Target audience segmentation and segment size: TAM, SAM and SOM with honest assumptions
Segment size answers the question “are there enough buyers for this brand and this batch”. TAM shows the whole category, SAM the part of it available to your price and channels, SOM what can really be taken in a year; the last is counted two ways, top-down by shares and bottom-up from the budget, and the smaller is taken. The document shows how to split the audience into segments, what each share consists of and where the calculation lies if the shares are taken by eye.
Why this document
Three decisions depend on the segment size: how many units to produce in the first batch, how much money can be spent on acquisition and whether to enter the niche at all. An error of several times is expensive: an overstated SOM gives a warehouse for two years, an understated one an empty shelf in the season. The document is read by the founder, the CFO and an investor; a retailer uses the same method to assess how many of the store’s buyers a new category or an own label gives.
At a glance
- What is this document and why is it needed?
- Segment size answers the question “are there enough buyers for this brand and this batch”. TAM shows the whole category, SAM the part of it available to your price and channels, SOM what can really be taken in a year; the last is counted two ways, top-down by shares and bottom-up from the budget, and the smaller is taken. The document shows how to split the audience into segments, what each share consists of and where the calculation lies if the shares are taken by eye.
- Who owns the document and how often is it rebuilt?
- Kept by the brand manager or an analyst; recalculated when the price, channels or budget change and once a season; the result goes to the founder and the CFO together with the assumptions.
- What mistakes are the most common?
- Brand: Take TAM as the target: the whole category is unreachable and the budget is not counted for it; plan by SOM. Brand: Set the share of the price segment by eye: it is the one that most often errs by several times; check it with a survey of 100–200 people. Brand: Not reconciling the top-down and bottom-up estimates: a gap of several times means an unreal capture share or an understated cost of the first order.
For whom
- Brand
- For a brand the segment size determines the first batch and the launch budget: the units are counted from SOM, the segments show whom to start with.
- Retail
- For a retailer the same calculation assesses the potential of a category or an own label among the store’s buyers and checks a brand’s order against how much it will really sell on this shelf.
What it consists of
How to split the audience into segments
- Brand: Geography: a million-plus city, another city, a delivery region; the shares of available channels and the cost of the first order depend on it.
- Brand: Age and life situation: 25–34 and 35–44 buy differently and in different channels; do not glue them into one row.
- Brand: Occasion and price level: work, every day, an event; the readiness to pay your price level sets the share of the price segment.
- Brand: No more than three to five segments: every row needs its own number of people and its own share, and a new brand has no data for ten rows.
- Retailer: The store’s segments are taken from CRM: age, receipt, frequency, categories; this is more precise than open statistics because the buyers are already yours.
What TAM, SAM and SOM consist of
- TAM: the segment’s people × the share buying the category × units a year × the average price of a unit in the market; this is a ceiling, not a sales target.
- SAM: buyers × the share of your price segment × the share reachable by your regions and channels × units a year × your price.
- SOM top-down: SAM × the capture share for the year; for a new brand these are fractions of a percent, not tens.
- SOM bottom-up: the budget ÷ the cost of the first order gives clients; clients × orders × units × price × (1 − returns) give revenue.
- The calculated SOM: the smaller of the two; a gap of more than a half means one of the estimates is unreal.
How to check the assumptions
- Brand: The share of the price segment is checked with a survey of 100–200 people: show the item with the price and ask about the readiness to buy; “maybe” does not count as “yes”.
- Brand: The cost of the first order is checked with a pilot on a small budget in two channels: real orders are taken, not clicks.
- Retailer: The capture share is checked against the store’s sales of a similar category last season and against how new brands sold in their first months.
- Retailer: The availability of channels is checked by where the store’s buyers buy: offline, site, platform; what is unavailable does not go into SAM.
Conclusions from the calculation
- Demand in units equals SOM divided by the shelf price; it is compared with the first batch: a batch above a year’s demand is an excess stock.
- Brand: The core segment is chosen by SAM per unit of spending: where there are the most reachable buyers and the cost of the first order is lower.
- Retailer: If a brand’s SOM is less than what the buyer plans to order, the order is cut or divided into a trial part and a repeat order.
Where the data comes from
| Data source | Fields |
|---|---|
| Open statistics and CRM data | The number of people by age and city, the structure of category purchases, units a year per buyer. |
| Survey and pilot campaigns | The share ready to pay your price, the cost of the first order, the number of orders per client. |
| Plan data | The shelf price, VAT, the acquisition budget, returns, the size of the first batch. |
How to build it step by step
- Step 1: Brand: choose three audience segments by geography and age and write down the number of people in each from open statistics or CRM.
- Step 2: Brand: for each segment set the share buying the category and the number of units a year; write the source of the figure next to it.
- Step 3: Count TAM at the average market price and make sure it agrees by order of magnitude with what you know about the category.
- Step 4: Brand: set the share of your price segment and the share of available channels; check both with a survey or a pilot.
- Step 5: Count SAM at your own price and show its share in TAM: a few percent is normal for a label.
- Step 6: Retailer: replace open statistics with CRM data and see how SAM changes for the store’s category.
- Step 7: Count SOM top-down by the capture share and then bottom-up from the budget and the cost of the first order.
- Step 8: Reconcile the two estimates: take the smaller one and explain the gap; if the gap is above a half fix the weak assumption.
- Step 9: Convert SOM into units at the shelf price and compare with the first batch; write the decision on the batch together with the assumptions.
Formulas
| Measure | Formulas | Note |
|---|---|---|
| TAM | TAM = Σ people × the share buying × units a year × the average market price | Summed over segments; the price with VAT, as on the shelf. |
| SAM | SAM = Σ buyers × the price-segment share × the reachable share × units a year × your price | The two shares in the middle are the shakiest: check them. |
| SOM top-down | SOM₁ = SAM × the capture share | For a new brand the capture share for a year is fractions of a percent. |
| SOM bottom-up | SOM₂ = budget ÷ cost of the first order × orders × units × price × (1 − returns) | Repeat orders in the first year are not assumed above what is proven. |
| Calculated SOM and demand | SOM = min(SOM₁; SOM₂); demand in units = SOM ÷ shelf price | The batch is compared with the demand: above 100% the stock exceeds a year. |
Formulas in notation
- Calculated SOM
SOM = min(SAM · g ; (B / CAC) · n · u · P · (1 − ρ))
g is the capture share; B the budget; CAC the cost of the first order; n orders per client; u units per order; ρ the return share.
Worked example
Segments, SAM and SOM, an illustrative case
| Segment | Buy the category, people | In your segment and reachable, people | Units a year | SAM, RUB m |
|---|---|---|---|---|
| Million-plus cities, ages 25–34 | 1,080,000 | 48,600 | 121,500 | 889.4 |
| Million-plus cities, ages 35–44 | 884,000 | 35,802 | 89,505 | 655.2 |
| Other cities, ages 25–44 | 840,000 | 5,040 | 10,080 | 73.8 |
| Total over segments | 2,804,000 | 89,442 | 221,085 | 1,618.3 |
TAM for the same segments is RUB 53.84 billion, SAM RUB 1.62 billion (3.0% of TAM). SOM top-down is RUB 8.1 million at a capture share of 0.5%, bottom-up RUB 7.3 million from 828 clients (budget RUB 1.2 million, first order RUB 1,450); we take the smaller: RUB 7.3 million with VAT, about 993 units. A batch of 1,200 units is 121% of the demand. All numbers are illustrative.
Figures
From TAM to SOM: how the circle narrows. Each step multiplies by a share: price, availability, capture; at the output the demand in units is compared with the batch.
- TAM 53.84 bn ₽ → SAM 1.62 bn ₽
- SAM 1.62 bn ₽ → SOM top-down 8.1 m ₽; SOM bottom-up 7.3 m ₽
- SOM top-down 8.1 m ₽ → Demand 993 units
- SOM bottom-up 7.3 m ₽ → Demand 993 units
- Demand 993 units
Data as a table
| Measure | Sum |
|---|---|
| SOM top-down | 8.1 |
| SOM bottom-up | 7.3 |
| Batch at the shelf price | 8.8 |
SOM in two ways and the batch at the shelf price, RUB m. The two estimates are close, the batch at the shelf price is noticeably above the calculated SOM: a stock above a year.
Data as a table
| Measure | SAM |
|---|---|
| Million-plus cities, ages 25–34 | 889 |
| Million-plus cities, ages 35–44 | 655 |
| Other cities, ages 25–44 | 73.8 |
SAM by segment, RUB m a year. The main contribution comes from million-plus cities: so channels and the cost of the first order are counted for them first.
Specifics by category
- Apparel
- In apparel segments are split by age and occasion; the number of units a year per buyer is higher than in footwear. The size policy affects SAM: a narrow size run cuts off part of the buyers and this must be subtracted from the reachable share.
- Footwear
- In footwear there are fewer units a year than in apparel and returns by size are higher, so the bottom-up SOM is counted with a larger return share. Availability depends on the size grid and the possibility of fitting: without it the reachable share is lower.
- Accessories
- In accessories there are more units a year and the average order is lower; the market price of a unit is two to three times below an apparel one. The segment is split by occasion (a gift, everyday, an event), and the share of gift purchases increases seasonality and the role of November and December.
- Kids
- In children’s the count is by parents, not by children: the number of people is the number of families with children of the right age, and units a year are higher because of growth. The share of repeat purchases is high but the item is worn briefly; age groups give separate rows.
- Men
- In the men’s segment the share buying the category is lower, units a year are fewer and the decision is faster and closer to the need; segments are made by occasion and reliability. The reachable share is more often determined by the channel: men more often buy on a platform or in a store nearby.
- Women
- In the women’s segment there are more buyers and more units a year, but also more competitors in each price segment; the share of your price segment matters more than the overall one. Segments are made by occasion and lifestyle, not only by age; seasonality is stronger and this shows in the media plan.
Common mistakes
- Brand: Take TAM as the target: the whole category is unreachable and the budget is not counted for it; plan by SOM.
- Brand: Set the share of the price segment by eye: it is the one that most often errs by several times; check it with a survey of 100–200 people.
- Brand: Not reconciling the top-down and bottom-up estimates: a gap of several times means an unreal capture share or an understated cost of the first order.
- Retailer: Take open statistics instead of the store’s CRM data: the share of buyers and reachable ones among your buyers is different.
- Retailer: Count a brand’s SOM by the market rather than by the shelf: the store’s floor limits sales by physical space and traffic.
Russian specifics
- Brand: The shelf price and the price for the calculation are taken with VAT of 22% (the general rate from 01.01.2026), revenue ex-VAT is obtained by division; check the rate for your item with an accountant. Rule: Law 425-FZ of 28.11.2025: Federal Law of 28.11.2025 No. 425-FZ (amendments to parts one and two of the Tax Code; VAT 22%).
- Brand: Open statistics by age and region are taken from official statistical publications and the year of the data is recorded; the figures in the illustrative case are made up.
- Brand: A survey of buyers with personal data requires consent and a clear purpose under Law No. 152-FZ; an anonymous survey without identifiers is simpler. Rule: Law 152-FZ of 27.07.2006: Federal Law of 27.07.2006 No. 152-FZ on personal data.
- Retailer: CRM and loyalty program data are used for segmentation within the clients’ consents; passing segments to a brand is only anonymized and under a contract; check with a lawyer.
- Retailer: In the EAEU countries the structure of statistics and VAT rates differ: for a calculation for Belarus, Kazakhstan, Armenia and Kyrgyzstan take local data and the local rate, check them.