Facebook Ad Cost in Bangladesh: Budget in BDT

Facebook ad cost in Bangladesh is not a single taka figure, and anyone who sells you one is smoothing over an auction. Meta does not publish a national CPM for Dhaka or Rajshahi. What you can read are guides from people who run accounts, and they do not match. This page lines those ranges up, separates the money you pay Meta from the money you pay a manager, and shows how to set a first budget without copying someone else’s ROAS.

Why the screenshots of “average CPC” disagree

Meta sells each impression in an auction. The price moves with the audience, the season, the placement and how often people hide the ad. A broad Bangladesh audience and a tiny retargeting pool are not the same product. Guides still help, if you keep the label on them.

IH Tahim’s mid-2026 budget note says the CPM ranges are what that author sees in accounts, not Meta’s blended numbers. Broad cold audiences ৳40–90, interest-targeted cold ৳60–150, lookalikes ৳90–180, retargeting ৳120–250. The same page gives restaurant and cafe figures as ৳20–80 per offline visit, with a note to optimise for offline conversions. That is one practitioner’s band.

Kuiperz’s 2025 Bangladesh benchmark calls its CPC range a rough guide: often ৳5–৳20, and says some industries sit outside it. Their industry table is marked as illustrative. Do not add their rows to Tahim’s rows and divide by two. That would invent a third number nobody measured.

Digital Marketing Dhaka’s 2026 cost page says Meta CPM in Bangladesh averages ৳60–180 depending on audience, placement and creative, with broad audiences at ৳60–110 and cold CPC at ৳4–14. “Average” on a vendor’s blog is not an audit. It is still a public claim you can open and argue with.

If your first week is cheaper than all three, do not scale on that luck. If it is more expensive, do not assume the guide is the law. Look at the result type. A message, a link click and a purchase are different auctions.

Spend and fee are different bills

A sentence that still gets lost in Bangladesh quotes: ৳15,000 a month can mean the fee, the spend, or a blend that hides both. Ask.

Marketivity’s Meta Ads management tiers in taka are Starter ৳8,000, Growth ৳15,000 and Scale from ৳25,000 a month. The ad budget is paid by you to Meta, from an ad account in your name. The pricing page also shows those tiers in dollars and pounds, as set prices, not as a conversion of the taka. Starter on the dollar card is a one-page, one-offer scope. It is not “unlimited campaigns until the money feels gone”.

A reasonable first test is a spend you can describe out loud. For a local shop, that might be a few hundred taka a day for two weeks, aimed at messages or profile visits you will actually answer. We are not going to turn that sentence into “the correct daily budget is ৳X”, because the guides above already show why that sentence fails. If you cannot answer the messages, do not buy messages.

Eid, Pohela Boishakh and other crowded weeks cost more because more advertisers show up. Digital Marketing Dhaka and others say this in different words. Plan the calendar. Do not discover it on the morning of the offer.

Boost versus Ads Manager, in taka

Boosting a post spends the same kind of money. It does not give you the same controls. Meta’s difference between boosted posts and ads applies in every country the tools exist: a boost starts from an existing post and has fewer targeting and creative options. Use it for a one-off if the post is already clear. Use Ads Manager when you need a radius around a Rajshahi or Dhaka shop, a result you can name, and a report that separates one photo from another.

Do not target by ethnicity. Meta does not offer that, and proxies are a bad idea and a policy risk. Location, language of the ad, and people who already interacted with the page are the tools you actually have.

How to hold a month of spend

Write the result on a piece of paper before you pay.

• Messages you can reply to the same day.
• Calls.
• Visits to a product page.
• A purchase you can see in a shop system, not only in Ads Manager.

Then refuse every other number as the score for that month. Click-through rate can explain a bad photo. It is not the goal if you asked for messages.

Meta says an ad set usually leaves the learning phase after about 50 results in the week after the last big edit. A small Bangladeshi budget often will not get 50 purchases a week. Say that in the brief so nobody panics on day four and edits the campaign into a new learning phase. Change the photo or the budget, then wait.

At the end of the month, export Ads Manager and, if the client counted sales outside it, keep that count in a second column. The only public sales line we use is client-reported: about 17–20 sales per messaging campaign on average, on campaigns where the client counted the sales. It is not a ROAS. It will not be your number unless your offer, your reply speed and your product are similar, and even then it is not a quote.

A budget you can defend

Try this order.

1. Pick one offer and one result. 2. Put the fee and the spend on two lines. Fee: the tier on pricing. Spend: what you will pay Meta. 3. Run it long enough to get a pattern, not a mood. Two weeks is a start. It is not a law. 4. Pause the ad people ignore. Leave the audience alone that day. 5. Only then decide whether next month’s spend goes up.

If the landing page is a Facebook album from 2022, fix that before you blame the CPM. A site with a current price and a WhatsApp button is part of the ad. The Rajshahi website guide is the build side of the same decision.

We will not fill this page with a fake national average. When your account has two weeks of its own numbers, those numbers replace every row in the diagram.

A worked week without a fake CPM

Suppose you choose messages, and you can reply from 10:00 to 19:00 Bangladesh time on the days you are open. Set a spend you will not need to pause on day three because the till was short. Write that amount down as media, then write the management fee under it if someone else is running the account. At the end of seven days, count threads that were real buyers, threads that were price questions you answered, and threads that were noise. The cost per real thread is your number. It will not match Tahim, Kuiperz or Digital Marketing Dhaka exactly. That is the point of doing the week.

If the threads are mostly people outside your delivery area, the radius or the first line of the ad is wrong. Fix the line. Do not “solve” it by doubling the budget. If the threads are people who want a product you no longer sell, the creative is wrong. Pause it. The auction did what you asked. You asked it to find people who would respond to that picture.

Keep the boost button for a post that is already doing well with the followers you have. The moment you need to know which photo, which area and which sentence did the work, you are in Ads Manager, and the cost conversation above is the one that applies.