It was a Tuesday in May of 2019, on the second floor of a cha-er dokan in Dhanmondi, the kind of place where the fan wobbles and the milk tea arrives in glasses the colour of wet cement. There were five of us at the table. One of my fellow founders was rolling a cigarette he had no intention of smoking, and the lead engineer was staring into his glass as if the chini at the bottom owed him an apology. I was watching the shadow of my own glass, long in the late afternoon light, crawl half an inch across the wood. We sat with the silence for the better part of ninety minutes before anyone said the thing.
"Bhai, ami mone kori, hoye gechhe."
Brother, I think it is done. That sentence, spoken in the unsteady light of a tea shop on Road 27, was the most successful thing PorashonaOnline ever shipped.
There is a whole genre of writing about how to start a company, and almost nothing serious in it about how to end one. Most founders I have known in Dhaka treat winding down as a thing that happens to them, on the order of a flood or a diagnosis, something to be survived rather than performed. The quieter argument I want to put to you, with the patience of an older man, is that winding down well is a separate skill from building. What people end up calling resilience seven years later is, in my long experience, simply the residue of someone who failed successfully and kept the receipts.
What does success inside a failure actually look like in practice? In my own count there are roughly four moving parts to it, less a checklist than the four corners of a tarp you are trying to hold flat against the wind on a Mohammadpur rooftop in monsoon. The first corner is team relationships preserved. When you wind down a startup you are also closing a small, weird, intense friendship that ten or eleven people built without quite realising it. If that friendship survives the closing weeks, almost everything else turns out, in time, to be salvageable.
The second corner is investors briefed honestly. PorashonaOnline was funded through TeamNinja Tech under the World Bank's HEQEP program, so our cap table was, in effect, a grant adjacent to government rather than a venture instrument. There is no clean institutional ritual for shutting down on public money. We wrote them a real letter, with the unit economics laid out plainly and the place where the math broke clearly marked. I would not have been able to look myself in the mirror in the years following had we not sent it.
The third corner is the IP left unpoisoned. We did not sue each other over the codebase, and we did not quietly fork it. The product is technically still on the Play Store today because two teammates wanted to keep it running on fumes, and we said go ahead, godspeed. That generosity costs almost nothing on the day, and compounds quietly for years. The fourth and final corner is a story you can tell. If you cannot describe your own failure to a stranger at a wedding without going pale around the ears, you have not, in my view, actually finished failing yet.
তোমার সেই কোম্পানি কেমন চলছে?
How is that company of yours doing? In the first year you will say "alhamdulillah, chaltese" most of the time, because the real answer takes longer than the wedding queue allows, and the aunty in front of you has six more relatives to greet before the milad starts. By the second year the answer has become "ah, oita to bondho hoye gechhe," and you will watch a small, secondhand grief move across her face on your behalf. By the third year you graduate to the longer version. "We shut it down, the math did not work, I learned a great deal." The aunty will nod the way she might nod at a minor misdemeanour the family has already forgiven.
The parts of this connected to government had their own particular shape. Meetings in offices where the air conditioning ran too cold, biscuits nobody touched, polite officials who wanted to know if anything could be salvaged out of the work. We told them, as gently as we could, that there was nothing left worth the next tranche. They were, to their lasting credit, gracious about it.
You also, if you are being honest with yourself in the small hours of the morning, grieve. It isn't the company itself you're grieving, so much as the version of yourself the company had implied, the man of 24 who had thought he was going to be on a magazine cover by the time he turned 30. That young man has to go somewhere when the company closes, and for a year he sulks in your chest and refuses to come down to dinner.
I came across Annie Duke's Quit in the autumn of 2023, on the recommendation of a friend recently retired from the bond desk. Her observation that people who quit on time look like quitters, while people who quit too late look like heroes who learned a lesson, is the most honest single sentence in modern business writing. I would put it next to a passage from David Whyte's Crossing the Unknown Sea, where he proposes that the antidote to exhaustion is not rest but wholeheartedness. Winding PorashonaOnline down was the most wholehearted thing I had done in the two years of running it, and that, oddly, was why it felt rather a lot like rest.
Nobody quite tells you that the team you shut a company down with can turn out to be useful later. Mine has, seven years on, become the most useful professional network I possess, and it isn't because we owe one another anything. We simply ended cleanly. Most of them would still pick up the phone tonight if I rang.
Two of them have referred me into roles I would not otherwise have heard about, in the quiet way that an introduction in a Banani drawing room used to be made in my father's day. A third sends me product critiques sharper than anything I currently pay analysts for. There is also one I sit with for cha twice a year at a small dokan off Kemal Ataturk Avenue, where the unspoken arrangement is that neither of us mentions the company by name, which has, over time, become its own quiet form of friendship.
I have watched the contrasting case play out up close. There is a particular class of breakup that begins with one bad equity conversation in a Banani coworking space and calcifies, over the better part of a decade, into two grown men no longer speaking at weddings they have both been invited to. Nobody is punishing them for having failed. What follows them for years is the running cost of having handled the failure cruelly while the small Dhaka founder community happened to be watching, and the market here will forgive almost any commercial outcome you can show up at a wedding with. It will not, in my experience, forgive what people did to one another in the closing weeks.
It took the better part of several years for clarity and peace to find each other again, and they did not arrive in any particular order. The gap between them is where most of the unhelpful advice in this country tends to live. People will rush to close it for you, telling you to "move on" or to "draw a line under it," advice built for a problem of clarity when what you are actually missing is mostly peace. What actually works, in my own slow experience, is the boring kind of time, paired with the small ongoing courtesy of keeping in touch with the people who happened to be in the room when the company died.
If you are winding something down as you read this, in a Dhanmondi cha-er dokan or in your own kitchen at one in the morning, I am not going to tell you that it gets better. I do not know that, and I would, if I am honest, resent anyone who promised it to me. The version of you who walks into the meeting with the team next Tuesday morning is the version your future career will quietly be built upon. That version is the closer who handles the last week with care, not the founder who pitched the deck three years ago in a spare bedroom. Be precise about what is true and kind about what is painful, pay people whatever you can actually pay them, write the honest letter even when nobody is asking for it, and hold the tarp flat at all four corners for one more week. Nobody has asked me about that cap table in seven years. What comes up instead, when it comes up at all, is the last month, and the people who bring it up are never the ones I expected.