WhatsApp Broadcast List: What It Can and Cannot Do

A broadcast list is a saved set of contacts inside the WhatsApp Business app. You write one message, choose the list, and WhatsApp delivers a separate copy to each person on it. Recipients do not see each other and do not know the message went to anyone else. If they answer, the answer arrives as a normal one-to-one chat in your inbox rather than in a shared thread.

Three things get called the same name, and they behave nothing alike:

• A group is one shared conversation. Everyone sees everyone, and every reply goes to all of them. Wrong tool for a shop, for obvious reasons.
• A broadcast list is many private conversations sent in one action. Nobody sees anybody. This is the thing most merchants mean.
• A one-to-many send on the WhatsApp Business Platform looks similar from the outside and works on completely different foundations — stored consent, segments and approved templates. That is the subject of how to send a WhatsApp broadcast safely.

The broadcast list is the oldest of the three and the only one that needs no setup at all. That is exactly why it is where most shops start, and why the disappointment tends to arrive at around the fiftieth customer rather than the fifth.

Because of one rule that sounds like a detail and is not. WhatsApp's own documentation states that a broadcast is delivered to the people on the list as long as they have your number saved in their phone's address book (WhatsApp Help Center, retrieved 2026-09-01). Saving works in one direction only, and it is not your direction. You having their number does nothing.

Now put that against how a Shopify store actually acquires phone numbers. A customer types their number into a checkout field so a parcel can reach them. Nothing in that moment causes them to open their contacts app and add your shop. Some will have messaged you first, about a return or a size, and those people usually do have you saved. The rest — the majority, and the ones a campaign is aimed at — do not.

The failure is silent, which is the expensive part. A number that is not saved produces no bounce, no warning and no separate line in any report. The send completes. From the sender's side a campaign that reached forty people out of eight hundred looks the same as one that reached everybody, and the conclusion a merchant draws — "WhatsApp does not work for us" — is drawn from a measurement that was never taken.

There is a second-order effect worth naming. The people who did save your number are your most engaged customers. A broadcast list therefore keeps talking to the segment that needed the least persuading, and keeps proving that WhatsApp performs brilliantly — on a sample selected for already liking you.

Rule of thumb: if you cannot say how many people on the list have you saved, you cannot say what your reach was. And you cannot say it — the app does not expose it.

Earlier than the delivery problem alone would suggest, because four other things are missing at the same time. None of them is a bug. A broadcast list was designed for a person messaging people they know, and it does that job well.

• No consent record. Membership of a list is not evidence that anyone agreed to marketing. Meta's policy requires an opt-in that names the business collecting it before you message someone (Meta for Developers, retrieved 2026-09-01), and a phone list stores nothing you could show.
• No segmentation. You can hand-pick names, and that is the entire mechanism. There is no way to ask for customers who bought a category twice in ninety days, because the app has never seen your orders.
• No shared inbox. Every reply is a separate private chat on one phone. Two hundred recipients is two hundred threads, and only whoever holds the device can answer them.
• No reporting. Ticks on individual chats are not a campaign result. There is no delivery rate, no read rate per send and no way to compare one message against the next.
• No opt-out you can honour. Someone who asks to stop hearing from you has to be remembered manually, on that phone, forever, by whoever is holding it.

There is also a plain operational risk underneath all of it: the list lives on one handset. A lost phone, a new device or a colleague leaving takes the list with it, and nothing about it was ever backed up as customer data. Compare that with a contact base that sits alongside your Shopify data and is visible to the whole team — which is what the platform side of Keaz is built around.

It replaces the address book with a record. That is the whole difference, and every practical consequence follows from it.

Delivery does not depend on being saved. A message sent through the Platform reaches a customer who has never added your shop to their contacts. What it depends on instead is consent and message type: outside an open customer service window, an approved template is the only thing that goes out at all. The rules that govern this — templates, categories, messaging limits — are collected on our WhatsApp facts page, with sources and retrieval dates.

Consent is a field, not an assumption. In Keaz every contact carries an opt-in level, so a marketing send can filter on it and a service message does not have to. When someone opts out, they stop being eligible everywhere at once rather than in one person's memory.

The audience is a query, not a hand-picked list. Segments are built from Shopify order data and channel behaviour — last-ninety-day buyers, one product category, people who never opened the last three sends. The mechanics are one screen; see Segment Builder in the help center. A segment also updates itself, which a phone list has never done.

Every send returns numbers. Delivered, read, replied, per segment, per send — which is what lets you find out that the narrow audience beat the broad one instead of assuming it. The send panel shows the segment size before you commit; Newsletter: Sending walks through it.

The honest trade is that the Platform asks for setup the app does not: a business account, approved templates, and an opt-in you actually collected. In exchange the reach stops being a function of who bothered to save you.

Five steps. The order matters more than the speed, and the second one is the step most shops try to skip.

1. Write down what the list was for. Restocks, launches, a VIP group, an apology when a delivery slips. Most phone lists are two or three real audiences that were merged because the app offered only one place to put them.
2. Re-ask for consent instead of importing the list. Numbers you have are not permissions you have. Ask once, in a place where saying yes is easy — checkout, an order confirmation, a QR code in the parcel — and store the answer. A smaller list you can prove beats a larger one you cannot.
3. Build one segment with a reason. Not "everyone who opted in". One audience, one thing you want to tell them, one reason it is relevant to them specifically.
4. Send one message and read the result. Delivery and read rates tell you the mechanics worked. Reply rate and revenue tell you the reason was real. Setting up a WhatsApp newsletter covers the first send end to end.
5. Retire the phone list. Running both means two consent states, two opt-out mechanisms and no idea which send caused which result. Keep the app for the handful of regulars who genuinely chat with you, and stop using it as a campaign tool.

The move is usually less work than it sounds, because the part that felt like an asset — the list itself — is the part you deliberately leave behind.

Six that come up again and again, in roughly the order they get made.

• Assuming the message arrived. Delivery depends on the recipient having saved you, and the app never tells you who did.
• Asking customers to save your number as the fix. It works for the few who read the request, which selects for engaged customers again and hides the problem for another quarter.
• Treating the list as an opt-in. A number collected for delivery is not permission to market. Carrying that assumption into a platform migration imports the problem rather than solving it.
• Sending to everyone because the tool offers nothing else. Unsegmented sends run at 15–25% engagement against roughly 30% for segmented ones (published benchmark band — how we calculate), and the difference is collected in opt-outs.
• Keeping the list on one person's phone. It is customer data with no backup, no access control and no continuity when that person is away.
• Concluding that WhatsApp does not work. The channel was never the variable. A tool built for messaging friends was being used to run campaigns, and it did exactly what it was designed to do.