Ocean import delay analysis

87,080US ocean import shipments from CBP manifest data, scored against each shipment’s own estimated arrival. Search for a company to see where its delays come from.

87,080
Shipments analysed
43.8%
Arrived late
38,179 shipments
5 days
Average delay, when late
median 2 days
396
Companies with a dashboard
of 9,398, those over 20 shipments

How late, when late

Every shipment placed in a band. Early and on-time are shown too, so the delayed share is read against the whole, not on its own.

6,892
7.9%
Early
42,009
48.2%
On time
24,421
28%
1–3 days
5,098
5.9%
4–7 days
5,244
6%
8–14 days
3,416
3.9%
15+ days

Delays by carrier — all companies

Every carrier in the dataset, ranked by volume.

% delayed avg days late, when laten = shipments
  1. Maersk
    20.1%
    1.4d
    15,3153,074 late
  2. MSC
    60.7%
    3.2d
    9,6315,845 late
  3. ONE
    25.7%
    1.5d
    8,8622,274 late
  4. CMA CGM
    65.7%
    4.6d
    8,2905,448 late
  5. Zim
    38.5%
    8.4d
    7,8073,003 late
  6. Hapag Lloyd
    35.8%
    2.2d
    6,5782,353 late
  7. Evergreen
    18.5%
    1.5d
    6,5091,206 late
  8. HMM
    84%
    6.9d
    4,2213,547 late
  9. COSCO
    49.8%
    4.4d
    3,4431,713 late
  10. Wan Hai Lines
    97.8%
    4.2d
    3,0783,011 late
  11. HEDE Shipping
    0%
    2,3910 late
  12. Kuehne + Nagel
    85.7%
    18.5d
    1,9961,711 late

Delays by vessel — all companies

The 12 busiest of 439 named ships. Vessel names arrived with the latest pull, so this covers 24,516 of 87,080 shipments rather than all of them.

% delayed avg days late, when laten = shipments
  1. Gerd Maersk
    2.8%
    3.9d
    78922 late
  2. Ain Snan Express
    8.1%
    4.2d
    72759 late
  3. Madrid Express
    0%
    5000 late
  4. Ever Mild
    0%
    4510 late
  5. Maersk Seletar
    0%
    3900 late
  6. Ym Width
    11%
    3.8d
    35439 late
  7. Ever Muse
    33.1%
    1d
    344114 late
  8. Zeal Lumos
    7%
    3.9d
    33023 late
  9. Good Prospect
    100%
    1.1d
    321321 late
  10. One Sprinter
    53.8%
    3.6d
    299161 late
  11. Conti Conquest
    22.3%
    7.8d
    29265 late
  12. Budapest Express
    100%
    2.7d
    279279 late

Delay rate by arrival date

Share of each day’s arrivals that came in late. A single late vessel puts the same delay on every container it carried, so a tall isolated bar is usually one sailing rather than a bad week — the count under each bar is the check on that.

49.4%
n=1950
2026-07-27
41.3%
n=1667
2026-07-28
45.4%
n=3232
2026-07-29
74.7%
n=2279
2026-07-30
47.7%
n=2100
2026-07-31
31.9%
n=4482
2026-08-01
23.1%
n=2459
2026-08-02
66.2%
n=1656
2026-08-03
36.2%
n=28324
2026-08-04
52.3%
n=22278
2026-08-05
64.5%
n=1922
2026-08-06
57.1%
n=3546
2026-08-07
51%
n=1582
2026-08-08
12.4%
n=3082
2026-08-09
50.6%
n=2500
2026-08-10
43.3%
n=2137
2026-08-11
37.9%
n=1884
2026-08-12

Delays by route — all companies

The 12 busiest lanes of 887.

% delayed avg days late, when laten = shipments
  1. Shanghai > Los Angeles, Ca
    21.9%
    10.1d
    5,4981,205 late
  2. Yantian > Long Beach, Ca
    14.9%
    4.7d
    3,670546 late
  3. Ning Bo > Los Angeles, Ca
    21.1%
    9.3d
    3,539747 late
  4. Yantian > Los Angeles, Ca
    32.7%
    1.8d
    2,620856 late
  5. Vung Tau > Los Angeles, Ca
    40.7%
    2d
    2,254917 late
  6. Yangshan > Long Beach, Ca
    7%
    5.3d
    2,230156 late
  7. Pusan > Tacoma, Wa
    38.1%
    1.5d
    2,108803 late
  8. Pusan > Los Angeles, Ca
    39.4%
    6.5d
    1,857731 late
  9. Ning Bo > Newark, Nj
    65.9%
    3.3d
    1,7601,159 late
  10. Mina Raysut > Newark, Nj
    48.3%
    1.1d
    1,411681 late
  11. Colombo Harbor > Newark, Nj
    1.9%
    2.3d
    1,24424 late
  12. Hai Phong > Los Angeles, Ca
    41.4%
    2d
    1,243515 late

How these are calculated

A shipment is delayed when its actual arrival is later than the estimated arrival on the same manifest record — so the comparison is against what was expected for that shipment, not against a lane average. Days late is the difference between the two, and averages of it are taken over delayed shipments only; including the on-time ones would dilute severity into something that reads like a small number when it is not.

Shipments that arrive early or exactly on the estimate both count as not delayed, and are shown separately in the distribution above. Companies with 21 or fewer shipments have no page: a delay rate off a handful of boxes moves ten points on one late arrival, which is noise presented as a finding.