Inventory how-to

How to use year over year data to forecast

Year over year data helps you catch seasonal swings that a short sales window misses, like holiday spikes or summer slowdowns.

Short answer: Year over year data helps you catch seasonal swings that a short sales window misses, like holiday spikes or summer slowdowns. You need at least a full year of sales history per variant to use it properly, and ideally around 13 months so you can compare the same period across two cycles. Until you have that much history, lean on your recent 28-day sales trend instead.

Why recent data alone falls short

A 28-day sales window is good for catching current demand, but it will not warn you that a variant sells three times as much in December as it does in July. If you only look at the last 28 days going into a slow month, you will underorder for the spike that follows. Year over year data fixes this by showing you what actually happened at the same point last year.

What you need before you start

You need clean, consistent sales history per SKU or variant, covering at least one full year, ideally 13 months so you can see the start of the comparable period twice. If you have had major assortment changes, new variants, or stockouts last year, flag those periods. A stockout last year will show as low sales, which understates true demand and will undercount your forecast if you take the number at face value.

How to apply it to ordering

Compare the same calendar window from last year to this year to get a seasonal multiplier. Take your current daily sales rate and adjust it by that multiplier for the weeks ahead that fall in a known seasonal period. Then feed that adjusted daily sales rate into your reorder point and order quantity formulas, rather than using a flat recent-average rate.

Step by step

  1. Pull last year's daily or weekly sales for each variant going back at least 12 to 13 months.
  2. Mark any weeks last year with stockouts, promotions, or listing changes that would distort the comparison.
  3. Calculate a seasonal multiplier: this year's recent daily rate divided by last year's rate for the same calendar window, excluding distorted weeks.
  4. Apply that multiplier to your current daily sales rate to get an adjusted rate for the upcoming seasonal period.
  5. Plug the adjusted daily sales rate into reorder point = daily sales x (lead time + safety days).
  6. Recalculate order quantity using the adjusted rate: order qty = daily sales x (lead time + days of cover) - (on hand + on order).
  7. Revisit the multiplier after the season starts and correct it if actual sales come in differently than last year.

Worked example

Last October, a candle variant sold 2 units a day. This October so far it is selling 2.6 units a day, so the seasonal multiplier is 2.6 / 2 = 1.3. Going into November, where last year's daily rate was 4 units a day, you apply the 1.3 multiplier to get an adjusted forecast of 5.2 units a day for this November. With a 14-day lead time and 5 safety days, reorder point = 5.2 x (14 + 5) = 98.8, so you reorder once stock drops to about 99 units.

How SteadyShelf helps

SteadyShelf applies seasonality automatically once a variant has about 13 months of sales history, adjusting the forecast behind the order-by dates and quantities it suggests. Before that threshold it forecasts off the 28-day window, so for new products or recent launches you will need to manually adjust quantities for an expected seasonal spike yourself.

$29/month flat, unlimited products and suppliers, 30-day free trial. Run the numbers now with the free reorder point calculator.

SteadyShelf is launching on the Shopify App Store. Join early access to get the install link first.

Related guides

All inventory how-to guides ยท Inventory glossary

FAQ

What if a product is less than a year old and has no prior year data?

Use a similar, older variant or product as a proxy for seasonality, applying its year over year pattern to your new product's recent sales rate. Otherwise, rely on the 28-day sales trend and adjust manually when you expect a season to hit, then refine the multiplier once you have a full cycle of data.

How do I handle a stockout period that messed up last year's numbers?

Exclude the stockout weeks from your year over year comparison rather than using the low sales figure as-is. If you can estimate what you would have sold based on the weeks before the stockout, use that estimate instead so your multiplier reflects real demand, not a supply gap.

Let SteadyShelf do this math for every SKU

Order-by dates per supplier, seasonal forecasts and draft POs for your Shopify store. $29/month after a 30-day free trial.

Get early access