Fazley RabbiMarket size is an estimate of how many potential customers exist for what you sell, and how much they could buy in a given period. For a small business, the honest way to estimate it is bottom-up: count the buyers you could realistically reach, multiply by how often they buy and how much, and present the answer as a range with every assumption written beside it. One impressive number copied from a report or a slide usually tells you very little about whether your shop, product or service will work.
Why one big number can mislead
Many pitches open with a huge figure: "the market for this in Bangladesh is enormous." That may be true, but it hides three problems.
- It is not your market. A national figure for, say, packaged food includes products, cities and price levels you will never touch.
- It looks exact but is not. A number with many digits feels scientific, yet it often rests on guesses several steps back that nobody can check.
- It leads to costly decisions. Plan stock, staff or rent around a number that was never yours and you find out the hard way.
False precision means presenting a rough guess as if it were an exact fact. The answer is not to stop estimating. It is to estimate in a way you can explain and test.
Three layers: total, reachable, winnable
You may see the terms TAM, SAM and SOM. In plain words:
| Layer | The question it answers | Hypothetical example |
|---|---|---|
| Total market | Who could ever need this? | Every pharmacy in the country that stores medicines needing a fridge |
| Reachable market | Who can you actually serve with your product, price and channels? | Pharmacies in Rajshahi division with a medicine fridge and a reason to monitor it |
| Winnable share | Who could you realistically win soon, given competitors and your capacity? | The pharmacies your two-person team can visit, install for and support |
The winnable layer is the one that pays your bills. The other two give context, and they matter more when you are preparing a plan for investors or a bank than in day-to-day decisions.
A bottom-up method, step by step
- Define the customer precisely. Not "shops" but "grocery shops in Natore town that sell packaged snacks and restock every week". A sharp definition makes counting possible.
- Count from evidence. Walk a few streets and count. Use Google Maps, Facebook groups, trade association lists, suppliers and distributors. Note where each count came from.
- Estimate how many could buy. Not everyone you counted fits. Some are too small, some are loyal to a competitor, some do not have the need. Talk to a handful and see how many respond seriously.
- Estimate how often and how much. Purchase frequency and typical order size come from conversations, your own sales records or similar products you already sell.
- Multiply, then write a range. Do the sum three times, as a cautious, a likely and an optimistic version, with the assumptions next to each.
- Check it against reality. Does the likely version make sense next to what existing sellers seem to manage? Would it need more delivery or staff capacity than you have?
Write the assumptions down
The list of assumptions matters more than the final number. For each one, record what you assumed, why, and how you could check it. For example: "We assumed most pharmacies we approach have a medicine fridge. Based on visits to a few in Bogura. Check by visiting more in two other towns." When an assumption turns out wrong, you change one line instead of losing faith in the whole plan.
Plain, rounded language helps too. Show a low and a high figure with the reasons for each, rather than one exact-looking number that invites false confidence.
Sources that work in Bangladesh
- Your own feet. Counting shops, schools, clinics or farms in a sample area, then scaling up carefully, is often more reliable than desk research.
- Suppliers and distributors. They know roughly how many outlets they serve and how often those outlets reorder, and some will share a picture.
- Search and social signals. Google searches and questions in Facebook groups, in Bangla, Banglish and English, show whether people are looking, though not how many will pay.
- Official and association data. Government statistics publications and trade bodies can help with population, households or registered businesses. Check the date and the definitions before relying on them.
- Customer conversations. Real buyers tell you about need, frequency and willingness to pay in a way numbers cannot. My guide to customer interviews explains how to ask without leading.
Common mistakes
- Starting from a national figure and assuming you will take "just a small share".
- Counting people who have the need but no money, access or authority to buy.
- Ignoring seasons, such as Eid, harvest time or exam periods.
- Forgetting your own limits: a market can be large while your delivery, staff or cash decide how much of it you can serve.
- Treating the estimate as finished. It should improve every time you learn something new.
What to do next
Pick one product or service and one area. Write a precise customer definition, count a sample, and fill a three-column table (cautious, likely, optimistic) with your assumptions beside it. Then test the weakest assumption with a few conversations or a small trial, as described in testing a new business idea cheaply. If you would like help turning the estimate into a plan, see the business development service.