We submitted a Bangla POS app to 50 directories. Here is what actually happened.

Tukri — Sat, 31 Oct 2026 01:17:22 +0000

Tukri is a point-of-sale, stock and dues-book app for small shops in Bangladesh. It has a free plan and works offline. When we wrote our outreach plan on 21 October, Search Console showed a single external link to tukri.bd.

So we took a list of 50 directories and launch sites and tried to get Tukri listed on every one of them, on free plans only.

This is the scoreboard from our own sign-up log, and what the numbers hide. The counts are as of 30 October 2026. We have no traffic or ranking results yet, so none are claimed here.

The scoreboard

What happened	Sites
A public Tukri page exists	6
Submitted, waiting in a review queue (not live)	13
Account or draft made, one step left	8
Stopped at a CAPTCHA, bot check or sign-in loop	9
Would not save (G2, plus the three sites it feeds)	4
Free only if a badge or backlink goes on our homepage	5
Paid only	4
Not a fit	1

That adds up to 50. The six live pages are Crunchbase, F6S, Peerlist, dev.to, Hashnode and GitLab. Two of them are thin: GitLab is only a profile, and Peerlist is a project page we have not launched. All six used Google sign-in.

What the numbers hide

Submitted is not live. PeerPush put us in a free queue of about three months. Uneed scheduled our free launch for 4 May 2027. whatlaunched.today gave us 27 January 2027. One small directory's own page says more than 82% of free listings there are rejected.

"Free" often means "free if you link to us". Five directories list a product for free only after their badge or a backlink is live on our homepage or footer. Four more are paid only. We bought nothing, and we held back on the badges: what goes in our footer is a product decision, not an outreach task.

Some listings ask for things we do not have.

One review platform wants a personal work address (not a shared info@ one), at least 100 customers and a LinkedIn page with 10 or more employees. We have none of those yet, so we did not apply.
One AI-tool directory wants a tick that the product is an AI tool and not a generic service. Tukri is a POS with an AI assistant inside it, and we could not honestly tick that.
Two developer registries want a public repository or a public MCP endpoint. Tukri's MCP server only works with a shop's own token, so there is nothing safe to list. RapidAPI got an account and nothing else, because Tukri has no public API.

Humans still hold the gate. Nine sign-ups stopped at a CAPTCHA, a bot wall or a sign-in loop, and a person has to pass the first two. We also stopped on purpose at boxes that declare something personal, such as "I am 16 or older" or "I declare this information is truthful". Those are a person's to tick.

Two sites failed for reasons we cannot see. G2's Add Product form took every field and returned a generic "Something went wrong while saving your product" on six attempts. It also feeds Capterra, GetApp and Software Advice, so all four are stuck behind it. StartupBlink refused our submission with "Your account email does not use the tukri.bd domain", although the account email is on that domain. We do not know the cause of either. The next step is their support teams.

What we refused to do

Ask for votes. One site's free launch expects ten upvotes on launch day. We will not ask anyone to vote.
Ask for reviews from people who have not used the product.
Keep numbers that are not ours. One directory had already scraped Tukri and showed a monthly-revenue badge and a starting price that we never reported. We corrected the listing.
Show Bangla script on English pages. Our own repo screenshots contain Bangla product names and the taka sign, so English listings got screenshots without them, and prices are written as BDT.

What we would do differently

Start with ten sites, not fifty. The other 44 ended in queues, puzzles, paywalls, or a decision about our own footer.
Read the terms before making an account. The badge, backlink and price rules are on the submit page and decide whether a site is worth it.
Log every site with its next step. A table with one status per site and the single thing left to do is what let us pick the work back up after every pause.

What comes next

We will watch the referrers in Cloudflare Web Analytics and the Links report in Search Console, and we will not claim a result until we see one.

If you run a directory with a free route that does not need a badge, tell us in the comments.

Tukri is at tukri.bd: a free plan, Standard at BDT 250 a month, Premium at BDT 400 a month, and a 14-day trial on the paid plans.

Disclosure: we are the Tukri team. This post was written by an AI agent from our own sign-up log and published by the Tukri team. The counts come from that log.

How to choose a POS and inventory app for a small shop in Bangladesh: a 12-point checklist

Tukri — Thu, 29 Oct 2026 14:51:20 +0000

Most "best POS" lists are written for shops in other countries. A grocery, pharmacy, garments or electronics shop in Bangladesh has its own needs:

Bangla on screen and on the printed bill;
customers who buy on credit (baki);
internet that drops at the worst moment;
an owner who runs everything from one phone.

Use these twelve questions to compare any app before you trust it with your shop.

The checklist

Is it fully in Bangla, including invoices? A Bangla menu isn't enough. Print a test bill with your shop name and a Bangla product name, and check the letters join correctly.
Does it keep working without internet? Ask specifically: can I bill and record a due while offline, and does it sync by itself afterwards?
Does it run on the phone you already have? You shouldn't need a special terminal. A barcode scanner or thermal printer should be optional.
Can it scan barcodes, including with the phone camera? Look for search billing too, for loose items without a barcode.
Does it have a real dues book (bakir khata)? You want a ledger per customer, a running balance, part payments, and a way to send the customer their statement.
Will it warn you before stock runs out? Low-stock alerts, and for pharmacies, batch and expiry tracking.
Can it handle your product shape? Garments need variants (size and colour); a grocery sells by weight; a wholesaler buys by the box and sells by the piece.
Can staff use it without seeing everything? Roles should let a cashier bill without seeing your profit.
Does it print the bills you actually use? A4 for some customers, 80 mm thermal at the counter, and a shareable digital receipt (SMS or WhatsApp).
Will it tell you your real profit? You want daily profit and loss, top products and stock value. If you need it, check for a VAT report.
Is your data yours? Can you export products, customers and records to CSV at any time? Can you delete your account?
What does it really cost as you grow? Read the limits of the free plan (users, products, bills a month), the yearly price, and which payment methods it takes (bKash, cards).

A comparison table you can copy

Question	App A	App B	App C
Bangla on invoices
Works offline (bill + dues)
Phone only, no extra hardware
Barcode / camera scan
Dues book with statements
Low-stock and expiry alerts
Variants, loose, packs
Staff roles
A4 + thermal + digital receipt
Profit and loss, VAT
CSV export
Free-plan limits / yearly price

Test the apps on a real day's work: ring up a few bills, record a part payment, switch the phone to airplane mode and bill again, print one invoice, then export your data. An app that passes that test is worth paying for.

This article comes from the Tukri team. Tukri is a Bangla POS, stock and dues-book app for small shops in Bangladesh: https://tukri.bd/.

DEV Community: Tukri