
From Keyword Search to Conversational Intent: The Blueprint for AI Search Dominance
How modern buyers use multi-variable natural queries and how structured schema makes your business the default recommended choice
Kenyan retailers have the same dilemma. They want to sell online. They know M-Pesa is non-negotiable. And they are staring at three options: Shopify, WooCommerce, or a custom build. They have read blog posts written by Americans who have never sent an STK push in their lives. They are confused.I have built e-commerce systems on all three platforms. I have integrated the Safaricom M-Pesa Daraja API more times than I can count. I have watched clients celebrate their first online sale and I have watched others burn money on the wrong platform. This is the honest breakdown I give every Kenyan business owner before we write a single line of code.
M-Pesa is not just a payment method in Kenya. It is the payment method. Over 96% of online transactions in Kenya happen via mobile money. If your e-commerce platform cannot handle M-Pesa cleanly, you do not have an e-commerce business. You have a brochure.
Shopify is a Canadian company. Its native payment gateways are Stripe, PayPal, and Shopify Payments. None of these work cleanly for Kenyan shillings and local mobile money.To accept M-Pesa on Shopify, you need a third-party app. Options exist—some Kenyan developers have built Shopify apps that connect to the Daraja API. But here is the reality:
There is a workaround using Shopify's manual payment methods, where you generate a Paybill or Till number and ask customers to pay offline, then upload a screenshot. That is not e-commerce. That is chaos.
WooCommerce runs on WordPress, and because it is open-source, Kenyan developers have built direct M-Pesa integrations. Plugins like WooCommerce M-Pesa by Osen or Lipa na M-Pesa by Kopokopo connect directly to the Daraja API.What works:
What breaks:
WooCommerce can work well with M-Pesa, but only if it is built and maintained by someone who understands both WordPress and the Daraja API.
A custom Next.js (or similar modern framework) build gives you complete control. You own the code. You own the checkout flow. You own the API integration.How we build it at AngazaDesk:
The advantages:
The downside:
This is where Kenyan business owners get shocked. Let me break down the real math.
Real annual cost for a small Kenyan store: KSh 50,000 to KSh 100,000 in subscriptions and fees alone, before you make a single sale. And you are paying in USD, which means your costs go up every time the shilling weakens.
Real annual cost: KSh 20,000 to KSh 60,000, depending on hosting and premium plugins.
Real first-year cost: KSh 200,000 to KSh 600,000.
Kenyan shoppers are impatient. We have measured this. If a checkout page takes more than three seconds to load on Safaricom 4G, abandonment rates jump to over 60%.
Shopify handles hosting and CDN. Their infrastructure is fast globally. But in Kenya, we have seen Shopify stores load in 4–6 seconds on mobile data. Why? Bloated themes, heavy apps, and unoptimized images. You can fix some of this, but you are locked into their architecture.
WooCommerce can be fast, but it is usually slow. Most Kenyan WooCommerce sites we audit load in 6–10 seconds on mobile. The culprits:
With proper optimization—good hosting, image compression, caching, and a lightweight theme—WooCommerce can load in under two seconds. But that requires expertise. Most store owners do not have it.
Next.js is built for performance. It has automatic image optimization, code splitting, and server-side rendering. A well-built Next.js store loads in under one second on desktop and under two seconds on Safaricom 4G.We built a fashion store for a client in Nairobi using Next.js. Their previous WooCommerce site loaded in 8.4 seconds. The Next.js site loads in 1.2 seconds. Their mobile conversion rate tripled. Not doubled. Tripled.
The standard WooCommerce checkout is a nightmare on mobile. Seven form fields. Billing address. Shipping address. Account creation. Password. Then payment. Kenyan mobile shoppers do not have the patience.
For both WooCommerce (via custom checkout plugins) and Next.js (via custom code), we build one-page or two-step checkouts:
No account creation required. No email field if they do not want to give it. No shipping address for digital products. We collect only what we need.For physical products, we use WhatsApp to confirm the delivery address after payment. It sounds informal, but it works. Kenyans trust WhatsApp. We have seen higher completion rates with a "Pay now, confirm details on WhatsApp" flow than with a rigid form that asks for everything upfront.
After building on all three platforms, here is what I tell clients.
You are selling digital products or physical goods to an international audience, and M-Pesa is a secondary payment method. Or you have a large budget, you value simplicity over control, and you can absorb the USD subscription costs. If you are targeting diaspora customers with credit cards, Shopify makes sense. For a pure Kenyan market, it is an expensive compromise.
You are a small to medium retailer with a budget under KSh 100,000 for the initial build. You need to launch fast. You have someone who can maintain WordPress updates and plugin conflicts. You want M-Pesa integration without building from scratch. You are willing to invest in good hosting and optimization.WooCommerce is the workhorse of Kenyan e-commerce. It is not perfect, but with the right setup, it works. We have built profitable WooCommerce stores for clients in Mombasa, Nakuru, and Nairobi. The key is not cutting corners on hosting and security.
You are doing serious volume—KSh 500,000+ in monthly online sales. You want complete control over the customer experience. You want a checkout flow that converts at the highest possible rate. You want a site that loads instantly on mobile. You have the budget for proper development and you see this as a long-term investment, not an experiment.We recommend custom builds for:
Sometimes we combine platforms. We build a custom Next.js frontend for speed and user experience, but connect it to WooCommerce or Shopify on the backend for inventory management. The customer sees a blazing-fast, custom-branded store. The business owner manages products in a familiar dashboard. This costs more, but it gives you the best of both worlds.
Key Takeaway: There is no "best" platform. There is only the right platform for your specific business, budget, and technical capacity. If M-Pesa is your primary payment method, Shopify will fight you. WooCommerce will work if you maintain it. A custom build will dominate if you invest in it. Be honest about where you are today, and build for where you want to be in twelve months.

How modern buyers use multi-variable natural queries and how structured schema makes your business the default recommended choice

A deep dive into vector embeddings, semantic proximity, and how LLM retrieval models mathematically select which business to recommend.