Store sales and receipts: traffic, conversion, ticket
What a store’s sales were made of: traffic, conversion, units per receipt and price per unit; plan versus actual by point with a pace forecast; the structure of receipts and the role of discounts in receipts.
About the topic
Store sales look like a single number but consist of four. How many people came in (traffic), how many of them bought (conversion), how many units they took on average (UPT, units per transaction) and at what average price (AUR, average unit retail). Revenue is the product: traffic x conversion x UPT x AUR. The topic teaches you to break a change in sales into these factors and see the cause, not only the consequence.
Example. Traffic of 10,000 people and conversion of 9% give 900 receipts, UPT 2 gives 1,800 units, AUR RUB 3,500 gives revenue of RUB 6.3 million. Next month traffic fell to 9,000 but conversion rose to 10%: the same 900 receipts, revenue unchanged. By revenue alone the store is stable, while the factors show fewer people and everything resting on staff and the till. Another month: a promotion. AUR fell from 3,500 to RUB 3,000, UPT rose from 2 to 2.2: revenue 900 x 2.2 x 3,000 = RUB 5.94 million, less than the earlier 6.3. The promotion brought more units but did not repay the price cut.
The second tool is pace. Say the month plan is RUB 12 million over 30 days, that is RUB 0.4 million a day. By day ten the plan expects RUB 4.0 million, but actual is 3.6 million, 90% of pace. If nothing changes, the month ends at 10.8 million, 1.2 million short. To reach plan, the remaining twenty days need RUB 0.42 million a day. A calculation like this on day ten is more useful than a review at month end.
Receipt structure shows what revenue rests on: many small receipts or few large ones; the share of discounted receipts and the depth of the discount show the price of promotions. If 35% of receipts carry a 25% discount, it costs more than the overall percentage suggests.
Start with definitions: what counts as a receipt (with or without returns), how traffic is counted (a door counter) and how a buyer is counted (one person with several receipts). Traps: comparing periods with different numbers of weekend days and different weather, looking at revenue alone, confusing prices with and without VAT.
Key ideas
Revenue is a product of four factors
Factor effects multiply, they do not add up. Say traffic fell 10%, conversion rose 5%, UPT rose 3% and AUR fell 4%. Adding the percentages gives minus 6%, but 0.9 x 1.05 x 1.03 x 0.96 = 0.9345, which is minus 6.5%: from a base of RUB 6.3 million about RUB 5.89 million remains. For small moves the gap is small, for large ones it grows.
The same revenue can hide different causes
Traffic minus 10% and conversion plus 11.1% leave revenue the same. But the store now depends on staff rather than flow. A season later such a store falls first if traffic does not return.
Measure a promotion on margin, not revenue
A promotion must be measured on margin, not revenue. A unit price of RUB 4,000 fell to RUB 3,200 (minus 20%), with a unit cost of RUB 1,600. A 25% rise in UPT holds revenue, but unit margin fell from 2,400 to RUB 1,600, and to keep margin UPT must rise 50%. A promotion that “broke even” on revenue can be well in the red on profit.
Pace matters more than the total
The price of delay grows. With a plan of RUB 18 million a month (0.6 million a day) and a 90% pace on day ten, actual is 5.4 million against 6.0: to reach plan you need 0.63 million a day, 5% above. The same 90% pace on day twenty leaves 7.2 million for ten days: 0.72 million a day, 20% above plan.
The price of a promotion shows in receipts
A discount hits margin harder than price. If 20% of receipts carry a 30% discount, the average price falls 6%. But with a unit cost of 40% of price, unit margin falls from 60 to 54, which is 10%. Count the price of a promotion in margin, and only then decide whether you need it.
The day-of-week index shows where to strengthen and where to pull up
In the example indices Mon–Sun are 0.71, 0.78, 0.74, 0.85, 1.14, 1.49, 1.29: Saturday is almost twice as strong as Monday.
Formula: Index = day average ÷ average over all days
What to do: Exclude holiday weeks.
Lift separates a link from popularity
A pair of A and B in 400 of 10,000 receipts: support 4.0%, confidence 26.7%, lift 2.22: more than twice as often as by chance.
Formula: Lift = support of the pair ÷ (support of A × support of B)
What to do: Do not rank pairs with support below 0.5%.
How to work through it
- Fix definitions: what counts as a receipt, how you count traffic and a buyer.
- Collect four factors for two comparable periods: traffic, conversion, UPT, AUR.
- Break down the revenue change by factor and find the leading one.
- Check plan and fact by store with pace: how much is needed per day to the end of the period.
- Break down receipts by amount band: where small and large are and who generates revenue.
- Calculate the share and depth of discounted receipts and compare a discounted check with a full-price one.
- Decide where to intervene: traffic, conversion, add-on selling or price; and check the result next period.
Where to start
A store: “Sales equation” over two periods, then “Plan versus actual” with the pace. To study the receipt: “Receipt structure” and “Discounted receipts”. Calculations of the topic in order: “Sales equation: traffic, conversion, units per receipt, price per unit”, “Plan versus actual by store: pace, period forecast, required daily pace”, “Receipt structure: ticket bands, average ticket, UPT and contribution to sales”, “Discounted receipts: share, depth and the ticket with a discount versus full price”, “Day-of-week and hour indices for schedules”, “Market basket affinity: support, confidence, lift”.
Calculations in this topic
Sales equation: traffic, conversion, units per receipt, price per unit
Splits the change in a store’s sales between two periods into four causes: how many people entered, what share bought, how many units per receipt and the price per unit. Sales = traffic × conversion × units per receipt (UPT) × price per unit (AUR). Shows what exactly pulls sales up or down.
What exactly changed a store’s sales: traffic, conversion, units per receipt or price per unit.
Plan versus actual by store: pace, period forecast, required daily pace
Compares the actual to date with the plan for the whole period and with the pace-plan for the same number of days, forecasts the end of the period at the current pace and calculates how much must be sold a day to reach the plan. Shows which stores lag behind the pace and by how much.
Are stores on the plan pace, what is forecast for the end of the period and how much must be sold a day.
Receipt structure: ticket bands, average ticket, UPT and contribution to sales
Splits receipts into amount bands: how many receipts and what share of sales fall into each, the average ticket, units per receipt and the price per unit. Shows which receipts the revenue rests on and where the median receipt lies.
Which receipts revenue rests on and where the median receipt lies.
Discounted receipts: share, depth and the ticket with a discount versus full price
Shows what share of receipts goes with a discount, the average depth of the discount in those receipts, how discounts eat into the list price and how a discounted receipt differs from a full-price one, including at list price. Helps see which purchases the discounts rest on.
What share of receipts goes with a discount, how deep the discount is and how a discounted receipt differs.
Day-of-week and hour indices for schedules
Shows how many times sales on each day of the week are above or below the average day: an index of 1.2 means a fifth above the average. Built from the weeks in the table (untick holiday weeks), optionally also by hour. Needed for the schedule of promotions and mailings (days with an index above one) and as an input to the staffing calculation.
How many times each day of the week and hour is stronger or weaker than the average.
Market basket affinity: support, confidence, lift
From the number of receipts with each item and with both at once, counts for pairs of items the support (how often the pair occurs), the confidence (what share of those who bought A also took B) and the lift (how many times more often the pair occurs than with an independent choice). Ranks the pairs by lift among those with enough data and suggests what to do: a bundle, adjacent placement or nothing.
Which pairs of items occur in receipts more often than by chance and what to do about it.
Self-check questions
- Do we count receipt and traffic the same way across stores?
- Which of the four factors changed most?
- Are the periods comparable in weekends, holidays and weather?
- What daily pace is needed to the end of the period?
- What does revenue rest on: small or large receipts?
- How many receipts carry a discount and how deep?
- Is UPT growing at the cost of a lower unit price?
- Are prices in the comparison with or without VAT?
Knowledge check
Multiple-choice questions. The correct answer and the explanation follow each question.
In a day a store sold 1,200 units in 800 receipts. What is UPT and what does it show?
- A. 0.67: how many receipts fall on a unit
- B. 1.5: how many units fall on a receipt
- C. 400: by how many units there are more than receipts
- D. 2,000: the total number of positions
Correct answer: B. 1.5: how many units fall on a receipt. UPT = units / receipts = 1,200 / 800 = 1.5. 0.67 appears if divided the other way round; 400 and 2,000 are the difference and the sum of two quantities of different meaning. Together with the price per unit UPT gives the average receipt: 1.5 × price per unit.
Base: traffic 5,000, conversion 10%, 2 units per receipt, price per unit RUB 2,000 excl. VAT. Now: 4,500, 11%, 1.9 and RUB 2,100. What is the conversion effect in the Lab order (traffic, conversion, units per receipt, price per unit; each factor is taken at the new values of the previous ones and the base values of the following ones)?
- A. −RUB 24,950
- B. +RUB 94,050
- C. +RUB 200,000
- D. +RUB 180,000
Correct answer: D. +RUB 180,000. The conversion effect = 4,500 × (0.11 − 0.10) × 2 × 2,000 = +RUB 180,000. The others: traffic −RUB 200,000, units per receipt −RUB 99,000, price per unit +RUB 94,050, in total −RUB 24,950 (sales RUB 2,000,000 → 1,975,050). +RUB 200,000 appears if the base traffic of 5,000 is taken instead of the new one; +RUB 94,050 is the price-per-unit effect; −RUB 24,950 is the total change of sales, not the contribution of one factor.
The stores’ plan is RUB 9,000,000 for 30 days, 12 days have passed, the actual is RUB 3,240,000. How much must be sold a day until the end of the period to reach the plan?
- A. RUB 270,000
- B. RUB 300,000
- C. RUB 320,000
- D. RUB 480,000
Correct answer: C. RUB 320,000. 30 − 12 = 18 days remain and 9,000,000 − 3,240,000 = RUB 5,760,000: 5,760,000 / 18 = RUB 320,000 a day. RUB 270,000 is the current pace (3,240,000 / 12): at it the period ends at RUB 8.1 million. RUB 300,000 is the planned daily pace (9,000,000 / 30), it ignores the lag behind the pace plan of RUB 3.6 million (−RUB 360,000). RUB 480,000 appears if the shortfall is divided by the days passed (5,760,000 / 12).
10,000 receipts for RUB 18 million excl. VAT. Up to RUB 1,000: 3,500 receipts, RUB 2.1 million; RUB 1,000–2,999: 4,500 receipts, RUB 9.0 million; RUB 3,000 and above: 2,000 receipts, RUB 6.9 million. In which band is the median receipt and what share of sales do the receipts of RUB 3,000 and above give?
- A. The RUB 1,000–2,999 band; 38.3% of sales
- B. The RUB 3,000 and above band; 38.3% of sales
- C. The RUB 1,000–2,999 band; 20% of sales
- D. The up to RUB 1,000 band; 38.3% of sales
Correct answer: A. The RUB 1,000–2,999 band; 38.3% of sales. The cumulative share of receipts is 35% after the first band and 80% after the second, so 50% falls in the RUB 1,000–2,999 band (the average receipt is 18,000,000 / 10,000 = RUB 1,800). Receipts of RUB 3,000 and above are 20% of receipts but 6.9 / 18 = 38.3% of sales. “20% of sales” confuses the share of receipts with the share of sales; the RUB 3,000 and above band is not the median one, it holds only the top fifth of receipts.
In a month there were 5,000 receipts, 2,000 of them with a discount. At list price the receipts with a discount are RUB 6,000,000, the discounts in them are RUB 1,500,000; the receipts without a discount are RUB 9,000,000 at list price. What share of the list price of all sales did discounts take?
- A. 25%
- B. 40%
- C. 16.7%
- D. 10%
Correct answer: D. 10%. The list price of all receipts is 6,000,000 + 9,000,000 = RUB 15,000,000, discounts are RUB 1,500,000: 10%. 25% is the depth of the discount in the discounted receipts only (1.5 / 6), 40% is the share of receipts with a discount, 16.7% appears if discounts are divided by the list price of receipts without a discount (1.5 / 9). Net sales are 15,000,000 − 1,500,000 = RUB 13,500,000.
What does a day-of-week index of 1.2 mean?
- A. 120% higher
- B. Exactly the average
- C. 12% lower
- D. Sales of the day are a fifth above the average day
Correct answer: D. Sales of the day are a fifth above the average day. The index is a ratio to the average day. Covered in: The day-of-week index shows where to strengthen and where to pull up.
What does a lift of a pair above one show?
- A. One item causes the purchase of the other
- B. The pair occurs more often than with an independent choice
- C. The pair is the most expensive
- D. The pair should be removed
Correct answer: B. The pair occurs more often than with an independent choice. Lift shows co-occurrence and not a cause. Covered in: Lift separates a link from popularity.
Common mistakes
- Looking at revenue only. Consequence: a loss of traffic goes unnoticed. Avoid: break down into four factors.
- Comparing non-comparable periods. Consequence: a false conclusion of growth or decline. Avoid: match by number of weekends and holidays.
- Judging a promotion by UPT growth. Consequence: units grow while revenue and margin fall. Avoid: look at AUR, check and margin together.
- Waiting for month end to review. Consequence: too late to fix anything. Avoid: calculate pace on day ten.
- Mixing prices with and without VAT. Consequence: gaps with accounting. Avoid: choose one base and label it.
Mixing holiday and ordinary weeks
What it looks like (illustrative figures): A holiday week distorts the indices.
Taking pairs with little support
What it looks like (illustrative figures): On rare pairs lift is large by chance.
Russia and the EAEU
The basis of retail sales in Russia is the till receipt under 54-FZ: every sale and every return goes through an online till and reaches the tax service. So till data is more reliable than manual reports, and it is the source for receipts, average check and UPT. Define a receipt the way it is rung up: a return receipt is separate, and if it is not subtracted, revenue and conversion are overstated.
The price on a price tag in Russia includes VAT. The general rate from 2026 is 22%, so a RUB 6,500 receipt contains RUB 1,172 of VAT and is RUB 5,328 without tax. For comparison with cost and margin take the price without VAT. On the simplified regime VAT is not charged (exemption below RUB 20 million of revenue, a special procedure above that), so revenue in the report may be higher than under the general regime; check with your accountant.
For marked goods a sale is accompanied by retiring the code at the till, and the code scanner and till must work together: code errors slow the queue and affect conversion. Traffic counter: if it only counts people it is not personal data; if it recognizes faces, 152-FZ applies, so check with a lawyer. Russian holiday peaks (23 February, 8 March, 1 September, New Year, marketplace sales) must be accounted for when comparing periods.
The sales formula is the same. Tills and taxes differ. As of 14.11.2026, VAT rates by country are on the page “EAEU and CIS countries”; check them with an accountant, so the same receipt carries different tax in different countries. Requirements for online tills, fiscal data and electronic invoices are local: indicative as of 14.11.2026, check the text of the act. Holiday peaks differ: Nauryz in Kazakhstan and Kyrgyzstan, national dates in other countries. The receipt currency changes psychological prices: compare checks in local currency, not by conversion.
Categories and audiences
- Apparel
- If it is apparel: store conversion is modest because people browse and try on, UPT of 1.5 to 2 depends on the skill of suggesting an add-on. Sales sharply change AUR. Measure fitting: the share of those who tried on and bought shows assortment quality.
- Footwear
- On footwear: purchase requires fitting and sizes, conversion is higher among purposeful visitors, UPT is close to one, so growth comes through the check and price. Add-ons of care products and socks lift UPT. Size returns reduce net revenue.
- Accessories
- In accessories: a high share of impulse purchases, UPT higher thanks to the till area, average check lower. Conversion depends strongly on display and footfall. Small checks create many transactions, so acquiring fees show in margin.
- Kids
- For children’s wear: the family comes in, traffic is counted by people, but the parent buys, so conversion is better counted by adults. UPT is high: seasonal sets. The pre-school peak gives a traffic spike that cannot be compared with an ordinary month.
- Men
- Men come in with a task: conversion is higher, time shorter, UPT lower unless a set is offered. Check growth comes through add-ons (a shirt with a suit, a belt). Traffic is less sensitive to promotions, and price is accepted faster.
- Women
- For the women’s line, conversion by the door counter is understated: some people come in to look and decide later. It is more useful to count conversion separately for visits with a fitting. The second mechanism is the set on display: top, bottom and an accessory in one look raise UPT, but the added accessory is cheaper and AUR falls. Check revenue against both factors at once.
Questions this topic answers
- What exactly changed sales: traffic, conversion, ticket or price?
- Are stores on the plan pace and what must be sold a day to the end of the period?
- Which receipts does revenue rest on?
- What share of receipts and sales do discounts take?
Key measures and terms
- Conversion
- Average transaction value (AOV)
- Units per transaction (UPT) and average unit retail (AUR)
- Comparable sales (comps)
- Revenue per m² and per employee
- Share of sales on discount and average depth
- Sales equation
- Pace-plan and pace forecast
- Receipt bands
- Discounted receipts: share and depth
- Store report
- Receipt, purchase and visit
- Average check, UPT and AUR: the identity
- Capture rate
Study first
What next
Related: “Markdown and exit price”, “CRM and the customer base: retention, segments, win-back”, “Season results: brand, category, SKU”.
Documents on this topic
- Sales plan
- Unit economics (CM1, CM2, CM3) with buyout
- Wholesale documents: pre-order, ship window, chargebacks, EDI/ASN
- Omnichannel: OMS, ship-from-store, BOPIS and unified stock
- Store KPI pack: traffic, conversion, ticket, revenue per m², payroll and shrinkage
- Men, women, unisex, plus size and adaptive
- Plus size, adaptive clothing and unisex: size grid, patterns, assortment and economics
- Descriptive statistics for retail: center, spread, outliers, intervals
- Unit economics with buyout: CM1-CM3, LTV/CAC, customer survival, channel payback
- Price and volume indices: Laspeyres, Paasche, Fisher, chain indices, deflator and the 4-5-4 calendar
- Bayesian estimates of conversion and buyout: small samples, prior distributions, shrinkage and comparison with the frequentist approach
- Unit economics by platform: how to build the calculation from deductions, compare schemes and find the break-even price
- Reader question: what price to set on a novelty when there is no elasticity
- AI in fashion and analytics: where to apply it, where not, how to check the result, which data are needed and which errors occur. A memo and a checklist
- Site, CRM and receipts: abandoned carts, the delivery threshold, payments, message frequency, the lapsed threshold, day indices and pairs of items