The consumption rate of an item is the average daily quantity used, across both sales and non-sales loss. It is one of the most important inputs to replenishment math: PAR, reorder point, days of stock, and safety stock all derive from it.
Quick answers
What is the consumption rate? Consumption rate is how fast an item is used per day, combining direct sales (units sold) and non-sales loss (spoilage, shrinkage, breakage, internal use). It is measured in units per day per item.
How do you calculate the rate of consumption? For an item sold directly: consumption rate = sum of daily sales over last 30 days / 30. For an ingredient that goes into recipes: consumption rate = Σ (recipe sales rate × recipe yield for this ingredient). Add a decay factor for perishables.
What's the consumption rate formula for a restaurant ingredient? ingredient consumption rate = Σ (recipe sales rate × recipe yield). If a sandwich uses 100g of bread and you sell 30 sandwiches/day, the bread consumption from that recipe is 3 kg/day. Sum across all recipes that use the ingredient, then add decay.
Why use a 30-day rolling window? Too short (7 days) is noisy and over-reacts to weekly spikes. Too long (90+ days) under-reacts to trend changes and is stale through season changes. A 30-day window is a practical starting point: long enough for stable means, short enough to catch trend.
Components of consumption
consumption rate = sales rate + decay rate × on-hand inventory
or, more practically:
- Sales rate — units sold per day, ideally pulled from your POS
- Decay component — daily loss to spoilage, shrinkage, breakage, internal use; see decay rate
How sales rate is computed
For an item sold directly:
sales rate = sum(daily sales over last 30 days) / 30
For an ingredient that goes into recipes (e.g. flour into bread):
ingredient sales rate = Σ (recipe sales rate × recipe yield for this ingredient)
If a sandwich uses 100g of bread and you sell 30 sandwiches/day, the implied bread consumption from that recipe is 3 kg/day. Sum across all recipes that consume the ingredient, then add direct sales (if any), then add decay.
The 30-day window
LineNow uses a 30-day rolling window for consumption calculation. This is a deliberate trade-off:
- Too short (7 days): noisy, over-reacts to weekly spikes
- Too long (90+ days): under-reacts to trend changes, stale through season changes
- 30 days: long enough for stable means, short enough to catch trend
For items with strong weekly seasonality (a Tuesday-spike vs Saturday-spike pattern), the system also uses day-of-week analysis to adjust short-term forecasts.
Intermittent demand
For items that don't sell every day (a specialty bitter that sells 6 times a month), the simple 30-day average understates true demand because zero-days dilute the mean. The Croston/SBA approach separates demand size from demand interval and forecasts them independently:
SBA forecast = (1 − α/2) × (smoothed demand size / smoothed inter-demand interval)
with α = 0.15 typically. This is the bias-corrected version of Croston's method, published in 2005, and is what LineNow uses for non-smooth demand.
Manual override
For items not connected to POS — toilet paper, cleaning supplies, office consumables — you can set the consumption rate manually. The system will use your value verbatim. You can update it any time.
Why this is the foundation
Consumption rate is the input that determines:
- Days of stock: on-hand / consumption rate (with decay adjustment)
- PAR level: consumption rate × order frequency + safety stock
- Reorder point: consumption rate × lead time + safety stock
- Order recommendation: PAR − on-hand + (consumption rate × lead time)
Get consumption rate right and the rest of the system falls into place. Get it wrong and every downstream calculation drifts. This is why we put POS connection at the very top of onboarding — without it, consumption is gut-feel, and the math is back to gut-feel.
Consumption rate vs sales velocity
Sales velocity and consumption rate are related, but they are not identical.
| Measure | What it counts | Where it breaks if used alone |
|---|---|---|
| Sales velocity | Units sold to customers | Ignores spoilage, shrinkage, and prep loss |
| Consumption rate | Units leaving usable inventory per day | Requires cleaner receiving and count data |
For direct retail products, sales velocity may be close enough. For restaurants, manufacturers, florists, and regulated retailers, consumption rate is usually the better operating signal. A restaurant does not sell flour directly; it sells menu items that consume flour through recipes. A florist does not sell every stem exactly as received; some stems decay, break, or get trimmed. A cannabis retailer may need lot-level expiry and shrink captured separately from sales.
That is why consumption rate belongs in procurement software, not only POS reporting.
How to sanity-check the number
Use three tests before trusting a consumption rate:
- Compare it to receiving cadence. If the item consumes 12 units/day and the supplier comes weekly, a normal order should cover at least 84 units plus buffer.
- Compare it to shelf reality. If the math says 10 days of coverage but the shelf looks empty after four days, shrink, recipe mapping, or sales mapping is wrong.
- Compare it to supplier constraints. If the supplier sells cases of 24, the recommendation needs pack rounding and MOQ handling.
The number is useful only when it can drive a purchase order your supplier can actually fulfill.