What Is Delivery Time Variance?

A carrier that delivers in three days on average sounds reliable. But if that average is built from a mix of one-day deliveries and five-day deliveries in roughly equal measure, the average is hiding a consistency problem that customers experience directly. Delivery time variance is what measures that inconsistency and makes it visible enough to act on.

What Is Delivery Time Variance?

Delivery time variance is the degree to which actual delivery times deviate from the expected or promised delivery window across a set of shipments. It measures not just whether deliveries are on time on average, but how consistent the delivery experience is across all shipments on a given carrier, lane, or service level.

Low variance means most shipments arrive within a tight window around the expected date. High variance means delivery times are spread across a wide range, with some shipments arriving significantly earlier and others significantly later than expected. From a customer experience perspective, high variance is a reliability problem regardless of what the average transit time looks like.

Why Variance Matters More Than Average Transit Time

Average transit time is a useful baseline measure. It tells you what a typical shipment experience looks like in the middle of the distribution. What it does not tell you is how far individual shipments deviate from that middle.

A carrier with a three-day average transit time and low variance delivers a consistent, predictable experience. A carrier with the same three-day average and high variance delivers an unpredictable experience where some customers get their order in one day and others wait five or six. Both carriers look identical on average transit time. They are very different operational partners.

Variance is the measure that captures the difference.

How Delivery Time Variance Is Measured

The standard statistical measure of variance in a dataset is standard deviation. Applied to transit times, it measures how tightly clustered individual delivery times are around the average.

A low standard deviation means most deliveries cluster close to the average. A high standard deviation means deliveries are spread across a wide range. For practical operational use, tracking standard deviation by carrier, service level, and lane over time gives a consistent measure of consistency that average transit time alone does not provide.

P95 transit time works alongside standard deviation as a practical variance measure. Where standard deviation describes the spread of the full distribution, p95 describes the threshold that 95% of shipments fall within. Both numbers together give a complete picture of typical performance and worst-case realistic performance on a given lane.

Where Variance Creates Operational Problems

Delivery promise accuracy suffers when variance is high. Promises set using average transit time will be missed on shipments that fall on the slow side of a high-variance distribution. Promises calibrated using p95 data are more conservative but significantly more reliable across the full customer base.

Customer service volume correlates with variance more than with average transit time. Customers whose orders arrive in one day rarely contact support. Customers whose orders arrive in five days when three were promised almost always do. High variance generates disproportionate support volume relative to its frequency because the customers affected are the ones most likely to reach out.

Inventory planning is also affected. Brands that use transit time data to time replenishment orders assume a predictable delivery window from suppliers. High transit time variance from logistics partners introduces uncertainty into inventory arrival timing that creates either buffer stock costs or stockout risk depending on how conservatively the planning accounts for it.

Using Variance to Make Better Routing Decisions

Variance data by carrier and lane creates a basis for routing decisions that average transit time analysis cannot support.

When two carriers have similar average transit times on a lane but different variance profiles, the lower-variance carrier is the more reliable operational choice for that lane. Routing decisions based on average transit time alone miss this distinction. Routing decisions that incorporate variance data select for consistency rather than just typical performance.

For lanes where delivery promise accuracy is critical, routing to the lowest-variance carrier even at a slightly higher cost may be the right operational decision when the cost of a missed promise, in customer service volume, reshipments, and customer lifetime value impact, is factored into the comparison.

Seeing inconsistent delivery performance but not sure which carriers or lanes are driving it? Talk to one of our shipping experts. Book a demo.