WhatsApp Business API Pricing in India: The 2026 Breakdown

Picture a D2C founder who budgets ₹15,000 a month for WhatsApp and opens a bill closer to ₹25,000. Nothing was wrong with the math on Meta's rate card. The gap was a BSP platform fee, a per-message markup, and 18% GST stacked on top of everything. WhatsApp pricing in India isn't one number, it's a stack, and most guides only show you the bottom layer.

A WhatsApp bill in India has three layers, four if you automate

Before any specific rupee figure, get the shape right. What you actually pay each month is a stack, and missing any layer makes your estimate wrong.

• Meta's messaging charges: what WhatsApp bills for the messages you send, priced per message by category.
• BSP fees: the platform subscription and any per-message markup your Business Solution Provider adds on top of Meta's rates.
• GST: 18% applied to the messaging spend and platform fees both, because this is a service consumed in India.
• Automation software: if you run an AI agent or chatbot, that's a separate line, and it's the one that quietly decides how many paid messages you actually need to send.

Most India pricing guides cover the first layer and stop. The last three are where budgets break.

How Meta charges now: per message, not per conversation

This is the big change to internalize. Until 2024, WhatsApp billed by conversation, a 24-hour window that bundled all your messages into one of four categories. Meta moved to per-message pricing for template messages, phased in through 2025 and fully in effect for 2026. Now each template message you send is billed individually, by category.

The message categories that set your rate

• Marketing: promotions, offers, and re-engagement blasts. The most expensive category by a wide margin, and the one that quietly runs up bills.
• Utility: order confirmations, shipping updates, payment reminders. Transactional messages tied to an existing order, much cheaper, and free when sent inside an open service window.
• Authentication: OTPs and login codes, priced low, with a separate rate for international numbers.
• Service: your replies to a customer who messaged you first. Free, within the 24-hour customer service window.

What's actually free

More than most people expect, if you use it well. When a customer messages you first, a 24-hour customer service window opens, and every service reply inside it is free. Utility templates sent inside that same window are free too. And a click-to-WhatsApp ad or a tap from your Facebook Page opens a 72-hour free entry-point window.

The pattern is clear once you see it: user-initiated and transactional traffic is cheap or free, and business-initiated marketing is where the meter runs. Teams that get their WhatsApp bill under control almost always do it by shifting volume toward the free side of that line.

India rate card: rough numbers for 2026

Meta publishes India-specific rates in rupees and revises them periodically, so treat these as approximate and check Meta's published rate card before you budget to the paisa. As of early 2026, the ballpark per message looks like this:

• Marketing: roughly ₹0.78 to ₹0.88 per message.
• Utility: roughly ₹0.11 to ₹0.16 per message, and ₹0 inside an open service window.
• Authentication: roughly ₹0.11 to ₹0.13 per message.
• Service: ₹0 within the 24-hour customer service window.

The spread is the whole point. A marketing message can cost six to eight times a utility one, so the mix of what you send matters far more than the headline rate. A team obsessing over a two-paisa difference on marketing while sending everything as marketing has the wrong problem in focus.

The BSP markup nobody quotes upfront

Almost every Indian business buys WhatsApp through a Business Solution Provider, because going direct to Meta is more setup than most teams want to own. The BSP is also where costs get murky. There are two charges to look for, and vendors don't always volunteer both.

• Platform subscription: a monthly fee for the dashboard, broadcast tools, and team inbox, commonly from around ₹999 to ₹5,000 or more per month depending on tier and contact volume.
• Per-message markup: some BSPs pass Meta's rates through at cost and make their money on the subscription, while others add a markup on every message. At scale, a small per-message markup dwarfs the subscription fee.

The honest read: BSPs earn their fee for smaller teams that don't want to touch the API directly. At high message volumes, that markup is worth negotiating hard or routing around entirely. Ask which model you're on before you sign, not after your first big month.

A note for fintechs: authentication and international rates

If you're sending OTPs and login codes at volume, authentication is its own line to watch. It's priced low per message, but a fintech verifying millions of logins a month feels it fast. Two details matter here. Meta added a separate authentication-international rate for messages sent to numbers outside India, and it's higher than the domestic rate, so a user base spread across countries costs more than the India figure alone suggests. And authentication messages don't get the free-window treatment that service replies do, because they're business-initiated by nature.

For regulated teams there's a second reason to track this closely. Every OTP is a logged, billable event, so reconciling the count on your BSP invoice against your own send logs each month catches both billing errors and delivery problems early. In a fintech context, that audit trail is worth as much as the per-message saving.

Don't forget the 18% GST

Short section, high value, because this line surprises people every single time. WhatsApp messaging and BSP platform fees are services delivered in India, so 18% GST applies, and it's charged on both the messaging spend and the platform fee. On a ₹20,000 monthly WhatsApp spend, that's ₹3,600 that appears on no rate card anywhere.

If your business is GST-registered you can usually claim it as input tax credit, which softens the blow. But if you're budgeting, add it before you sign off, not after the invoice lands.

What a mid-size D2C brand actually pays

Numbers make this concrete. Picture a D2C brand on Shopify sending a realistic monthly mix, the kind of volume a growing ecommerce brand hits fast:

• 10,000 marketing messages at about ₹0.82: roughly ₹8,200.
• 6,000 inbound service conversations, answered inside the free window: ₹0.
• 4,000 utility updates, half inside the service window and half outside, at about ₹0.13: roughly ₹260.
• BSP platform fee: ₹2,999.
• Subtotal before tax: about ₹11,459.
• GST at 18%: about ₹2,063.

All-in, that's roughly ₹13,500 a month, before any automation software. Now change one variable. Double the marketing blasts and the bill jumps by more than ₹8,000, while the utility and service lines barely move. That single lever, marketing volume, explains most of the difference between two brands with identical customer counts.

Questions to ask before you sign a BSP contract

Most of the surprise in a WhatsApp bill traces back to something the buyer never asked about. Before you commit, get plain answers to these:

• Do you pass Meta's rates through at cost, or add a per-message markup? If there's a markup, what is it for each category?
• Is GST shown separately or bundled into the rate? You want it itemized so you can claim input tax credit cleanly.
• What's included in the platform fee versus metered on top? Extra team seats, additional numbers, and API call limits are common add-ons.
• How are failed and undelivered messages billed? Policies vary, and at high volume this quietly adds up.

A provider that answers these plainly is one you can budget around. A provider that gets vague is telling you something worth hearing.

Where the automation cost fits, and where it doesn't

An AI agent or chatbot is a separate line from everything above, and it's easy to conflate the two. Here's the clean split. Meta and your BSP charge for moving messages. The automation layer charges for resolving them. Robylon, for instance, uses credits-based pricing with no separate per-agent AI add-on, and it sits on top of your existing WhatsApp and BSP setup rather than replacing it.

What automation changes about the messaging bill is indirect but real. A good WhatsApp AI agent resolves 60–80% of incoming conversations autonomously, and it does most of that inside the free service window, since the customer messaged first. When you resolve more conversations in that free window, you lean less on paid marketing re-engagement to drive the same outcomes. It won't make Meta's per-message charges disappear, because no vendor can do that. What it does is shift your volume toward the cheap and free categories.

Practical levers to cut your WhatsApp bill

If your bill is already too high, these move the needle, in rough order of impact:

1. Shift marketing to utility where you honestly can. An order-update message is utility, not marketing, and costs a fraction. Don't miscategorize, Meta will reject it, but many sends teams tag as marketing are really transactional.
2. Answer inside the service window. Every reply within 24 hours of a customer's message is free. Fast response times aren't only good service, they're cheaper.
3. Use click-to-WhatsApp ads for the free 72-hour entry window instead of paying to re-engage cold contacts with marketing templates.
4. Automate the high-volume, low-value queries so your team sends fewer paid follow-ups and resolves more in the free window.
5. Negotiate the BSP markup or move to a pass-through provider once your volume justifies the switch.

Want to spend less on paid WhatsApp messages and resolve more for free? Robylon's AI agent resolves 60–80% of incoming conversations inside the service window, taking action across your order system, Shopify, and 60+ other integrations. Start free at robylon.ai