Saying yes is easy. Saying no is where most projects fail.
When you're building a platform, the temptation to do everything is overwhelming. A marketplace? Yes. A mobile app? Yes. A social commerce feature? Absolutely. An affiliate programme? Of course. A blog system? Why not.
I've watched projects collapse under the weight of their own ambitions. The team spends 18 months building and never ships anything because they're always adding one more thing. I've also built enough software to know that every feature you add is a surface area that can fail, a thing that needs to be maintained, a complexity that slows down everything around it.
So before I wrote a single line of Manmerce code, I wrote a different kind of document. Not a list of what to build. A list of what I was willing to commit to — and what I was explicitly choosing to ignore, at least for now.
What's In: The Core Commitment
Manmerce is a self-hosted e-commerce platform for brands that sell products — physical and digital — both online and at a physical counter. That's the one-sentence version.
Self-hosted means you install it on your own server and you own your data completely. This is a deliberate choice. It means you're not paying a monthly SaaS fee forever. It means your store doesn't go down because a third-party platform has server problems. It means you can customise anything — your store is your code.
The core scope breaks down into six groups, in order of how essential they are:
1. Storefront & Customer Experience — Themes, visual builder, multi-language, product pages, the cart, the checkout. This is what customers see. It has to be perfect.
2. Catalog & Inventory — Products, variants (sizes, colours, materials), brands, labels, stock tracking across multiple locations. This is the backbone of what you sell.
3. Checkout & Payments — The checkout flow, COD handling, mobile OTP, bKash/Nagad/Rocket/SSLCOMMERZ/card integrations, subscriptions, EMI. This is where money moves.
4. Operations & Logistics — POS for physical stores, courier API integrations, fraud detection, thermal printing, returns. This is how you fulfil what you've sold.
5. Marketing & Retention — Coupons, loyalty points, referrals, abandoned cart recovery, WhatsApp/SMS/email communication, SEO tools. This is how you grow.
6. Finance & Admin — Bookkeeping, expense tracking, role-based access control, activity logs, team management. This is how you run the business behind the store.
What's Out (For Now)
A marketplace — where multiple sellers list products on one platform — is out of scope. Marketplaces are a completely different product with completely different dynamics. Building a great single-brand store first is the right call.
A mobile app is out. I know this seems surprising for 2025. But a well-built mobile-optimised web store performs better than many native apps for the buying journey, and building a separate iOS and Android app alongside the platform would double the development surface area. It's coming eventually — but not in version 1.
Social commerce integrations — selling directly through Facebook Shop, TikTok Shop, Instagram — are out. These platforms change their APIs constantly and they're not stable enough to build a core dependency on.
A built-in blog or content management system is out. Manmerce is for selling products, not publishing content. If you want a blog, put it on a separate domain and link to your store.
Why This Matters
Saying no to these things meant I could say an unambiguous yes to everything that was in scope. It meant I could build the checkout properly instead of rushing it. It meant the courier integration could be deep and reliable instead of shallow and fragile.
Scope is not a limitation. Scope is focus. And focus is the only way to build something that's actually excellent rather than something that does everything poorly.
Next week: how I plan before I build — and why most planning goes wrong.
Mustafa Kamal Hossain
Founder & Principal Engineer at Manfi. Passionate about Laravel, SaaS architecture, and high-performance engineering.