What it means
A customer orders a particular brand of milk, the store is out of stock and offers another brand; if the alternative is supplied, the item has been substituted. Instacart documents replacement flows that involve preferences and communication with the customer, and different platforms allow approval, automatic alternatives or refunds, so a retailer should follow the customer's chosen rules.
A fictional grocery service fills 10,000 ordered items and replaces 400, so if the denominator is all originally requested items its completed-item substitution rate is 4%, and the definition should say whether cancelled items are included. Orders and items are different bases, since one large order may contain several replacements and a rate based on orders with any substitution cannot be compared directly with item substitution rate.
A substitution may preserve a basket when the original item is unavailable, but it can also disappoint the buyer if size, ingredients or price differ, so a lower rate is not the only customer goal. Instacart's help material lets customers specify preferred replacements and item instructions; clear preferences reduce guesswork, though a picker still needs to check whether an alternative fits those preferences.
A fictional shopper marks a peanut-free snack as do not replace, and the store does not send a visually similar product that may contain peanuts, because safety and the customer's instruction outweigh preserving the sale. Rules differ across products and markets, since allergens, age restrictions and regulated goods require special care and substitution should never bypass applicable safety or consent requirements.
A fictional pharmacy delivery service distinguishes a customer-approved brand swap from a staff-initiated one and follows applicable dispensing rules rather than treating medicines like ordinary groceries, because the generic metric alone cannot govern that decision. Stock accuracy can reduce avoidable substitutions, since an app that shows unavailable items as in stock creates false expectations, so synchronise availability with picking and replenishment where possible.
Some substitutions are requested by customers after ordering and others are proposed by the merchant, so segment them if the rate is meant to diagnose stockouts, and track substitutions by category, store, time and supplier because fresh produce and branded products may behave differently and aggregate rates can hide repeated gaps in key items. A picker needs realistic time to search for the original, as substituting too quickly can mask stock-location errors, and training and store layout can improve accuracy.
A fictional grocery chain sees high substitutions late on Sundays, compares inventory counts and replenishment schedules, and finds the problem is shelf restocking, not customers choosing unusual items. Data capture matters: record original item, proposed alternative, customer response and final outcome where the system permits it, because counting a proposal as completed replacement inflates the rate.
Measure customer acceptance and refund rates beside substitution, since a high replacement rate with high rejection signals poor alternatives and a low rate with many cancelled items can be worse. A product-size change may alter the price, so explain how the final charge or refund is calculated, and do not assume consent to a different item includes consent to any price.
A fictional produce retailer offers a larger package after a smaller one runs out, the buyer declines because it costs more, and the service removes the item and records a lost sale rather than calling it a successful substitution. Measure the effect on basket value and repeat purchase, since a replacement can save one order while reducing trust; a fictional meal-kit provider that notices repeated herb substitutions changes supplier planning and recipe notes, because the best fix is better availability, not merely a larger list of alternatives, and a useful rate has a clear item definition, denominator and completion status, paired with stockout, rejection and refund measures.
In practice
Real-world examples.
Example
A grocer replaces an unavailable milk brand with an approved one.
Example
A shopper rejects a more expensive alternative.
Example
A chain investigates Sunday replacement spikes.
Formula
Calculation
Completed-item substitution rate = originally requested items fulfilled with accepted alternatives / originally requested items in scope x 100%. State how cancelled and partial items are treated.
Worked example. A fictional service has 10,000 originally requested items in scope in a week. Alternatives are proposed for 500 items; customers accept 400 and reject 100, and the rejected items are refunded.
- Completed-item substitution rate = 400 / 10,000 x 100 = 4%.
- Proposal rate = 500 / 10,000 x 100 = 5%.
- Acceptance of proposals = 400 / 500 x 100 = 80%, and rejection = 100 / 500 x 100 = 20%.
- The proposals that were rejected are not counted as completed substitutions, and the 100 refunded items are tracked as a separate refund measure.Case study
Seen in the real world.
In this fictional case, River Grocer's substitution rate rises in one store. Its records show frequent alternatives for the same cereal size. The manager finds a replenishment gap and corrects shelf stock. It then checks whether customer rejections and refunds fall as well as the headline rate.
The store also reviews which categories drive the rise and compares weekend with weekday patterns before changing any picker guidance. It keeps the definition of the denominator the same across weeks so that the before and after figures can be compared fairly. The store and its figures are invented for illustration.
Watch out
Common mistakes.
- Counting proposed alternatives as completed substitutions.
- Ignoring customer preferences or important product differences.
- Comparing item and order rates without their definitions.
Questions
People also ask.
Is a substitution the same as a refund?
No. A substituted item is replaced; a refunded item is not supplied.
Is a higher rate always bad?
It may preserve orders, but often signals availability problems.
What else should be tracked?
Customer acceptance, rejections, refunds and stockouts.
From the founder's library

Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%