E-commerce development

We build online stores as part of web and app development — WooCommerce, Shopify, OpenCart and Magento, with the payment gateway, tax handling, inventory and shipping integrated rather than listed as future phases. A store is a system that has to reconcile against money, which makes it a different job to a website.

Choosing the platform is the decision that sticks

Everything else on an e-commerce project can be changed later. The platform effectively cannot — migrating means moving products, customers, order history and URLs, and doing it without losing search rankings.

• WooCommerce — you want control over the checkout, the data and the customer relationship, and you accept that you own the maintenance. The most common right answer for Indian SMEs, especially anyone already on WordPress.
• Shopify — you’d rather pay monthly than maintain anything, and you can live inside its rules. Fastest to launch and least to go wrong, at the cost of flexibility at the edges. Its defaults need work for Indian tax and shipping, which is most of what we do on a Shopify build.
• A custom platform — when neither of the above can model how you actually sell. B2B pricing tiers and approval flows, deep coupling to an ERP or inventory system, subscription or rental mechanics, or a catalogue that isn’t really a list of products. Custom costs more up front and is justified by fit, not by preference.

We’ll ask about catalogue size, variants, tax treatment and who runs it day to day before answering — and if the honest answer is the cheapest option, that’s the one we’ll say. Custom is the right call less often than the people selling it suggest.

We also maintain stores on OpenCart and Magento, usually for clients who arrived on them. If you’re already running one, a migration is a real option but not an automatic one — see below.

What gets skipped and then hurts

Store builds tend to fail in the same few places, all of them after launch rather than during:

• Payments tested only in sandbox. A gateway that works on test credentials can still fail on live settlement, refunds or partial captures. We test with real transactions before launch.
• GST configured afterwards. Tax rates, HSN codes and invoice formats set up at the first filing rather than during the build means reconciling months of orders by hand.
• No plan for who runs it. Somebody has to add products, handle a stuck order and answer a customer. If that person isn’t named and trained at handover, the store degrades within a quarter.
• Maintenance treated as optional. Stores carry payment credentials and customer data, which makes unpatched plugins a different class of risk to a brochure site. That belongs in an AMC with a defined schedule.

FAQ

Frequently asked questions