WhatsApp is the operating system for Indian business communication. The numbers are not in dispute — by mid-2026 most B2C operators in India route more inbound through WhatsApp than through email, phone and chat combined. What is in dispute is what "WhatsApp business automation" actually means.
Search the term and you will find lists. Ten BSPs. Twelve automation tools. The same four names — Wati, AiSensy, Gallabox, Interakt — ranked in a slightly different order on each page. None of those lists answer the actual question an operator is asking, which is: what should the system do, and where does the leverage sit?
This post is a field guide written from the WhatsApp work we have done for Indian operators in the INR 5 crore to INR 100 crore band. The short version is that the BSP is the rail, not the automation, and the real leverage sits in a thin custom layer that almost nobody is selling.
The three layers of a real WhatsApp setup
Strip away the marketing copy and a serious WhatsApp operation has three layers, in this order.
Layer one — the rail. This is the Meta WhatsApp Business Platform itself, accessed through the Cloud API or a Business Solution Provider. It handles the actual sending and receiving of messages, template approvals, the display name and verification badge, the quality rating that governs your messaging tier, and the per-message billing. This layer is commoditised. Wati, AiSensy, Gallabox, Interakt, Gupshup, Pickyassist — they all wrap the same Meta API. The differences are in pricing, the team-inbox UI, and the depth of the chatbot builder. None of them are differentiating on the rail itself.
Layer two — the conversation surface. This is the shared inbox, the chatbot, the broadcast tool, the template manager. Most BSPs ship a competent version of this layer. It is what you see when you log into Wati or AiSensy. It is genuinely useful for an operator who needs to start sending and receiving structured WhatsApp messages tomorrow. The ceiling on this layer is reached when your team starts copy-pasting information out of the WhatsApp inbox into another system — your CRM, your dispatch board, your clinic software, your Tally voucher screen — and copy-pasting structured replies back in.
Layer three — the system-of-record bridge. This is where almost every Indian SMB WhatsApp setup is thin or absent. It is the layer that turns an inbound WhatsApp message into a structured ticket, lead, order, appointment or voucher in the business's real system of record, and then triggers the right outbound template when that record changes state. It does not show up in BSP feature lists because BSPs cannot ship it generically — it is specific to your CRM, your dispatch software, your clinic management system, your Tally instance. This is where custom work earns its keep.
A useful diagnostic: if your sales head, your clinic manager or your operations lead is logging into the WhatsApp inbox more than twice a day to "see what is going on", you do not have layer three. You have layer two with humans bridging the gap.
The five workflows where automation pays back fastest
Across the Indian operators we have built for, five workflows produce most of the WhatsApp automation return. The order matters — start with the one that touches the most revenue.
Appointment and service reminders. Clinics, salons, HVAC operators, diagnostic chains, automotive service centres. A no-show or a missed AMC renewal is a direct revenue loss, and a utility template sent 24 hours before the appointment with a confirm-or-reschedule reply reduces no-show rates measurably. This is also the cheapest category to send under Meta's 2026 pricing.
Order confirmation, dispatch and delivery updates. Direct-to-consumer brands, F&B operators with delivery, regional logistics. The template flow is almost always cheaper than a transactional email at India volumes, and the open rate is materially higher. Tie the templates to events in your order management system rather than firing them on a schedule.
Collections and payment follow-ups for B2B services. CA firms, law firms, agencies, service businesses where invoices age. A utility template referencing a specific invoice with a payment link and a reply path to flag a dispute is one of the highest-leverage uses of WhatsApp Business we have seen, and one of the most under-built. Read our process automation between Tally, Zoho, Stripe and GST playbook for how the bridge to the accounting system should look.
KYC and document collection. CA firms during tax season, law firms on new matters, NBFCs and lending businesses, clinics collecting prescriptions or imaging. The WhatsApp document flow with structured intake — name the document, attach the file, confirm receipt — is faster than email and dramatically faster than chasing the client over a phone call. The hard part is the bridge: the document needs to land in the right matter folder or patient record, not in the operations team's downloads folder.
Lead capture from paid ads. Real estate developers, broker firms, education businesses, high-ticket retail. A Click-to-WhatsApp ad pushes a warm prospect into a conversation that needs to be captured as a structured lead with source attribution within seconds, routed to the right sales person, and slotted into the CRM with SLA timers. Off-the-shelf BSP click-to-WhatsApp flows handle the capture but not the routing and the attribution.
What the BSPs do well — and what they do not
A fair read on the current Indian BSP market, as of mid-2026.
Wati is reliable, has the cleanest team inbox of the bunch, and is the most flexible on per-message volume. The pay-as-you-go plan is genuinely cheap for low-volume operators. The chatbot builder is competent but not deep. The integration story beyond Shopify and Zapier is thin.
AiSensy is the most aggressive on price for mid-volume operators, ships native INR billing, and has the broadest set of pre-built integrations including HubSpot, Razorpay and Shopify. The campaign tools are strong; the team inbox is workmanlike. Activation is fast.
Gallabox leans into broadcasts, click-to-WhatsApp ads, drip sequences and an AI-assisted chatbot. The 2025 per-message pricing move made the maths cleaner for high-volume senders. The integration depth into Indian SoR systems remains shallow.
Interakt publishes per-conversation pricing transparently and has improved on chatbot flows in the last two release cycles. Branched flows only appear on the Advanced plan, and the single-channel limit on Starter catches some operators by surprise.
The honest framing across all four: they are competent rail-plus-conversation-surface providers. Pick the one whose pricing and inbox UI fit your team, do not spend more than two weeks on the selection, and move on. The differentiating work happens in layer three.
What a thin custom layer over a BSP actually does
The phase one we typically scope at Tanvora for an Indian SMB operator already on a BSP looks like this.
A webhook handler that consumes inbound message events from the BSP, parses them for the top three or four intents the business actually sees, and writes a structured record into the CRM or system of record — with source attribution, customer match against existing records, SLA timer started, and the conversation thread linked back into the SoR.
A routing engine that assigns each conversation to the right team or human based on intent, customer segment and load — and reassigns when the SLA timer breaches. This sits in front of the BSP inbox, not inside it.
A consent and DPDP audit layer that captures opt-in evidence at source, records the legal basis for each outbound template, holds an opt-out registry that the outbound flow checks before every send, and produces an audit trail a Data Protection Officer can defend. Meta's WhatsApp Business Platform docs describe the technical mechanics; the DPDP Act 2023 describes the legal obligation. Most BSPs cover neither in production-grade form.
A templated outbound engine that fires utility messages based on events in the SoR — an appointment booked, an invoice raised, a delivery dispatched, a renewal due — rather than on a manual broadcast trigger. This is where the cost optimisation actually lands, because event-triggered utilities tend to keep you inside open service windows and out of marketing-tier billing.
A small set of structured intake flows — appointment booking, order status, document upload, lead capture — that use WhatsApp Flows or a chatbot to collect data in named fields, not as free-text the operations team has to re-key. See our internal tools work for the patterns we use.
A worked example
A 60-person consumer healthcare operator in Bengaluru, four clinics plus a tele-consult line, was on AiSensy doing roughly 80,000 outbound template messages a month and receiving about 14,000 inbound conversations. The AiSensy bill was reasonable. The hidden cost was four full-time operations associates whose job was to read inbound WhatsApp messages and copy structured data — appointments, billing queries, prescription refill requests — into the clinic management system, and then trigger the right outbound template back.
The phase one we built, eight weeks, INR 12.5 lakh fixed price, plus INR 45,000 a month retainer:
- Inbound webhook handler pushing structured records into the clinic management system with patient identity match, intent classification (appointment / billing / prescription / other), and SLA timers
- Outbound utility templates triggered by clinic-system events (appointment booked, prescription ready, follow-up due, invoice generated)
- A WhatsApp Flow for appointment booking that wrote directly to the doctor's calendar in the clinic system
- DPDP consent capture at the patient registration step, with audit log
- Routing rules that sent prescription queries to the pharmacist queue, billing queries to the finance desk, and clinical questions to a nurse triage queue
Outcome over the first quarter: two of the four operations associates moved to higher-value work, prescription refill turnaround dropped from a day-and-a-half to under three hours, and the marketing-template share of the WhatsApp bill fell because most outbound now rode utility-tier event triggers. We have written about adjacent clinic patterns in running a multi-location clinic chain on actual software and software for multi-doctor clinics.
How to start
If you are early — under 5,000 monthly outbound messages and no BSP yet — start on Wati pay-as-you-go or AiSensy's entry plan, set up your top three utility templates, and spend the next quarter learning what your customers actually message about. Do not commission custom work yet.
If you are mid-stage — already on a BSP, 20,000+ monthly outbound, operations team manually bridging WhatsApp and your SoR — that is when a thin custom layer pays back. The phase one scope above is the right starting shape.
If you are late — heavy WhatsApp volume, a BSP that is straining, multiple operations people whose job is now mostly copy-paste — the custom layer is overdue and the maths is almost certainly in your favour. Get a scoping conversation booked.
If you want a fixed-scope, fixed-price read on what a layer-three build would cost for your specific stack, book a discovery call. We will tell you straight whether this is a custom-build problem or whether you are better off staying on the BSP for another two quarters and revisiting.