Architecture respected
Plugins, observers and service contracts – never core edits. A patched core is why so many of these stores cannot be upgraded.
For catalogues and rules that outgrew everything simpler.
Magento is heavy, expensive to run and genuinely the right answer for large catalogues, multi-store operations and B2B pricing rules. I build on it, tune it and, where the honest answer is that you have outgrown the need for it, help you leave.
Magento exists for stores whose complexity would defeat everything simpler. Tens of thousands of products. Multiple storefronts sharing one catalogue but differing in currency, language, tax and pricing. Customer groups with negotiated prices, tiered quantities and quote workflows. Attribute sets where different product families genuinely need different data.
None of that is unusual in wholesale, distribution, industrial supply or multi-market retail. The problem is that the same capability makes it a poor choice for a store with two hundred products and one price list – the hosting bill alone is several times what a simpler platform costs, before anyone writes a line of code.
The platform also demands more of its infrastructure than most: indexing, caching layers, search services and cron all have to be configured and monitored rather than assumed. A store on inadequate hosting is not slow because the platform is slow; it is slow because it was never given what it needs.
So the work divides. Building and extending stores where the complexity is real. Tuning stores that are struggling with catalogue size or infrastructure. Upgrading installs that have fallen behind on security releases. And, where a business has simplified since the store was built, migrating to something cheaper to run – which is a legitimate outcome, not a failure.
Modules written to the platform’s architecture – dependency injection, plugins and service contracts – rather than core edits.
Functionality that survives upgrades instead of blocking them.
Customer-specific pricing, tiered quantities, quote requests, company accounts and purchase approval flows.
Wholesale rules enforced by the store rather than handled over email.
Indexing strategy, search configuration, caching layers and query work for catalogues that have grown past comfort.
Category and search pages that stay usable at real catalogue size.
Store views, websites and scopes configured so several storefronts share a catalogue without duplicating it.
Several markets run from one admin rather than several stores kept in sync by hand.
Security releases and major version moves, with third-party extension compatibility checked before anything is promised.
Security support without an upgrade that takes the store down.
Moving onto the platform where complexity warrants it, or off it to WooCommerce or Shopify where it does not.
A platform that matches the business, with the catalogue and URLs carried across.
On a platform this large, discipline matters more than cleverness.
Plugins, observers and service contracts – never core edits. A patched core is why so many of these stores cannot be upgraded.
The infrastructure requirement is stated up front. A store on shared hosting will not be rescued by optimisation.
Code and configuration in version control and deployed, because production changes made by hand are unrepeatable.
Performance tested against a realistic catalogue, not a demo dataset where everything looks fine.
If your store no longer needs this platform, you will hear that with the cost of moving to something cheaper to run.
Custom modules, configuration and integrations written down, because complexity without documentation traps you with one developer.
Stores where the rules are complicated, not just the catalogue.
Thousands of SKUs with attribute sets that differ by product family.
Negotiated pricing, tiered quantities, company accounts and approval workflows.
Several storefronts differing in currency, language, tax and price, sharing one catalogue.
Deep product data, dealer pricing and ERP integration as a requirement rather than a nice extra.
Catalogues that have grown past what the current setup handles.
Businesses now running a fraction of the complexity they are paying for.
Indian requirements matter more on a store of this size, not less. GST with correct interstate handling, e-invoicing where turnover requires it, HSN codes against products, and the difference between a B2B invoice and a retail one are all configuration rather than afterthoughts.
Payment and logistics follow the same pattern: Razorpay, PayU and Cashfree rather than assuming Stripe, cash on delivery treated as a first-class method, courier integrations for the services your warehouse actually books, and pin-code serviceability checks at the cart.
I work with businesses across Bangalore and India, and with clients in the UK, USA, Canada, the UAE and Singapore – where the tax and logistics differ but the platform work does not.
Tax, HSN codes and invoice formats configured correctly.
Razorpay, PayU and Cashfree, plus cash on delivery.
Courier integration and pin-code serviceability.
India, the UK, USA, Canada, the UAE and Singapore.
Serving Chennai, Bangalore, India, Canada, UK, USA, Dubai, Singapore.
Sectors where the catalogue or the pricing is genuinely difficult.
Typically needs: Vast catalogues, deep specifications and dealer-specific pricing.
How I help: Attribute sets per product family, customer-group pricing, and specification data that is searchable rather than trapped in PDFs.
Typically needs: Company accounts, negotiated rates, quote flows and purchase approvals.
How I help: B2B account structures with approval chains and quote requests, tied into the ERP.
Typically needs: Fitment data where the wrong part is worse than no part.
How I help: Vehicle compatibility as structured attributes, driving search and filtering that can be trusted.
Typically needs: Several storefronts with different branding on a shared catalogue.
How I help: Website and store-view scoping so one catalogue serves several fronts without duplication.
Typically needs: Specification-led comparison and fast-moving stock across channels.
How I help: Comparison views from real attribute data, with stock synchronised to the systems of record.
Typically needs: Category restrictions, documentation and controlled fulfilment.
How I help: Rules enforced at cart level, with the documentation requirements built into checkout.
Six stages, with infrastructure settled before anyone writes a module.
I audit what exists or, for a new build, what the catalogue and rules actually require: product data shape, pricing logic, integrations, traffic and order volume.
You get: A written assessment, an honest platform recommendation and a fixed quote.
Why it matters: This platform is expensive to run. Confirming you need it is the cheapest hour of the project.
Attribute sets, category structure, customer groups, pricing rules and store scopes are modelled, and the hosting and service requirements settled.
You get: A data model and infrastructure plan you have signed off.
Why it matters: Catalogue structure and hosting are the two decisions that are painful to change once a store is live.
Design for the templates that sell: category, product, cart and checkout, built against real product data at phone width first.
You get: Designs for the key templates, approved before building.
Why it matters: A demo catalogue hides every layout problem your real product data will cause.
The build: theme, custom modules, integrations and data import, in version control and deployed to staging rather than changed in production.
You get: A working store on staging, deployable repeatedly.
Why it matters: Configuration changed by hand in production is why these stores become impossible to replicate.
Full order testing across devices and customer groups, pricing rules verified, tax and shipping checked against real scenarios, and performance measured at full catalogue size.
You get: A tested store, the performance numbers and verified pricing rules.
Why it matters: B2B pricing errors are expensive and silent – the customer rarely reports being undercharged.
Launch at a quiet hour, with monitoring, indexing and cron verified, then training on catalogue, orders and pricing administration.
You get: A live store, trained staff and documentation of the custom work.
Why it matters: Indexing or cron left misconfigured is the most common reason a new store degrades within weeks.
The platform, the services it depends on, and the simpler alternatives.
Scoped against your catalogue and your pricing rules.
A store this size creates SEO problems by sheer volume: layered navigation generating endless filter combinations, the same product reachable through several category paths, and thousands of URLs for products nobody searches for.
Most of it is configuration rather than content. Canonical rules, robots directives and crawl management are decided once and then apply across the catalogue. Nobody can guarantee a ranking – the aim is that crawl budget goes where the revenue is.
A well-modelled catalogue is already structured data: every specification lives in a named attribute rather than inside a description. That makes product information straightforward to mark up accurately and easy for anything automated to read – provided the attributes are actually populated.
The practical reasons, rather than the adjectives.
A great many of these stores are running complexity the business no longer has. If that is yours, you will hear it with the cost of moving somewhere cheaper to run.
Plugins and service contracts only. A patched core is the single most common reason one of these stores cannot be upgraded, and I will not leave that for the next person.
This platform needs real infrastructure. You get told that before the build, not after three months of optimising around a server that was never adequate.
GST, e-invoicing, HSN codes, Indian gateways, cash on delivery and pin-code serviceability configured as part of the work, not bolted on afterwards.