Antor logo
Antor
👋 Start a project
antor. / about / fail-forward
🔥 TRANSPARENCY

Fail Forward:
My Graveyard of Ventures.

Raw, unfiltered post-mortems of failed projects. Because you don't build 20+ shipped companies and 14 years of experience without collecting a few scars along the way.

Most portfolios only show the wins. They show the exit numbers, the press releases, and the hockey-stick charts. But that is only half the story. The grit is forged in the failures. Every successful venture in the NextBangla ecosystem is built directly on the lessons of a project that didn't make it. Here is my raw graveyard—unfiltered, plain, and fully analyzed.

The Post-Mortems

Since starting my journey in 2012, I have shipped over 20+ ventures and 500+ client projects. Some scaled to thousands of users, while others crashed on contact with the market. Here are four case studies from the archive.

2018 💔 FAILED

MewMela: The Pet Marketplace

A premium local pet marketplace, community, and adoption platform for animal lovers in Bangladesh.

What went wrong:

We built a stunning, highly-interactive UI and acquired 5k+ community members. But the local pet supply chain was unorganized, and transactions shifted to peer-to-peer offline deals, leaving us with zero commission leverage.

The Lesson:

A beautiful community without a built-in escrow or centralized transactional mechanic is just an expensive hobby. We now bake monetization deep into the product wireframe, never as an afterthought.

2020 💔 FAILED

HelSebe: Healthcare Booking

A localized digital booking and appointment directory connecting regional clinics with rural patients.

What went wrong:

We underestimated the operational friction of medical calendars. Regional clinics were highly non-technical, and doctor schedules changed on a whim, leading to canceled appointments and high support overhead.

The Lesson:

Avoid heavy offline operations in early MVPs. If you cannot automate 95% of the sync loop via APIs, do not build the product. This directly inspired the automated, low-overhead architecture of VOXLY and Pawport.

2016 💔 FAILED

KaruBazaar: Artisan Textiles

A traditional hand-crafted textile marketplace connecting regional weavers in Nilphamari with international buyers.

What went wrong:

The shipping and logistics costs for single-item boutique textile exports from Nilphamari to Europe were completely prohibitive. We could not achieve price-competitiveness on retail orders.

The Lesson:

Weightless software beats physical supply chains every time for early-stage solo builders. This realization caused us to pivot NextBangla Ltd's roadmap heavily toward high-margin digital SaaS and AI-native products.

2019 💔 FAILED

ShadGhor: Food & Discovery

A premium gourmet local restaurant directory and food-spot discovery engine for foodies.

What went wrong:

We built a top-tier visual directory, but local restaurants refused to pay listing retainers. Larger delivery aggregators swept in with massive capital and locked down exclusive distribution networks.

The Lesson:

Never compete directly with capital-heavy aggregators on generic consumer plays. Find a specific, high-intent niche wedge (like enterprise tools) rather than broad consumer directories.

The Grit Behind the Numbers

Building 7 AI startups in parallel today is only possible because we paid our dues. We know exactly where the traps are because we fell into them in 2016, 2018, and 2020. Our partner network, our infrastructure, and our 14 years of mathematical timeline compliance represent real lessons bought with hard cash and sleepless nights.

Looking for a partner who has been in the trenches?

We don't bring agency theory. We bring battle-tested operational execution.

Partner With Antor