Seven Changes That Keep Your WhatsApp Bill Near Zero, and Three That Do Not Work

Do these two things before you optimise anything

First, attach a payment method to your WhatsApp Business Account if you have not. Without one, delivery stops the moment your number passes into billable territory, and an operation that has always been free is exactly the kind that never attached a card. Optimising a number that has stopped delivering is not optimising.

Second, write down how many phone numbers you actually run, because the free monthly allowance is granted per number and everything below depends on knowing your baseline. Most Moroccan stores discover they have one number doing ninety percent of the work and two that were connected for a campaign and forgotten.

1. Change where conversations start, not what they say

This is the largest single lever and nothing else is close. A conversation that begins with a click on a Click-to-WhatsApp ad, a post on your Facebook Page, or the WhatsApp button on your Page or Instagram profile is a free entry point. Reply inside the first 24 hours and Meta opens a 72 hour window in which every message costs nothing.

A conversation that begins with someone typing your number into WhatsApp gets no such window. The very same sales script, to the very same customer, selling the very same product, is free in one case and billed message by message in the other.

So the cheapest thing you can do to your WhatsApp bill happens in Ads Manager, not in WhatsApp. Set the destination of your Facebook and Instagram campaigns to WhatsApp rather than to a landing page or a form, and put the WhatsApp button on your Page and profile so that organic discovery also lands on a free entry point.

2. Set a reply deadline you can actually hold

The free window is not granted by the click. It is granted by your reply, and only if that reply lands inside the first 24 hours. An ad click that nobody answers until the next evening has bought you a lead and no free messaging at all.

This is where money disappears quietly in Morocco, because ad delivery does not stop at closing time. Clicks arrive at eleven at night, on Friday afternoon, during Ramadan evenings, and on the days your team is at a wedding. A store answering only during office hours forfeits the window on a large share of exactly the traffic it paid for.

There are two honest ways to hold a 24 hour deadline: staff the inbox around the clock, or automate the first reply. Most stores cannot do the first. What matters is that the first reply is real and useful, not an autoresponder saying you will be in touch, because the window opens on your reply whether or not the customer finds it helpful, and you only get one first impression.

3. Treat day three as the deadline it is

The 72 hours run from your reply, not from the sale. A conversation that is still open on day four is outside the free window, and every message from then on is billed, including the order confirmation and the tracking update.

Look at your own inbox and ask how many deals are actually decided on day four. Usually very few, and almost all of them were decided much earlier and then left hanging on a follow up nobody sent. A conversation that has gone quiet for two days rarely comes back because of a message on day five, and that message costs money.

The practical rule: ask for the decision on day one, follow up once inside the window, and either close it or let it go. A clean close costs nothing. An open conversation kept warm for a fortnight is a recurring charge.

4. Fewer templates, better written

From 30 September 2026, a utility template sent while the 24 hour window is open is billed like any other. The exemption that makes order and shipping notifications free inside a live conversation goes away, so every template you send will have a price on it.

Most Moroccan COD stores are sending three or four status templates per order out of habit: confirmed, packed, dispatched, out for delivery. At $0.0230 each that is around 0.9 MAD per order in notifications alone, which on 500 orders is 450 MAD a month for information most customers did not ask for.

Audit the list and keep the ones that reduce refused deliveries. In cash on delivery, the message that earns its price is the one sent shortly before the courier arrives, because it is the one that gets the customer to be home with cash. The other three are habit.

• Keep: the pre delivery notice, because it directly lowers your refusal rate
• Keep: the confirmation, if your confirmation rate depends on it
• Question: packed, dispatched, and every intermediate status the customer cannot act on
• Remove: any template that exists because a carrier webhook happened to fire

5. Stop broadcasting to everyone

At $0.0414 a marketing template is now the most expensive thing on the Moroccan rate card, almost twice a service message. A blast to 5,000 contacts costs around 2,030 MAD before a single person replies, and the replies themselves are billed too if those conversations are outside a free window.

The arithmetic of a broadcast changes more than people realise. While marketing templates cost 0.22 MAD each, a two percent response rate is fine. At 0.41 MAD each, with the replies billable as well, the campaign has to work considerably harder to pay for itself.

Segment before you send. A list of 800 people who bought in the last ninety days will out earn a list of 5,000 that includes everyone who ever messaged you, and it costs a sixth as much.

6. Decide what your 1,000 free messages are for

Every phone number gets 1,000 free delivered service messages per calendar month. It resets monthly, nothing carries over, and it is consumed by whichever message happens to be sent first, which is almost never the message you would have chosen.

For a store doing a few hundred conversations a month, the allowance covers everything and the rate rise is theoretical. For a store doing thousands, it covers the first day or two and then stops mattering. Work out which you are, because the advice is different.

If you are in the second group, the allowance is best thought of as a small buffer rather than a plan. If you are in the first, the practical implication is that you should stop worrying about the rate card entirely and spend that attention on conversion instead.

7. Measure delivered messages per conversation, every month

This is the one number that predicts your bill, and almost nobody tracks it. Take the messages your business delivered in a month, divide by the number of conversations, and write it down.

Under four, you are in good shape. Between four and seven, there is an easy third to remove. Above seven, something structural is wrong: usually an automation that greets in one message and answers in another, or a team habit of sending each product photo separately.

Track it monthly next to your order count. If conversations go up and messages per conversation stays flat, your costs scale linearly with your growth, which is manageable. If messages per conversation is creeping up, you have a problem that will get expensive at exactly the moment you start scaling.

Three things that look like savings and are not

Every rate rise produces workarounds that circulate in seller groups. These three come up constantly in Morocco and none of them survive contact with the arithmetic.

The honest summary: there is no clever trick here. The savings come from fewer messages and faster closes, which are the same things that make you sell better.

• Registering extra phone numbers to collect more free allowances. Each number needs its own warm up and carries its own quality rating, your conversation history gets split across numbers so your team loses context, and a set of numbers behaving identically is exactly the pattern Meta polices. You are risking delivery on your main number to save a few dirhams.
• Sending a marketing template to reopen a closed conversation. A marketing template costs $0.0414, which is more than the service message you were trying to avoid, and it does not reopen a free window. It only gives the customer a chance to reply, and their reply restarts the 24 hour service window, not the 72 hour free one.
• Going back to the WhatsApp Business app on a phone to escape per message fees. The app genuinely has no message charges, and for a seller doing twenty orders a month it is a reasonable answer. Past that it stops working: no proper multi agent inbox, no reliable automation, no integration with your store or your carrier, and one phone that has to be physically present. Most stores that go back come back within a quarter, having lost the order history in between.

Frequently Asked Questions

What is the fastest change I can make this week? Point your Facebook and Instagram ad campaigns at WhatsApp as the destination, and make sure something answers within 24 hours at any hour. Those two together decide whether most of your conversations are free or billed, and neither requires changing anything about how you sell.

Do I need to stop using templates altogether? No. Templates are how you reach a customer outside the 24 hour window, and some of them pay for themselves many times over. The pre delivery notice in cash on delivery is the clearest example: at $0.0230 it is trivially cheaper than one refused parcel. The point is to send the ones that change an outcome and drop the ones that report a status.

Is WhatsApp still worth it in Morocco at these rates? For cash on delivery, yes, and by a wide margin. A confirmed COD order in Morocco typically carries tens of dirhams of margin, and even a poorly run WhatsApp operation spends around one or two dirhams per confirmed order in Meta fees. The rate rise changed a rounding error into a line item, not into a reason to leave the channel.

Should I move to SMS instead? Compare the real numbers before you do. Moroccan SMS pricing for business sending is generally higher per message than $0.0230, it carries no free entry point window, no images, no carousels, no buttons, and far lower engagement. WhatsApp at the new rates is still the cheaper channel for the same job.

How do I know if my ad traffic is actually opening free windows? Check whether your first reply reliably lands inside 24 hours, since that is the condition. If your platform reports the window state per conversation, look at how many conversations show an open free window in the hours after an ad click. If a large share never opens one, the problem is reply speed, not the rate card.

Does any of this apply to the free WhatsApp Business app? No. These rates apply to the WhatsApp Business Platform, the API used by software. The app on a phone has no per message fees, and also none of the automation this checklist assumes.