How to choose a software agency in Dhaka: a checklist

Dhaka has a large and growing number of software companies, from one-person freelancers to big outsourcing firms. That gives you choice, and also makes choosing hard: almost every portfolio looks good, and almost every proposal promises quality and speed.

This checklist covers what actually predicts a good project. Use it whether you are a local business or a company abroad looking for a team in Bangladesh.

1. Can they explain your project back to you?

Before talking about price, a good agency asks questions: who the users are, what job the product must do, what already exists and what success looks like. After the first call, they should be able to explain your project back to you, in plain language, better than you explained it.

Red flag: a detailed quote after one short call, with no questions about users or goals.

2. Do they propose a small first release?

Strong teams push back on scope. They suggest a first version that does one job well, and a plan for what comes after. That protects your budget and gets real users involved early.

Ask: "What would you leave out of the first release, and why?"

Red flag: everything you asked for is in the first release, at a fixed date, with no trade-offs discussed.

3. Will you see working software every week?

The safest projects are the ones where you can see progress constantly: a test link or a test build on your phone, updated every week. That way problems surface in days, not months.

Ask: "When will I first be able to click through something real?"

Red flag: months of work before anything you can try.

4. Who owns the code, designs and accounts?

You should own everything from the first commit: the code repository, design files, domain, hosting and app store accounts. If the relationship ends, you should be able to hand everything to another team without negotiation.

Ask: "Whose name will the repository and hosting accounts be in?"

Red flag: the agency hosts everything under its own accounts and "maintains it for you", with no access for you.

5. How do they make sure it works?

Quality is not something added at the end. Ask how they test: automated tests, testing on real Android and iOS devices, and checking performance on mid-range phones over mobile data, which is how many customers in Bangladesh will use your product.

Ask: "How do you check speed and quality before each release?"

Red flag: "We test everything manually before launch" and nothing more.

6. Do they care about speed and search?

A website that loads slowly or cannot be found on Google is a cost, not an asset. Good web teams build search basics in from the start: fast pages, readable HTML, proper titles and descriptions, and structured data. Ask them to show the speed of a site they built, live, on a phone.

Red flag: "SEO is a separate service after launch."

7. Is communication clear and in writing?

Especially for clients abroad, communication matters more than location. You want a team that writes clear updates, keeps decisions in shared tools, answers in reasonable time and plans calls around your time zone. Dhaka is GMT+6, which overlaps well with the Middle East, Europe's mornings and Asia-Pacific afternoons.

Ask: "What will a normal week of updates look like?"

Red flag: updates only when you ask, or only by phone.

8. Are the people in the sales call the people who build?

In some companies, senior people win the project and junior people you never met build it. That is not always bad, but you should know who will work on your product and talk to them before signing.

Ask: "Who exactly will design and build this, and can I meet them?"

9. Can they show real, checkable work?

Look for work you can actually use or verify: a live product, an app you can install, a case study with concrete decisions and trade-offs. Be cautious with portfolios of screenshots only, or results that cannot be checked.

Ask: "Can I try something you built, and can I speak to someone you built it for?"

10. What happens after launch?

Launch is the start, not the end. Ask how bugs are handled, how fast they respond, how improvements are planned and how a handover to your own team would work.

Red flag: no plan, or a support contract you only learn about after launch.

Making the final choice

Shortlist two or three teams, give them the same short brief, and compare how they think, not just what they charge. The cheapest proposal is rarely the cheapest project: unclear scope, missing pieces and rework cost far more than a fair price for a team that asks the right questions.

If you want a second opinion on your project or a proposal you have received, book a call. We will tell you honestly what we would do, even if that means a smaller project than you expected.