There are three honest routes to an online store, and the right one depends far more on your catalogue, your margin and your team than on any feature comparison table. Below is a five-year total-cost view of each, which local payment gateways work with which, and the specific signals that mean you have outgrown a hosted platform.
The short version
- Shopify if you want to be selling in three weeks with no technical staff, and your margins can absorb roughly 3–6% of revenue in platform and app costs.
- WooCommerce if you want ownership and low running costs, already have a WordPress site, and have someone who will keep it updated. It is not maintenance-free.
- Custom build only when your business model does not fit a product catalogue — complex configurators, B2B pricing tiers, rentals, subscriptions with unusual rules, or deep ERP integration.
- Most stores under about 500 SKUs should not be custom-built. Most stores over PKR 5M/month in revenue with unusual logic should not stay on a template.
- The deciding factor is usually who maintains it, not what it can do.
Five-year cost, side by side
Platform fees look small monthly and large over five years. Custom builds look large upfront and flatten out. Here is the comparison we walk clients through, for a store with roughly 200 products doing moderate volume.
| Shopify | WooCommerce | Custom build | |
|---|---|---|---|
| Initial build (PKR) | 150,000 – 500,000 | 250,000 – 800,000 | 1,800,000 – 6,000,000 |
| Time to launch | 2–4 weeks | 4–7 weeks | 12–24 weeks |
| Platform fee / month | USD 39–399 | None | None |
| Hosting / month (PKR) | Included | 4,000 – 25,000 | 15,000 – 80,000 |
| Apps / plugins / year (PKR) | 60,000 – 300,000 | 25,000 – 120,000 | Minimal |
| Maintenance / year (PKR) | Low | 80,000 – 250,000 | 15–20% of build |
| Transaction fee beyond gateway | 0.5–2% unless using Shopify Payments | None | None |
| Who can operate it | Anyone | Semi-technical | Needs a developer |
The line that surprises people most is Shopify apps. The platform is genuinely excellent, but the business model pushes functionality into paid apps. A realistic store ends up with a reviews app, a subscription app, an upsell app, an advanced-shipping app and a reporting app — each USD 15–60 per month. Five apps is comfortably USD 150/month before you have sold anything.
The line that surprises people second most is WooCommerce maintenance. "Free and open source" describes the licence, not the workload. WordPress core, WooCommerce, the theme and every plugin update on independent schedules, and occasionally break each other. Budget for someone to own that, or budget for the day it goes down during a sale.
Shopify: what it is genuinely good at
Shopify's real product is not the storefront. It is the removal of every operational problem you did not want: hosting, PCI compliance, uptime, checkout security, card-testing fraud, CDN, SSL, scaling on a spike. That checkout has been optimised against more traffic than any custom build will ever see, and you inherit that work for a monthly fee.
Choose Shopify when:
- You need to be trading quickly, and speed to market beats flexibility.
- You have no technical staff and no appetite for maintenance windows.
- Your catalogue is a conventional product catalogue — variants of size, colour, capacity.
- You sell internationally and want multi-currency handled for you.
- Your margins comfortably absorb the platform cost.
Where it constrains you: checkout customisation is limited outside the enterprise tier; unusual pricing logic (customer-specific rates, complex bundles, rental periods) fights the data model; your data lives with a vendor whose pricing you do not control; and deep integration with a local accounting or ERP system usually means building a middleware service anyway.
Payment gateways in Pakistan
This is the practical detail that decides many stores. Shopify Payments is not available in Pakistan, so you integrate a local gateway or a third-party processor — which is where the extra Shopify transaction fee applies. Local options such as JazzCash, Easypaisa, PayFast and bank gateways generally have official or community WooCommerce plugins and documented APIs suitable for custom builds. Verify current availability and rates directly with each provider before you commit to a platform — this landscape changes faster than any article.
WooCommerce: ownership with a maintenance bill
WooCommerce is the pragmatic middle. You own the code and the database, hosting is cheap, local payment integrations are well supported, and the plugin ecosystem covers almost any requirement — often with a one-time licence rather than a monthly fee.
Choose WooCommerce when:
- You already run WordPress and your content and blog drive traffic.
- You want to avoid recurring platform fees and own your data outright.
- You need a specific local payment or courier integration that exists as a plugin.
- You have — or will hire — someone semi-technical to keep it updated.
Where it hurts: performance degrades as plugins accumulate; a store with 25 plugins is slow by default and needs deliberate work (our speed checklist has a WordPress section). Product counts above roughly 10,000 with heavy filtering need real database tuning. And you are responsible for security — an unpatched plugin on a store taking card details is a genuine liability, not a theoretical one.
Custom: when it is actually justified
We build custom e-commerce, and we talk clients out of it more often than into it. The honest test is whether your business logic fits the shape of a product catalogue. If it does, a platform will serve you better and cheaper. If it does not, you will spend years fighting the platform and paying for plugins that half-solve the problem.
Genuine reasons to build custom:
- Configurable products with dependent options — made-to-measure, print-on-demand with live previews, machinery with compatibility rules.
- B2B pricing — per-customer price lists, negotiated contract rates, credit limits, quote-to-order workflows, purchase orders.
- Rentals or time-based inventory — availability calendars, deposits, late fees, overlapping bookings.
- Deep ERP or POS integration where stock, pricing and invoicing must stay in sync in real time across branches.
- Marketplace models with multiple vendors, commission structures and split payouts.
- Regulatory constraints requiring specific data residency or audit trails.
A custom store is a software product. It needs the same discipline as any other: a requirements document, a role matrix, a QA cycle and a maintenance budget. Treat it as a website and it will behave like a failed project.
Six signals you have outgrown your platform
Migration is disruptive and should be triggered by evidence, not ambition. These are the signals we treat as real:
- Platform and app fees exceed the annualised cost of a custom build. Do the arithmetic; it happens sooner than people expect at high volume.
- Three or more apps exist purely to work around one structural limitation. That limitation is not going away.
- Someone is doing daily manual data entry between the store and your accounting or inventory system.
- You are turning down business because the store cannot express a pricing arrangement a customer wants.
- Checkout conversion is capped by a flow you are not permitted to change.
- Your catalogue or traffic has grown past what the platform performs well at, and tuning options are exhausted.
One signal is not enough. Three is a business case.
What matters regardless of the platform
Stores fail on these far more often than on platform choice:
- Product photography and descriptions. The single highest-return investment in e-commerce and consistently the most under-budgeted. Good photos beat a better platform.
- Mobile checkout. Most of your traffic, and where most abandonment happens. Test it on a real mid-range phone on a real mobile connection.
- Shipping and returns clarity. Unclear delivery cost and timing is a top abandonment cause everywhere, and returns policy drives trust in markets where online buying is still cautious.
- Speed. E-commerce is the category where load time most directly touches revenue.
- Abandoned-cart recovery and transactional email. Cheap to set up, measurable return.
- Honest stock accuracy. Selling something you do not have costs a customer permanently.
We have seen a plain WooCommerce store with excellent photography and two-day delivery outsell a beautifully engineered custom platform with stock photos. The platform was never the variable.
If you are choosing right now, tell us your catalogue size, your monthly volume, your margin and any pricing rule that feels unusual. Those four answers decide it in about ten minutes. See our e-commerce development page, or get in touch — including if the answer turns out to be "stay on Shopify", which it often is.