The WhatsApp agent that answers when the message arrives

Can AI agents use WhatsApp?

Yes, and here is what we have and have not shown. We have no WhatsApp-only result to show you yet. Our recorded results come from voice, CRM and booking work, and what carries over to WhatsApp is the tool list, the gates and the trace, not the number itself. The price follows the number of write tools and approved templates, and we quote it after an audit.

A business connects an AI agent to its WhatsApp number through the WhatsApp Business API, not the ordinary consumer app. The API is what allows automated replies, routing and a shared record that several people can see, which the ordinary app does not. The agent reads each message, calls tools such as order lookup or booking, and replies within the platform's messaging rules.

The tool list is the product

Most of the effort in a WhatsApp agent, and most of its value, sits in the tool list: the things it can actually do, and above all the tools that write to your systems. An agent that can look up an order, read live availability and create a booking is useful to somebody. The same agent with a better personality and no tools is a search box that types slowly.

Each tool is a contract: what it takes, what it returns, what it is permitted to change and what happens when it fails. Reading tools are cheap to add and low risk. Writing tools change something in the business, and every one of them deserves a limit, a confirmation and a log entry.

The refusal is the part that gets left out. An agent with no tool for a question should hand over rather than improvise. A polite guess about a refund policy is the failure that costs a customer, and it looks exactly like a correct answer until somebody checks it against the policy.

• Read: order status by reference, delivery date, opening hours, stock, price for a listed item
• Read: whether this number belongs to an existing customer, and what they last bought
• Write: create or move a booking, inside a stated range of times
• Write: raise a support ticket, with the conversation attached to it
• Write: send a payment link, for an amount the agent was handed rather than one it produced
• Refuse: anything with no tool behind it, routed to a person instead of answered from memory

How it decides, and how you find out when it decided wrong

When somebody reports that the bot said something strange, the only useful question is which tool it called and what came back. Without a record of that, the available fix is to rewrite the instructions and hope.

Every turn should leave a trace: the message received, what the agent took it to mean, which tool it called with which arguments, what the tool returned and what went back to the customer. That trace turns a vague complaint into a specific fix. It also gives you two counts that describe the agent more honestly than any containment figure, which is the share of conversations closed without a person and says nothing about whether the customer was helped: how many conversations ended with the customer repeating themselves, and how many ended with no resolution and no handover. Once every turn is logged, those counts are a reporting job rather than a guess, and they can sit on an AI business dashboard that the owner reads without opening a single chat.

A wrong answer and a wrong action are not the same size of mistake. Telling somebody the wrong opening hours is embarrassing. Moving the wrong booking is a phone call and an apology. Gates should be proportionate: read tools run freely, write tools confirm the specifics back in plain language before acting, and anything involving money either stops for a person or uses a value the agent was given.

Voice notes, screenshots and two languages in one sentence

A large share of what arrives on WhatsApp is not typed text. Customers send voice notes because talking is faster than typing, photographs of the thing they want, and screenshots of a payment they say they have made. An agent that only reads text will answer a fraction of its inbox and look broken for the rest of it.

Voice notes are transcribed before the agent sees them, which adds a second place for errors to enter. Names, addresses and reference numbers are precisely what transcription gets wrong, so the agent should read the important detail back in text and wait for confirmation before acting on it. A booking made against a misheard date costs more than a request for clarification ever will.

Payment screenshots need a rule of their own. An image showing a completed transfer is a claim rather than a verification. The agent can read the reference and the amount out of it and check them against the payment record, and where it cannot, the correct behaviour is to acknowledge receipt and pass it to a person instead of confirming an order that has not been paid for.

Language is the other half. Customers routinely mix English and Urdu inside one message, sometimes in Urdu script and sometimes typed phonetically with no fixed spelling. An agent built to expect careful English will mishandle a lot of ordinary traffic, and the reply should come back in the language the customer used. That is a comprehension requirement rather than a courtesy.

Designing inside the window and the templates

WhatsApp Business API messaging runs under rules that shape the agent more than any design preference does. Customers have to opt in. After a customer's message the business can reply freely for a limited window, and once that window closes anything the business sends has to be a template approved in advance. Getting onto the platform itself is a separate job, covered in WhatsApp Business API automation, and whether you need the API at all is covered in the app against the API.

The practical consequences are large. A follow-up sequence cannot be improvised, because the message going out well after the conversation has to exist and be approved before the conversation starts. Templates are written up front with variables where the specifics go, so a sequence is designed as a small set of shapes rather than free conversation.

The agent has to know which side of the window it is on before it composes anything at all. Inside the window it can write. Outside it, the only decisions available are which approved template to send and what to put in the variables, and an agent that does not check will produce a message the platform simply refuses to deliver.

One design move follows directly from this. Give the customer a reason to reply, because a reply reopens the window. A short question that is easy to answer at the end of a message does more for a follow-up sequence than a better-written template will.

Handover, and the shape of a good one

The handover is the part customers judge. Done badly it is a dead end where the agent stops replying and nobody arrives. Done properly the customer is told what is happening and roughly when, and is met by somebody who has already read the conversation.

The triggers belong in writing rather than in the agent's discretion, because an agent asked to judge whether it is out of its depth will usually decide that it is not.

What the person receives matters as much as when. A notification saying a customer needs help is a worse handover than a short summary of what the customer wants, what has been established and what the agent could not do, with the conversation attached underneath.

Out of hours needs a stated answer rather than silence. If a handover fires in the middle of the night, the customer should be told when somebody will be there, and the item should be at the top of the queue when they arrive. A handover that disappears into an empty office is the failure to design against.

• The customer asks for a person, in any wording, including plain frustration
• A complaint, a refund request or any dispute about money
• Anything above a stated value threshold
• Back-to-back turns where the agent did not understand
• Anything touching health, safety or somebody's legal position
• Any request with no tool behind it

When an agent should not answer for you

An agent is the wrong first move when any of these is true.

• Nobody is available to take a handover. The agent becomes a wall rather than a door, and the escalation path is a staffing commitment. Keep the agent narrow enough never to need one.
• The business cannot answer its own common questions consistently. Where a policy lives in several people's heads and they disagree, writing it down is the project and the agent comes afterwards.
• The number is a personal handset shared between a family and a business. Moving it onto the Business API changes how it behaves and somebody will mind, so have that conversation before the migration rather than during it.
• The incoming questions are mostly unique. An agent that checks with a person every second turn is worse than a list of saved replies. Improve how quickly messages reach the right human and leave the answering to them.

What is already running, what moves the number, and who holds the number itself

A first WhatsApp project would start with the audit, which produces the written tool list: every read tool, every write tool and the limit on each. The first agent is then allowed a few read tools, such as order status and opening hours, and a single write tool, such as creating a booking, which is enough to answer the routine questions and take bookings. Before customers see it, we read its traces on test conversations and check which tool it called and what came back. Further write tools are added one at a time, each with its own confirmation and its own trace.

The table below shows how our method runs in practice on voice, CRM and booking work, and it says nothing about what a WhatsApp agent will achieve. Each row comes from our Work page and keeps the figure with the case it came from. None of them is a WhatsApp-only number.

Business	What runs	Recorded result
Real estate business, USA	AI voice agents replacing a human cold calling team	$114k saved per year, set against that team. A saving, not new revenue
Real estate business, USA	Inbound and outbound AI voice agents with automatic CRM categorisation	More than 500 calls a day
Another real estate business, USA	AI calling	$19.79 cost per lead, 40% conversion lift after AI calling, 60% lower outreach cost
Eye care business, USA (New York)	AI booking across calling agents, SMS and patient messaging	82 conversations via AI booking

Across 25+ engagements in 6 countries, run from Karachi, we have shipped 120+ automations. The method behind the figures is the same one a WhatsApp agent uses: tools defined first, handover written down, every turn logged. The wider system it plugs into is described in AI for business operations.

What moves the number on a WhatsApp agent is the tool list rather than the conversation. Read tools are cheap and low risk. Write tools each need a limit, a confirmation and a log entry, and they are where the work actually goes. After that it is how many approved templates a sequence needs, whether your order and booking systems will let anything write to them, and who maintains the agent when the platform changes a rule. A supplier quoting on message volume alone has not looked at your write tools.

The number, the templates, the opt in list and the conversation history are yours and sit in accounts under your own logins. You own what we build, so you can move it to another supplier or run it yourself without asking anybody. Monthly services stop any month, with no lock-in.

Running it yourself becomes realistic the moment the tool contracts are written down, and a business with an engineer who enjoys this should take the agent in-house at that point. That is the natural end of the arrangement rather than a failure of it.

How much does a WhatsApp AI agent cost?

We do not publish a fixed price for a WhatsApp AI agent, because the cost follows the tool list: how many write tools, how many approved templates, and how open your order and booking systems are. Any figure given before that list is written down would be a guess, so we quote after the audit, and a price for an unrelated product is no guide. The audit is paid work, not a free sample, and we tell you its price before it begins.

You supply what we cannot invent: access to your order and booking systems, your policies written down, the opt-in list with the source of each opt-in, and a named person who takes handovers. Meta's own messaging charges are separate and set by the platform, so ask for them on their own line in any quote you receive.

How we bill depends on the build. Some projects are a one-time build, some are a monthly service, and some are both, so we will not tell you in advance that it is one or the other. New ideas that come up become the next phase, quoted first and never added quietly.

Common questions

What is a WhatsApp AI agent?

Software that reads incoming WhatsApp messages, works out what the customer wants, uses a defined set of tools to act on it, and replies. What an AI agent is covers the general idea in plain terms.

What is the difference between a WhatsApp AI agent and a chatbot?

A scripted chatbot follows a menu or fixed flow. An agent reads free text, chooses a tool, acts and checks the result. The agent is more flexible and carries more risk, which is why its write tools need limits. The full comparison is in AI agent against chatbot.

Can a WhatsApp agent handle voice notes and images?

Yes, provided anything important is confirmed back in text before it is acted on; the reasoning is under voice notes, screenshots and two languages.

How does the WhatsApp messaging window affect the agent?

The WhatsApp messaging window is the limited period after a customer's message during which a business can reply freely. Inside it, the agent can write free text. Once it closes, anything the business sends must be a template approved in advance, so follow-up sequences are designed and approved before they run. The agent checks which side of the window it is on before it composes anything.

When should a WhatsApp agent hand over to a person?

A WhatsApp agent should hand over whenever a written trigger fires: the customer asks for a person in any wording, a complaint, refund or dispute about money, a value above a stated threshold, back-to-back turns it did not understand, anything touching health, safety or legal position, or any request it has no tool for. The person receives a short summary plus the full conversation.

Can the agent call the customer instead of messaging?

WhatsApp supports business-initiated calling, though not everywhere: Meta excludes the United States, Canada, Egypt, Vietnam and Nigeria. Availability is a platform rule that can change, so confirm it for your own market before designing around it. A design where a missed call triggers an immediate message achieves most of the same outcome without depending on it.

Will customers know they are dealing with an agent?

They should be told, and telling them costs nothing. The outcome to avoid is a customer discovering it in the middle of a complaint and feeling handled. Disclosure also makes the handover read as a feature, because somebody who knows they were talking to software expects a person for the difficult part.