Choosing a niche and an item on a platform: demand, competition, the unit economics of an item, the entry point and the pilot batch
How to choose which item to launch on a platform and how much of it to order for a pilot: estimate the capacity of the niche from demand and price, see competition through the number of sellers and the leaders’ share, count the newcomer’s expected sales, the unit contribution, the entry point and the size of the pilot batch. The document turns a choice “by feel” into a table with the decision “go, test or no” and a criterion by which the pilot is checked after a few weeks.
Why this document
Newcomers choose an item because it “sells well” at others, and forget that the turnover of a niche is shared by the leaders and hundreds of sellers, while a newcomer without reviews gets a small part. Then the batch lies, advertising eats the contribution, and the failure is blamed on “the platform”. The head of e-commerce, the platform manager, the category manager and the founder read this document before choosing an item and the first batch. It gives an order: capacity, competition, the unit contribution, a pilot, the decision and the stop criterion. The “Niche and item on a platform” calculation does this for several candidates at once.
At a glance
- What is this document and why is it needed?
- How to choose which item to launch on a platform and how much of it to order for a pilot: estimate the capacity of the niche from demand and price, see competition through the number of sellers and the leaders’ share, count the newcomer’s expected sales, the unit contribution, the entry point and the size of the pilot batch. The document turns a choice “by feel” into a table with the decision “go, test or no” and a criterion by which the pilot is checked after a few weeks.
- Who owns the document and how often is it rebuilt?
- The head of e-commerce or the platform manager prepares, the category manager and the CFO approve the size of the pilot and the stop threshold; counted before every new item and reviewed after the pilot and with a change of season.
- What mistakes are the most common?
- Brand: takes the niche capacity as its own sales → orders a batch for a year → the goods lie, advertising eats the contribution → count the newcomer’s sales by the share of an average seller and take a pilot for one or two months. Retail: does not look at the leaders’ share and the number of sellers → enters niches where little is left → compare sellers per 1,000 units of demand and the leaders’ share before ordering. Brand: does not count the fixed costs of an item → the unit contribution seems enough, while at small sales it does not cover them → count the entry point and compare with the expected sales.
For whom
- Brand
- A brand chooses which model to bring to the platform first and in what run: its own cost and price, trademark rights, the possibility of a re-order and the speed of production matter; the pilot protects against a batch that “nobody noticed”.
- Retail
- A retailer or reseller chooses what to buy from a supplier for a platform: turnover, the possibility of returning the remainder to the supplier, the purchase price and competitors with the same goods matter; the pilot lowers the risk of stale wholesale stock.
What it consists of
Three questions about a niche
- Capacity: how many units and how much money a month sell in a narrow group of goods (demand a month × price). This is the ceiling of the niche, not your sales; the data come from the platform’s analytics and are averaged over several months allowing for the season.
- Competition: how many sellers have sales, what share the ten leaders take and how many sellers there are per 1,000 units of demand. The higher the leaders’ share and the more sellers, the less is left to a newcomer.
- Entry: how you will stand out to get at least the share of an average seller: a price in the core of the niche, reviews, photos and a card, a unique style, the brand, sizes in stock. Without a reason it is not worth entering a niche with high concentration.
The unit economics of an item and the entry point
- Contribution per unit: price without VAT minus cost, commission, logistics, storage, returns and advertising. The exact holdbacks come from the “Marketplace unit economics” calculation; in the niche calculation they are set as one share of the price.
- The entry point: how many units a month must be sold for the unit contribution to cover the fixed costs of an item (photos, content, subscriptions, manager time). The expected sales must exceed it with a margin.
- Brand: count the contribution at your price and your cost and check whether the price will hold the platform’s discount; compare the platform’s warehouse scheme and your own warehouse separately, it changes the holdbacks.
- Retail: take into account the purchase price from the supplier and the terms of returning the remainder: if the unsold cannot be returned, allow a larger reserve for the remainder and take a smaller pilot.
The pilot and the decision criteria
- The pilot batch: the expected sales for one or two months, rounded up to the batch step; the investment equals the batch at cost plus the advertising test; the profit of the pilot is counted at a full sell-out.
- The decision “go”: the contribution is positive, the return of the pilot is not below the threshold, the margin over the entry point is not below the threshold, the leaders’ share is not above the maximum. “Test with a small volume”: the contribution is positive but the margin is between one and the threshold. “No”: otherwise.
- The stop criterion is set before the start: how many units must sell by the control week (see the document on collection gates); without it the pilot is “prolonged” and money is lost on hope.
Where the data comes from
| Data source | Fields |
|---|---|
| The platform’s analytics | demand a month in the niche, the number of sellers with sales, the share of the ten leaders, core prices, seasonality |
| Production or supplier | cost in the warehouse without VAT, the batch step, the term, the possibility of a re-order and of returning the remainder |
| The platform’s offer and the advertising plan | holdbacks (commission, logistics, storage, returns), the test budget, the fixed costs of an item |
How to build it step by step
- Step 1: Brand: make a list of three to five candidates: models you can make, with the price of the niche core; compare several, not one favorite.
- Step 2: Retail: for each niche collect the demand a month, the number of sellers with sales and the share of the ten leaders from the platform’s analytics; average over several months and mark the seasonal peaks.
- Step 3: Brand: count the unit contribution at your price and cost; if it is not positive, the niche drops out before any demand calculation.
- Step 4: Retail: set the criteria in advance: the minimum return of the pilot, the margin over the entry point, the maximum leaders’ share; write them down before looking at the result.
- Step 5: Brand: run the niche calculation and sort by the return of the pilot; look separately at the expected sales against the entry point and the time to sell out the pilot.
- Step 6: Retail: for the niches “go” and “test” check whether you have a special reason to enter (price, brand, photos, uniqueness) and whether the card will withstand the leaders’ competition.
- Step 7: Brand: order the pilot batch in the size of the calculation, no more; record the control week and the stop threshold in numbers.
- Step 8: Retail: at the control week compare the fact with the expected sales, replace the assumptions with the fact, recalculate and decide: a re-order, improving the card or a stop.
Formulas
| Measure | Formulas | Note |
|---|---|---|
| Capacity and the share of an average seller | Capacity = Q × P; the share of an average seller = (1 − L) ÷ (N − 10) | Q: demand a month, P: price, L: the share of the ten leaders, N: the number of sellers with sales; at N of 10 or less one place remains. |
| Newcomer’s expected sales | S = Q × the share of an average seller × k | k: the newcomer coefficient to an average seller; without reviews usually below one. |
| Unit contribution and the entry point | u = P ÷ (1 + t) − c − h × P; E = F ÷ u | c: cost, h: the share of the platform and advertising, F: the fixed costs of an item per month; S ≥ E with a margin is needed. |
| The pilot and its return | B = ⌈S × m ÷ step⌉ × step; ROI = (B × u − A) ÷ (B × c + A) | m: months of sales in the pilot, A: the advertising test budget; the profit is counted at a full sell-out of the pilot. |
Formulas in notation
- The share of an average seller outside the leaders
s = (1 − L) / (N − 10), S = Q·s·k
L: the share of the ten leaders, N: sellers with sales, Q: demand a month, k: the newcomer coefficient; S: the newcomer’s expected sales.- The entry point and the margin
E = F / u, margin = S / E
F: the fixed costs of an item a month, u: the unit contribution; going is possible at a margin not below the set threshold.
Worked example
Four illustrative niches: price with 22% VAT, holdbacks and advertising 35% of the price, newcomer coefficient 60%, a pilot for 2 months, an advertising test of RUB 40,000; all values conditional
| Niche | Capacity a month, m RUB | Expected, units a month | Unit contribution, RUB | Batch, units | Return of the pilot, % | Margin over the entry point, times | Decision |
|---|---|---|---|---|---|---|---|
| Cotton shirt dress | 68.6 | 72 | 851 | 150 | 34 | 2.0 | go: order the pilot batch |
| Palazzo trousers | 85.8 | 47 | 682 | 100 | 18 | 1.1 | test with a small volume and check |
| Linen suit | 46.8 | 65 | 1,082 | 130 | 30 | 2.3 | go: order the pilot batch |
| Basic oversize hoodie | 158.4 | 4 | 650 | 10 | −68 | 0.1 | no: do not enter |
The shirt dress: a capacity of RUB 68.6m a month, but a newcomer gets about 72 units; at a contribution of RUB 851 a pilot of 150 units invests RUB 257,500 and gives RUB 87,709 of profit (a return of 34%). The hoodie with its huge demand drops out: the leaders hold 71% and there are about two thousand sellers. The numbers are illustrative.
Figures
- positive
- negative
- total
Data as a table
| Step | Change | Level |
|---|---|---|
| Shelf price | +4,900 | 4,900 |
| VAT | −884 | 4,016 |
| Cost | −1,450 | 2,566 |
| Platform and advertising | −1,715 | 851 |
| Unit contribution | — | 851 |
The unit contribution of the shirt dress: from the shelf price to profit. Of RUB 4,900 on the shelf, VAT, the cost and the platform’s holdbacks with advertising leave RUB 851: this is how much is earned on every unit.
Data as a table
| Measure | Return of the pilot |
|---|---|
| Cotton shirt dress | 34 |
| Palazzo trousers | 18 |
| Linen suit | 30 |
| Basic oversize hoodie | −68 |
The return of the pilot by niche. The decision is made by the return together with the margin over the entry point and the leaders’ share: the hoodie is negative at high concentration and a small newcomer share.
Data as a table
| The share of the ten leaders, % | Expected, units a month | |
|---|---|---|
| 40 | 72 | Cotton shirt dress |
| 50 | 47 | Palazzo trousers |
| 50 | 65 | Linen suit |
| 71 | 4 | Basic oversize hoodie |
The leaders’ share and the newcomer’s expected sales. The higher the share of the ten leaders, the less is left to a newcomer at the same demand; a large niche with high concentration may give less than a small one with an even market.
Specifics by category
- Apparel
- Apparel: demand is seasonal, the size run is wide, size returns are noticeable; take demand in and out of season and include returns in the holdbacks; the pilot is usually across a few sizes.
- Footwear
- Footwear: returns by size and last are higher, the unit price is larger, the pilot costs more and there are more sizes; allow a larger returns percentage and a lower newcomer coefficient, marking is mandatory.
- Accessories
- Accessories: a low unit price and impulse demand, fewer returns; the fixed costs of an item matter more, the entry point in units is higher, so a larger sales volume is needed.
- Kids
- Children’s: the school and holiday seasons, quick growth of the child and certification; a reduced VAT rate is possible by a list and changes the price without VAT; make the pilot on the popular sizes.
- Men
- Men’s: demand is steadier, concentration at the leaders is often higher, basic items give a stable turnover; a special reason to enter is needed (fabric, fit, price), otherwise a small share remains.
- Women
- Women’s: there are many niches, more sellers, fashion changes quickly; the pilot batch is small, the time to sell out is shorter and the stop decision is taken earlier so as not to slip into discounts.
Common mistakes
- Brand: takes the niche capacity as its own sales → orders a batch for a year → the goods lie, advertising eats the contribution → count the newcomer’s sales by the share of an average seller and take a pilot for one or two months.
- Retail: does not look at the leaders’ share and the number of sellers → enters niches where little is left → compare sellers per 1,000 units of demand and the leaders’ share before ordering.
- Brand: does not count the fixed costs of an item → the unit contribution seems enough, while at small sales it does not cover them → count the entry point and compare with the expected sales.
- Retail: sets the price by the cheapest competitors → the contribution goes negative → take the price of the niche core and check whether the card will hold this price.
- Brand: does not set the stop threshold before the start → the pilot is “prolonged” against the data → record the control week and the threshold in units before the first sale.
Russian specifics
- Brand: on a platform the seller usually works under an offer and a contract: commissions, logistics, storage and penalties change and are not named by numbers in Lab; in the niche calculation they are set as one illustrative share of the price, put your own from the offer and the reports.
- Retail: the resale of goods is limited by trademark rights and contracts with suppliers; imported and sold light-industry goods need conformity confirmation under the technical regulation and marking before turnover. Rule: CU Commission decision 876 of 09.12.2011: TR CU 017/2011 on safety of light-industry products (Customs Union Commission Decision of 09.12.2011 No. 876).
- Brand: the law on the platform economy (31.07.2025 No. 289-FZ) sets the rules of the relations between a seller and a platform, including notice of changed terms: check the contract and the terms when going to a platform. Rule: Law 289-FZ of 31.07.2025: Federal Law of 31.07.2025 No. 289-FZ on the platform economy.
- Retail: the platform’s analytics (demand, the number of sellers, the leaders’ share) is given under the rules and tariffs of the platform itself and is sometimes incomplete; the data are treated as an estimate and not a fact and are revised after the pilot.