WhatsApp Business app or the API: how to decide

There are three WhatsApp products and the names do very little to tell them apart. Personal WhatsApp is the one everybody has. The WhatsApp Business app is a free download built for one person running a small business from a phone. The WhatsApp Business Platform, still widely called the API, is an interface for software, with no app of its own.

Choosing between the last two is a real decision with real costs on both sides, and the usual advice ("upgrade when you scale") is not specific enough to act on.

What the Business app actually gives you

It is free, it installs in a minute, and for a business where one person answers everything it is genuinely good software.

• A business profile with hours, address, website, and a catalog
• Labels for organising chats
• Quick replies for the answers you type constantly
• Greeting and away messages
• Broadcast lists and status posts
• Linked devices, so the same account works on a laptop as well as the phone

What it does not give you follows from its design. It is built around one account on one primary phone. Several people cannot work the same queue with any real sense of who is handling what. There is no way to connect it to your order system. There is no reporting worth the name. And automation stops at away messages.

The broadcast list trap

This deserves its own paragraph because it produces more confusion than anything else in the app.

A broadcast list in the Business app only delivers to recipients who have your number saved in their own contacts. Someone who messaged you last week but never saved you will not receive it. The list also has a hard cap on recipients, and it is small.

So the common experience is: build a broadcast list of several hundred customers, send an offer, watch almost nothing happen, conclude that WhatsApp marketing does not work. The messages went to a fraction of the list, silently. This is not a bug and it is not fixable inside the app.

What the Platform gives you

The API is not a product you use. It is an interface software talks to, which means the first thing you need is software. What you get through it:

• Several people answering the same number at once, with ownership and assignment, in a shared inbox
• Approved message templates, so you can reach people outside the 24-hour window
• Real broadcasts to anyone who opted in, whether or not they saved your number
• Connections to your other systems, so a reply can quote a real order
• Automation that goes beyond an away message
• Reporting, delivery status per message, and an audit trail
• Higher messaging limits as your number earns them

And what it costs you:

• Per-message charges from Meta for templates you send
• A software subscription on top
• Template review before you can send anything outside the window
• Business verification for higher limits, on Meta's timeline
• A setup that takes longer than a minute

The three questions that actually decide it

Most of the comparison tables miss the point. The decision usually turns on three things.

How many people need to answer at the same time? This is the most common reason to move, and it is decisive. Two people sharing one phone works badly and stops working entirely by three. If your constraint is people rather than volume, move.

Do you need to start conversations at scale? Answering people who message you is fine on the app. Reaching a few thousand customers with an order update, a restock, or an appointment reminder is what the Platform is for. If your outbound is a handful of manual messages a week, the app is enough.

Does the conversation need to know things? If answering a question means opening another system to look up an order, and doing that fifty times a day, the value of connecting the two is easy to calculate. If your answers do not depend on data held elsewhere, that value is close to zero.

If all three answers point at the app, stay on the app. Moving because a vendor says you should is how a business ends up paying a subscription to do what a free app already did.

What changes when you move

Your number becomes a platform number. A phone number registered on the Platform is generally no longer usable in the Business app in the ordinary setup. Plan for that before you migrate the number your customers already know.

Chat history does not simply come with you. Treat old conversations as reference material rather than as something the new inbox will contain.

You gain the rules. The 24-hour window, template approval, quality rating, messaging limits. These exist in the app too, in a form you rarely notice; on the Platform you meet them directly.

Somebody owns the account. A WhatsApp Business Account holds the number and the templates. Whether it sits inside your own Meta business or inside a vendor's decides what you keep if you leave.

Coexistence, and why it changes the calculation

Meta supports running one number on the Business app and on the Platform at the same time, described as coexistence. Messages sent from the app appear in the platform too, so history stays in one place, and staff who prefer the phone keep using the phone.

It suits the common middle case: a shop where the owner answers from their phone and a small team also needs to see the queue. Availability and behaviour are set by Meta and it is not offered to every account, so treat it as something to ask about rather than assume.

Who should hold the account

Ask any prospective vendor one question: will the WhatsApp Business Account be in our own Meta business, and can we see it?

Embedded Signup is Meta's flow for connecting an account you own to a platform. The customer signs in to their own Meta account, selects or creates the business, and selects or verifies the number, and it stays in their name throughout. A vendor using it is connecting to your account.

The alternative is a provider who holds the account and resells you access. That arrangement is legitimate and common, and it means the number, the approved templates, and the opt-in history live somewhere you do not control. Changing vendor usually means starting again: new number or a migration you cannot run yourself, templates resubmitted, quality rating rebuilt.

Neither answer disqualifies a vendor. An evasive answer does.

What to sort out before you migrate

If you have decided to move, most of the pain is avoidable and almost all of it is preparation rather than technical work.

Decide which number. Migrating the number your customers already have is the right answer for most businesses, and it is also the irreversible one. Starting on a fresh number keeps the app working while you set up, at the cost of retraining everyone who saved the old one.

Start business verification now. It gates the higher messaging limits and the timeline belongs to Meta. Businesses that leave it until the week of a campaign discover that nobody can accelerate it.

Write your first templates before you need them. Review takes time and rejections are common on a first attempt. Having three approved utility templates ready on day one is worth more than any feature comparison.

Export what you want to keep. Labels, saved quick replies, and the numbers of people who have been messaging you. The app can export chats; do that before the number moves rather than after.

Agree what happens to your catalog. If you sell through the app's catalog, check how the platform you are moving to handles product lists, because a catalog set up in one place does not automatically appear in the other.

Tell the people who answer. The largest cost of migration is usually a week of a team learning a new inbox while customers keep messaging. Move at a quiet time, not before a sale.

A short decision guide

Stay on the Business app if one person answers, your outbound is occasional and manual, and your answers do not depend on data in another system.

Move to the Platform if two or more people need the same queue at once, or you need to send to a list larger than the people who saved your number, or answering well requires looking something up.

Consider coexistence if the honest answer is "both": the owner wants their phone, and the team needs the queue.

And whichever way you go, the thing to get right first is not the software. It is who owns the number.