To evaluate logistics performance, an importer needs a small set of numbers that are defined the same way every month. Without that, a forwarder can report a healthy on-time rate while one late shipment stops a production line, and a buyer can feel that costs are rising without knowing where. This guide gives seven metrics with formulas, works through eight made-up shipments to show why the average can mislead, explains how India's National Time Release Study measures customs speed and ends with a scorecard to review with your forwarder. Define each metric in writing before you collect data, because a figure with no definition cannot be compared next quarter.
For tracking tools, see how shipment tracking improves supply chain visibility. For the wider picture, see how businesses can improve end-to-end supply chain efficiency.
Seven Metrics and How to Calculate Them
| Metric | Formula | What it shows |
|---|---|---|
| On-time rate | Shipments delivered by the agreed date divided by all shipments | Reliability against a promise |
| In-full rate | Shipments delivered complete and undamaged divided by all shipments | Cargo condition and accuracy |
| Transit time | Delivery date minus departure date, reported as mean, median and spread | Speed and predictability |
| Customs release time | Time from arrival to Out of Charge | Clearance speed |
| Landed cost per kilogram or per unit | Total landed cost divided by kilograms or units | Cost efficiency |
| Document correction rate | Shipments needing a correction divided by all shipments | Quality of data and paperwork |
| Response time | Hours from your query to a useful answer | Service quality |
Choose a tolerance for on-time before the quarter starts, such as plus or minus one day, and keep it for the whole quarter.
A Worked Example: Eight Made-Up Shipments
Suppose the plan for a lane is 22 days, and eight shipments take 21, 22, 24, 22, 30, 23, 22 and 25 days.
| Measure | Calculation | Result |
|---|---|---|
| Mean transit time | 189 days divided by 8 | 23.6 days |
| Median | Middle of the sorted list: 21, 22, 22, 22, 23, 24, 25, 30 | 22.5 days |
| On time within one day of plan (23 days or less) | 5 of 8 | 62.5% |
| Spread (population standard deviation) | Square root of (57.875 divided by 8), where 57.875 is the sum of the squared gaps from the mean | 2.7 days |
| Latest minus planned | 30 minus 22 | 8 days |
Three points follow. The mean (23.6) is higher than the median (22.5) because one shipment took 30 days, so the mean describes a typical shipment less well than the median does. The on-time rate of 62.5% hides the size of the miss, so report the worst case too. And the spread of 2.7 days tells you how much buffer stock the lane needs, which is why a predictable 24-day lane can be easier to plan around than an unpredictable 22-day lane. Our guide to inventory management and logistics shows how transit variability turns into safety stock.
How India Measures Customs Speed
The Central Board of Indirect Taxes and Customs publishes the National Time Release Study 2025. It defines Average Release Time as the arithmetic mean of the duration between the arrival of goods and the granting of Out of Charge by Customs. It also reports a standard deviation for bills of entry released within the national target, and says a low deviation shows release times that are more consistent and closely aligned with the average, while a high deviation shows more unpredictability. In its tables, the deviation for imports at seaports was 13 hours 26 minutes in 2025 against 36 hours 37 minutes in 2024.
The same study reported that in its sample of bills of entry filed from 1 to 7 January 2025, seaport bills filed in advance averaged 71 hours 23 minutes. The study identifies advance filing, Risk Management System facilitation and Authorised Economic Operator accreditation as the "Path to Promptness", and reports that Direct Port Delivery containers had an overall average of 65 hours 33 minutes against 84 hours 3 minutes for the average at container freight stations. These are national figures from a one-week sample, so use them as a reference for the questions to ask and not as your own target. The WTO's Trade Facilitation Agreement encourages its Members in Article 7.6 to measure and publish average release times of goods periodically, and the same idea works for a company: measure arrival to release on your own shipments.
India's national position gives context for why this matters. A PIB backgrounder records that India ranked 38th of 139 countries in the World Bank's 2023 Logistics Performance Index and aims to be among the top 25 by 2030.
Cost per Kilogram: A Second Worked Example
A total can rise for a good reason. Suppose, with made-up figures, that landed cost for one month was 12,400 on 3,100 kg, which is 4.00 per kilogram. The next month it was 13,500 on 2,700 kg, which is 5.00 per kilogram. The total rose 8.9% (13,500 divided by 12,400 is 1.089) while the weight fell 12.9% (2,700 divided by 3,100 is 0.871), so the cost per kilogram rose 25% (5.00 divided by 4.00 is 1.25). The total alone would have hidden the problem, and the unit cost alone would not tell you whether the cause was a surcharge, a part-empty container or a switch from sea to air. Split the figure by cause, as in hidden shipping costs importers should include in their budgets.
Avoid These Measurement Mistakes
- Changing definitions. If on-time moves from "delivered by the promised date" to "delivered within three days of it", the trend becomes meaningless.
- Reporting only the mean. As the example shows, one late shipment can lift the mean while the median stays low.
- Ignoring the starting point. Measure from the date you needed the goods or the date the carrier promised, and say which.
- Mixing lanes. A China-to-India sea lane and an air lane belong in separate rows.
- Counting too few shipments. Eight shipments are enough to illustrate a method but not to settle a dispute, so build up a longer record before you act on a small difference.
Setting Targets From a Baseline
Targets set before a baseline exists are guesses. Collect at least a quarter of data, then set each target as an improvement on your own median and your own spread, not as a copy of a national figure. In the transit example, a first target might be to bring the spread from 2.7 days to 2.0 days and to keep the worst case within 5 days of plan, because those two numbers describe predictability, which the mean hides. These targets are made up. Write each one with its definition, its owner and its review date, and drop a target if you cannot measure it.
One Page for Each Metric
Keep a one-page definition for every metric: its name, its formula, the data source, the owner, the reporting period and the tolerance. When a number changes, the first question is whether the data, the definition or the performance changed, and the sheet answers the first two.
A Quarterly Scorecard to Use With Your Forwarder
- Agree the seven metrics and their definitions in writing.
- Fill in the baseline from the last quarter's records before you set any target.
- Add the median and the worst case next to every average.
- Record the reason for every late or damaged shipment, such as a booking roll, congestion, a document correction or a hold.
- Review the scorecard each quarter, and change one thing at a time so you can see its effect.
- Ask for the data behind each figure, not just the summary.
See how to avoid documentation errors during import customs clearance for the main cause of document corrections.
Measure Your Shipments With RR Brothers and Logistics
RR Brothers and Logistics arranges sea freight and air freight from China and handles customs clearance in India. Tell us which metrics you track and we will discuss which status and timing data we can provide.
Frequently Asked Questions
Choose a small set of metrics, such as on-time rate, in-full rate, transit time and its spread, customs release time, landed cost per kilogram, document correction rate and response time, define each in writing and review them every quarter.
One late shipment can raise the mean while the median stays low. In the made-up example, eight shipments gave a mean of 23.6 days, a median of 22.5 days and a worst case of 30 days against a 22-day plan.
The National Time Release Study defines Average Release Time as the arithmetic mean of the duration between the arrival of goods and the grant of Out of Charge by Customs.
The study says a low deviation shows release times that are consistent and close to the average, and a high deviation shows more unpredictable release times.
Eight shipments illustrate a method but are too few to settle a dispute. Build a longer record, keep the definitions fixed and record the reason for each late or damaged shipment.
Metric definitions, a baseline from the previous quarter, the median and worst case next to each average, reasons for late or damaged shipments and a quarterly review with your forwarder.
Sources
- CBIC: National Time Release Study 2025 (released 20 June 2025)
- WTO: Trade Facilitation Agreement
- PIB: Logistics, India's growth engine (backgrounder, 16 August 2025)
This article is general information, not legal, tax or customs advice. Rules, rates and procedures change; confirm current requirements with the relevant authority or a licensed customs professional before you ship. Regulatory references were checked against the official sources listed above on 28 October 2026.
Image credits: header photo by William Warby on Unsplash; photo by Luke Chesser on Unsplash. Used under the Unsplash License.
Request a Quote
