How Moncho sizes a market you can defend
Moncho answers "how big is this market, and can I show my work?" with a named claim, sourced inputs, and a calculation you can inspect. This page covers the trust rules across the product, then the market-sizing method we use today. Value chain, organizations, and product pricing follow the same trust rules. Their full methods will be added here when those chapters are ready. They do not reuse sizing math as a stand-in.
On this page
- Market size: full method below
- Also in the product (same trust rules; methods on this page later): value chain analysis, organizations, product pricing
Shared trust rules
- Prefer named publishers (statistical offices, customs, industry surveys, audited company filings, published price lists).
- A Moncho analyst reviews anything labeled official.
- Show gaps. Missing volume or missing price is a research state, not a license to invent.
- Public platform knowledge and private client work stay separate.
- AI assists discovery and structuring; humans own publication.
- Two honest methods stay two numbers. We do not average incompatible scopes into a fake consensus.
Market size
1. What are you measuring?
Before any formula, we fix the market claim: what economic object, in which country, at which sale point (farm, factory, shop, border), and for which year.
That claim is not always "TAM." It may be:
- Observed current market: what already sells at that transaction point
- Eligible market (TAM): eligible units × a sourced price, after an eligibility rule
- Reachable slice (SAM): TAM under named constraints
- Observable activity around the problem: counts, volumes, or transactions we can name when a full eligible-market size cannot close yet
- Supporting flows: import value, value added, or government spend. Related evidence that is not automatically the whole commercial market
Import value is not the domestic market. GDP value added is not the grocery bill. Capacity is not demand. Unmet need is not paying demand. Value-chain stages must not be summed into one "market."
We size concrete markets we can defend (for example potato at farm gate in Bangladesh, or a named garments export segment). We do not claim a global sizing engine.
2. Where do the numbers come from?
Evidence is ranked by how directly it measures the claim:
- Official statistics and statistical yearbooks
- Customs and trade databases
- Industry association surveys and published sector reports
- Company filings and public list prices
- Published government household or establishment surveys
We record publisher, year, unit, geography, and sale point. We reuse evidence already on Moncho before searching for new sources. Analogues (nearby product, another country) are allowed only as labeled inputs to a working range, never as silent substitutes for a missing home-country price.
Public curated facts on the platform stay separate from private client datasets.
3. How do you build a size?
- A
Define
Scope in and out. Name the unit (tonnes, seats, accounts, households, facilities). Name the sale point.
- B
Choose the economic calculation that fits the evidence
Moncho does not run one universal formula. When public evidence exists, we select one or two fitting families. The families buyers see on market-size cards are in the table below.
- C
Close only when required inputs exist
A TAM-style claim needs eligible units and a sourced price (or another complete family above). If the fee or price half is missing, we search for a real source. We do not multiply a census by a made-up ARPU and call it market size.
- D
Triangulate when two paths are honest
Keep both. Explain scope differences. Do not blend them into one fake midpoint.
- E
Review
A Moncho analyst signs the card (method, sources, year, sale point, limitations) before it is official.
Calculation families on market-size cards
Reported Sales/Revenue
A publisher already printed seller output
Use the published total; do not rebuild from thin air only because we can
Value added
National-accounts contribution
Use GVA / SNA value added as that claim, not as retail spend
Import value
Customs line
Import value for that HS scope; not the whole domestic market
Harvest value
Physical output × harvest price
Tonnes (or quintals) × sourced farm-gate price, same year
What buyers spend
Demand build-up
Eligible buyers × take-up × frequency × sourced price
What accounts pay
Installed base × fee
Seats, wallets, connections, or accounts × annual fee
What capacity earns
Used capacity × tariff
Utilized capacity × sourced tariff (not empty nameplate capacity)
Fees on spending
Take-rate on flow
Gross flow × sourced take rate (the fee pool, not the underlying GMV)
Slice of a larger market
Top-down share
Parent total × a published share (invented percent fails)
Net Domestic Market Size
Apparent supply
Production + imports − exports at a consistent price lens (not the retail grocery bill)
| How we label it | Economic idea | Typical equation shape |
|---|---|---|
| Reported Sales/Revenue | A publisher already printed seller output | Use the published total; do not rebuild from thin air only because we can |
| Value added | National-accounts contribution | Use GVA / SNA value added as that claim, not as retail spend |
| Import value | Customs line | Import value for that HS scope; not the whole domestic market |
| Harvest value | Physical output × harvest price | Tonnes (or quintals) × sourced farm-gate price, same year |
| What buyers spend | Demand build-up | Eligible buyers × take-up × frequency × sourced price |
| What accounts pay | Installed base × fee | Seats, wallets, connections, or accounts × annual fee |
| What capacity earns | Used capacity × tariff | Utilized capacity × sourced tariff (not empty nameplate capacity) |
| Fees on spending | Take-rate on flow | Gross flow × sourced take rate (the fee pool, not the underlying GMV) |
| Slice of a larger market | Top-down share | Parent total × a published share (invented percent fails) |
| Net Domestic Market Size | Apparent supply | Production + imports − exports at a consistent price lens (not the retail grocery bill) |
Worked example (potato, farm gate)
Yearbook harvest tonnes × a sourced farm-gate price bulletin for that crop and year becomes Harvest value. If the price bulletin is missing, the size stays unfinished or a labeled working range, not "tonnes × a guessed taka."
What we do not do
- Census × invented fee: Looks precise, is fiction; destroys trust when a buyer checks the fee.
- Treating one HS import line as the whole market: Not without an exact match or a published share.
- Calling customs value or GDP value added "TAM" by renaming: Wrong claim type dressed as market size.
- Averaging two incompatible scopes into one "consensus" number: Two honest methods stay two numbers.
- Summing every value-chain stage into one market: That double-counts the same activity.
- Primary surveys, interviews, and econometric models we design ourselves as the default beta path: Those stay Enterprise or later when the engagement funds them. Published government surveys remain in scope.
- Multi-year forecast as the product headline: Moncho's focus is a current or historical size you can inspect, not a syndicated outlook.
None of this is withheld as a trade secret. What stays internal is day-to-day research operations, not the economic methods themselves.
4. What is official vs draft vs custom?
- Official: Published market-size card after analyst review. Reports and Sherpa quote this as Moncho's size for that claim.
- Working range / draft: Sourced or analog inputs with named assumptions; or a chat answer from available facts. Labeled. Not sold as official.
- Related evidence: Jobs, trade lines, inflation, or other facts that support understanding but are not the headline size.
- Enterprise / custom: Interviews, primary research, extra geographies, or tighter eligibility under contract. Not automatically everyone's platform number.
5. What if the data is thin?
Thin data is normal in developing-market cells. We name the missing input (volume, price, eligibility, share). Sherpa and the product surface thin coverage instead of inventing a neat total.
A smaller honest card, or a labeled working range, beats a round number with a hollow fee. When a better source arrives, we correct the card.
Coverage is beta and uneven by country and sector. What we cover today.
6. How we use AI (and what is not AI-made)
AI helps find sources, clean tables, propose classifications, and draft calculations from evidence already in hand. It does not invent missing prices, overwrite trusted facts without review, or publish an official size.
Official market sizes and curated market facts on the platform are sourced and reviewed by a Moncho analyst. They are not synthetic fill and not model-made numbers treated as published truth. That matches the sitewide editorial disclosure.
Chat answers and share what-ifs can use AI or labeled scenarios. Those stay labeled. They are not the official card.