> For the complete documentation index, see [llms.txt](https://docs.dapta.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.dapta.ai/ai-text-agents/dapta-and-whatsapp.md). # Dapta and WhatsApp Business Dapta works with the WhatsApp Business number you already use. What changes for your number, and what your agent adds on top of WhatsApp. Dapta does not replace WhatsApp. It works with it. Your customers keep writing to the same WhatsApp Business number, from the same app, the same way they always have. What changes is who is on the other side: instead of a person who has to be free, awake and holding the phone, a Text Agent answers — in your words, with your prices, your policies and your catalogue. You can keep the WhatsApp Business app on the phone while Dapta is connected. Your team keeps replying from the app whenever they want to, and those replies appear in Dapta alongside the agent's. Nothing has to be deleted, and you do not give up your number. ## What you keep and what you gain | | Before Dapta | With Dapta | | -------------------------------- | ----------------------------- | ----------------------------------------------------------------------- | | **Your number** | Your WhatsApp Business number | The same number | | **What the customer sees** | WhatsApp | WhatsApp | | **Who answers** | Whoever is free | Your agent, instantly, at any hour — and a person the moment it matters | | **Where the conversation lives** | On one phone | In [Inbox](/ai-text-agents/inbox.md), where your whole team can see it | *** ## What a Text Agent adds on top of WhatsApp WhatsApp on its own can greet people, give your opening hours and answer a handful of questions you typed in advance. That is often enough to start. A Text Agent is built for what comes after the first answer — knowing your real inventory, doing something in your systems, recording the result, and involving a person at the right moment. | What you want to do | WhatsApp on its own | With Dapta | | -------------------------------------------------- | ------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- | | **Greet, give hours, answer FAQs** | Yes | Yes | | **Answer at 2 a.m., or during a launch** | Someone has to be holding the phone | The agent replies on its own, as many conversations at once as arrive | | **Answer from your catalogue, PDFs or website** | Quick replies you typed by hand | [Brains](/ai-text-agents/setting-up-your-text-agent/brains-for-text-agents.md) built from your own sources | | **Check stock, create an order, book a slot** | Not available | [Actions](/ai-text-agents/setting-up-your-text-agent/text-agent-actions.md) that call your own API | | **Record what each conversation produced** | Open the chat and read it | [Post-conversation analysis](/ai-text-agents/setting-up-your-text-agent/post-conversation-analysis.md) extracts the fields you define | | **Move the lead in your CRM** | Copy it across by hand | [CRM Pipeline Sync](/ai-text-agents/setting-up-your-text-agent/crm-pipeline-sync.md) moves the HubSpot deal or GoHighLevel opportunity on its own | | **Chase someone who went quiet** | Remember to do it | [Follow-ups](/ai-text-agents/setting-up-your-text-agent/follow-ups.md), up to three per agent | | **Message a list of contacts** | Broadcast lists, capped, only to people who saved your number | [Sequences](/ai-text-agents/campaigns.md): import, schedule, retry, fall back to a call | | **Work as a team on one number** | The app is built for one business phone | [Inbox](/ai-text-agents/inbox.md): assign, take over, release, tag, filter | | **Use the same agent on your website** | Not available | The [widget](/ai-text-agents/channels/widget-embed-your-text-agent-on-a-website.md), into the same Inbox | | **Choose the AI model and test before going live** | Not available | A model per agent, and the [Playground](/ai-text-agents/setting-up-your-text-agent/playground-testing-your-text-ai-agents.md) | | **Measure it** | Message counts | [Metrics, history and exports](/ai-text-agents/analytics-and-logs.md) | | **Tell Meta Ads which chats became customers** | Set it up yourself | The [Meta Ad Conversions](/integrations/meta-ads-conversions.md) credential lives inside the agent | {% hint style="info" %} All of this still runs on WhatsApp. Meta's rules about templates, quality and the 24-hour window apply exactly the same way — Dapta just handles them for you. See [template types](/ai-text-agents/channels/template-types.md). {% endhint %} *** ## What about WhatsApp's own AI assistant? Meta has built an AI assistant into WhatsApp Business. It is a **separate product** from Dapta: connecting your line to Dapta does not switch Meta's assistant on, and using Dapta does not require it. {% hint style="warning" %} **One number, one AI.** If Meta's assistant is enabled on a number you want Dapta to answer on, turn it off before you connect — Dapta's onboarding requires it. Your team can still reply by hand from the WhatsApp Business app exactly as before; that is a person writing, not an AI. {% endhint %} Meta's assistant needs very little setup and is right there in the app. If what you need is a greeting, your opening hours and answers to a handful of common questions, it will do that. A Dapta Text Agent is built for what comes after the first answer — and for the conversations you **start**: * Answering from your real catalogue, files and website, not only what you can type into Meta's tools. * Doing something in your systems mid-conversation, and recording the outcome afterwards. * Reaching out first: [Sequences](/ai-text-agents/campaigns.md) to a contact list, with the line protected while you send. * A team operating one number together in [Inbox](/ai-text-agents/inbox.md), and the same agent on your website. {% hint style="info" %} To be fair about it: Meta's platform *can* call external APIs and *can* hand a conversation to a person. The honest difference is how much work it takes to get a working result, and whether outbound campaigns are part of the picture at all. {% endhint %} {% content-ref url="/pages/J1S2an7vkYskmcLqMb93" %} [Your Dapta agent and Meta's Business Agent](/ai-text-agents/dapta-and-whatsapp/dapta-agent-vs-meta-business-agent.md) {% endcontent-ref %} *** ## What it costs Dapta charges credits for what the **agent** does: reading a message, working out the answer, replying. WhatsApp delivery is charged separately, by whoever provides the line. Nothing is charged twice for the same thing. The parts people expect to be expensive usually are not: incoming messages are free, an answer costs the same however long it is, failed messages are free, and testing in the Playground is not a customer conversation. {% content-ref url="/pages/4UzZiIFvS8tHwW343c2h" %} [WhatsApp billing](/ai-text-agents/channels/whatsapp-billing-and-meta-charges.md) {% endcontent-ref %} *** ## Where to go next {% content-ref url="/pages/Fos4bKY11uC0JD7tNfmx" %} [What people use Text Agents for](/ai-text-agents/use-cases.md) {% endcontent-ref %} {% content-ref url="/pages/8yBxnycfBbLwmHztlM7I" %} [Prepare your Meta account](/ai-text-agents/channels/prepare-your-meta-account.md) {% endcontent-ref %} {% content-ref url="/pages/anfdTczN1uZyNz5LkS0L" %} [Connect your Agent with WhatsApp](/ai-text-agents/channels/connect-your-agent-with-whatsapp.md) {% endcontent-ref %} --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.dapta.ai/ai-text-agents/dapta-and-whatsapp.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.