The most common question we have been asked since announcing Exit.bd is some version of the same thing: why not just build a Bangladeshi Flippa? A public directory of businesses for sale, browsable, filterable, with seller revenue and asking price visible up front. It is the obvious model. It works in the United States, in parts of Europe, in Southeast Asia1. And it is precisely the model we deliberately chose not to copy.
Public listings work where the cost of disclosure is low
On Flippa, the cost to a seller of having a listing visible publicly is roughly nothing. The asset is usually a domain, an e-commerce store, a small SaaS — businesses with no physical staff to spook, no supplier contracts that contain change-of-control clauses, no bank covenants triggered by ownership intent, and no concentrated customer who might terminate on the news. The seller can declare an exit intent on the open internet and lose, at most, some small amount of dignity.
In Bangladesh, the cost of public disclosure of an exit intent is materially higher and almost always asymmetric — it falls on the seller. A garment factory whose largest buyer learns the owner is exploring a sale will, with very high probability, see purchase orders slow within thirty days. A pharma manufacturer whose distributors learn the same thing will see credit terms tighten. A restaurant chain announcing it is for sale will lose its best floor managers to competitors within a fortnight. The information leak does not just affect the deal. It damages the underlying business, sometimes permanently, and the damage is locked in whether the deal happens or not.
Public marketplaces transfer this cost to sellers and call it transparency. We do not think that is the right trade in this market.
Why public listings fail in low-trust markets
The pattern is not unique to Bangladesh. Public business-for-sale marketplaces have repeatedly failed to gain traction in low-trust, low-frequency emerging-market environments where exit-intent disclosure carries asymmetric cost.
- Pakistan, 2019–2021. Two attempts to launch domestic business-for-sale directories quietly wound down within eighteen months, both citing low qualified-buyer throughput against high seller-side reputation cost3.
- Vietnam, 2017–2020. A regional listing platform with strong VC backing pivoted away from public SME listings to closed M&A advisory after observing that listed sellers experienced measurable customer-flight post-listing4.
- Nigeria, 2020–2022. A widely-publicised Lagos-based marketplace launched with public listings, faced repeated complaints about leaked teasers and tyre-kicker buyers, and quietly converted to a closed-broker model in 20235.
- Bangladesh, 2014–2017. An earlier attempt at a domestic listing site (now defunct) reached a peak of approximately forty active listings before stalling on near-zero qualified-buyer activity; the site closed without recording a single completed transaction6.
The trust gap is the real product problem
Bangladesh M&A is a low-trust, low-frequency market. A typical SME owner will sell exactly one business in their lifetime. They do not have repeat-game intuition for what is normal. They cannot easily verify whether the buyer they are speaking to is qualified, funded, serious, or simply gathering competitive intelligence. Buyers face the mirror version of the same problem — listings that exaggerate revenue, hide customer concentration, or describe a business that, on inspection, does not match the teaser.
Solving this is not a directory problem. It is a trust-infrastructure problem. It requires KYC verification of every participant, not just at sign-up but at progressive stages7. It requires buyer briefs that get scored against listings before either side is exposed to the other. It requires a structured progression where neither party reveals more than the current trust level warrants. And it requires enforced consequences — a permanent ban for a buyer who tries to extract information without intent to transact, a verified-history record that follows participants across deals.
How we modelled trust — the progressive-disclosure ladder
The Exit.bd interaction model has six stages, each gating more sensitive information than the last. The progression is deliberately slow and asymmetric — the seller controls every gate upward, the buyer controls every gate downward.
- NDA. Every buyer signs a sector-templated NDA before seeing a teaser. NDA is generated server-side from a seller-provided template, executed via an authenticated portal, and produces a timestamped breach-detection record.
- Q&A. Structured, scoped questions across revenue, customer concentration, regulatory exposure. No identity disclosure. The seller can decline any question.
- Secure chat. Scoped messaging with contact-information regex filtering. Off-platform leakage is the fastest way to lose access; we screen for emails, phone numbers and social handles in every message.
- Deal room. Watermarked documents, granular permissions, full audit trail per access. Q&A threads attach to specific files. Every download is timestamped and traceable to the requesting buyer.
- Offer. Offers and counter-offers chain inside the platform with a full audit trail and a structured term sheet. No off-platform offer is recognised by the workflow.
- Escrow. On acceptance, funds move into Hold.bd escrow at Prime Bank PLC8 and release on milestone — Form 117 lodged9, board resolution passed, share certificate endorsed. No wire-transfer risk, no cross-border FX to bridge by hand.
At every stage, both parties must consent to advance. A seller can engage with five interested buyers at the Q&A stage and choose to progress only one to NDA. A buyer can request escalation; the seller can decline without explanation. The information gradient is deliberate. It is also, we think, the only model that fits the actual risk profile of Bangladeshi M&A.
The cost of closure
We are aware of the trade-off. A closed marketplace has slower liquidity than an open one. A buyer who wants to browse a directory of a hundred restaurants for sale will not find that on Exit.bd. A seller who wants the broadest possible audience will get fewer total impressions. That is the cost of the model, and we have priced it in.
What we get in exchange is, we think, the only thing that actually matters at this stage of the market: deals that close. Liquidity is a vanity metric if no transaction completes. Browse counts are meaningless if they leak the seller's exit intent. We optimised for transaction completion at acceptable terms for both sides, and accepted that the upper-funnel numbers will look smaller than they would on an open platform.
What this means in practice
Concretely: there is no public list of businesses for sale on Exit.bd. There never will be. Sellers list confidentially and are surfaced only to buyers whose investment brief has been verified as a credible match. Buyers do not browse — they receive curated profiles. Both parties operate inside a closed environment with audited trails, watermarked documents, and KYC-verified identities. Hold.bd handles the money. The Companies Act compliance is built into the closing workflow9.
It is a more deliberate, slower-feeling product than a public marketplace. It is also, we are quite confident, the only kind of product that makes Bangladeshi M&A actually work. The trust infrastructure has to come first. The volume follows.