits a suggestion, that reply stays as a reference for how that particular business responds. With use, the suggestions sound more like your way of speaking and less like generic text.

The difference with a flow-based chatbot is the starting point. A flow has to be built and maintained: if a price or a policy changes, someone has to go in and edit it. Here the system starts from your real replies, so it adapts on its own with normal use. There's nothing to program.

Watch out for the expectation: it's not that the AI replies on its own without supervision. By default it suggests and you approve with one click. That's on purpose: in customer service, an uncontrolled mistake is expensive.

Concrete example: from inquiry to closed sale in a restaurant

A restaurant receives via WhatsApp: 'do you have a table for 4 on Saturday night?'. No one on the floor is watching the phone at 15:00. The auto-reply says they'll be attended to shortly and states the hours when they respond.

At 19:00 the team comes in. In the inbox they see the conversation and an AI suggestion that proposes: 'Yes, we have space on Saturday at 21:00. Who should I book it under?'. The waiter approves it with one click, or edits it if Saturday is full.

On Saturday morning they want to fill the tables that are left: that's where a template (HSM) approved by Meta with the promo of the day comes in, sent as a campaign to the tagged contacts. The customer replies and the conversation continues in the same inbox.

Everything stays on the same contact record, with tags, and the statistics show which campaign brought replies. None of this requires anyone to program a flow.

Common mistakes when starting (and how to avoid them)

Almost every stumble is about configuration, not the tool itself. These are the ones that come up most often:

• Thinking the AI answers on its own: by default it suggests and the team approves. If you expect full autonomy, that's not how it works.
• Sending campaigns without an approved template: first Meta has to approve the HSM, and that can take up to 48 hours. Plan with a margin.
• Not setting business hours: without that, the auto-reply doesn't know when to activate.
• Using the template to answer something the customer already asked: inside the open conversation it isn't needed, and it's confusing.
• Not tagging contacts: later you can't segment campaigns or read the stats in a meaningful way.
• Assuming Meta's approval times are shorter than they really are: the only timeline that can be stated is 'up to 48 hours'.

What can I NOT state about Meta and WhatsApp without checking the documentation?

Meta's conversation prices change by country and by conversation type. There's no fixed number that works for everyone: if you need the exact figure, go to Meta's official documentation. Here you won't find a made-up Meta price.

Same with template validation rules, webhook formats, or what the algorithm prioritizes. Meta doesn't publish everything with the same level of detail and changes often. Anything you read as a settled fact on those topics, verify it at the official source before taking it as true.

What you can be sure of: the API Wando uses is Meta's official WhatsApp API (WhatsApp Cloud API), and the only template approval timeline that can be stated is 'up to 48 hours'. Everything else, to the documentation.

How does Wando connect with the rest of my tools?

Through webhooks. Wando can notify Make, Zapier, n8n or Slack when something happens, and those tools do the rest: add a contact to Google Sheets, notify a Slack channel, trigger a flow in n8n.

For payments, the integration is with Mercado Pago. For contacts, with Google Sheets. There are no other integrations than those in the profile: if someone promises you another one, it's not from Wando.

The fine details of each connection depend on the tool on the other side, not on Wando. If something doesn't trigger, first check the webhook on the Make or Zapier side before assuming the problem is here.

Which industries have their own landing page and why does it matter?

Wando has specific pages for food service, e-commerce, retail stores, beauty, real estate, health and academies. It's not just marketing: each industry has a different use for WhatsApp.

In food service, reservations and the daily promo matter most. In e-commerce and retail stores, order tracking and cart recovery. In beauty, appointments and reminders. In real estate, inquiries about a property and pre-filtering. In health, appointments and confirmation. In academies, enrollment and class start notifications.

If your industry is on that list, the landing page saves you the work of translating the example. If it's not, it still applies: the underlying pattern is the same, AI suggestions plus human approval, and you adapt the content.