Article

WhatsApp Business Operations for Kenyan SMEs

A WhatsApp operations platform connecting customer conversations, M-Pesa payments and orders into one system — with automated assistants handling routine work while money and decisions stay under human control.

WhatsApp business operations platform — conversations, M-Pesa payments and orders

A Kenyan business receives customer messages on WhatsApp, takes payments through M-Pesa, and manages orders across spreadsheets and chat. A single customer query often means checking three systems before anyone can say whether an order was paid or what should happen next.

We built a WhatsApp operations platform that connects customer conversations, M-Pesa payments and orders into one system — with automated assistants handling routine work while money, customer data and consequential decisions stay under human control.

The problem Kenyan businesses face

WhatsApp is where Kenyan customers actually are. But it is also where business operations fall apart:

  • A customer messages on WhatsApp to ask about an order.
  • The team opens WhatsApp to read the message, M-Pesa to check if payment arrived, and a spreadsheet to find the order.
  • If anything does not match, someone calls the customer back — hours later.

This friction costs sales. A customer who does not get a fast answer buys from a competitor who does.

The platform we built

The platform connects WhatsApp conversations, M-Pesa transactions and order records into a single shared inbox and business record. Instead of three systems, there is one place where a team member can see the customer, the conversation, the payment and the order together.

Automated assistants handle routine work — confirming order details, answering pricing questions, reminding customers about pending payments — while money, customer data and consequential decisions remain under human control.

How it works

Customer journey:

  1. A customer messages your business on WhatsApp.
  2. The conversation appears in your shared inbox alongside their order and payment history.
  3. An automated assistant answers routine questions (price, availability, delivery) and flags anything that needs a person.
  4. When the customer pays via M-Pesa, the payment is matched to their order automatically.
  5. If anything is unclear — an incomplete payment, a wrong amount — the system flags it for review instead of guessing.
  6. The team confirms the order, and the customer gets confirmation on WhatsApp.

What it includes

Shared inbox and business records

One inbox for WhatsApp conversations, connected to contacts, orders, permissions and history. Every team member sees the same customer record — no more asking "did anyone reply to this customer?"

M-Pesa payment matching

Transactions are checked, matched to orders, and exceptions are investigated. Exact matching (receipt, merchant reference, expected amount, time window, customer details) runs first. Unclear payments are flagged for human review — the system never guesses that a payment succeeded.

Automated assistants

Assistants answer routine questions, check lead details and recover stalled sales. They work within clear access levels: they can help and suggest, but they cannot confirm payments or change orders without a person.

Reports and dashboards

Revenue and operations dashboards show conversation volume, payment match rates, pending orders and response times — so you know where the bottlenecks are.

Why it matters for Kenyan SMEs

M-Pesa and WhatsApp are the two tools Kenyan small businesses already use every day. This platform connects them instead of keeping them separate. The result is faster customer responses, fewer missed payments, and less time spent copying information between spreadsheets.

If your business sells through WhatsApp and collects payment through M-Pesa, this is the system that should already exist.

Run your business on WhatsApp and M-Pesa — together

We build WhatsApp operations platforms that connect conversations, payments and orders for Kenyan businesses — with automated assistants and human control where it matters.

Explore AI Engineering & Infrastructure

What we're learning,
while we build it.

One email a month. Practical notes on software, cloud and production AI, written by the engineers doing the work. No filler.