Reading Kenyan customs data without fooling yourself

Customs records are the most detailed public evidence of what a company actually imports. They are also routinely misread, usually in the direction the reader was hoping for.

An entry is not a shipment and not a sale

A customs entry is a declaration. One shipment can generate several entries when it contains goods under different tariff headings. A company with more entries than a competitor is not necessarily moving more goods; it may simply be importing a wider range of products.

No declared value means no declared value

The aggregated customs sets available for Kenya frequently omit declared values. That absence is not an invitation to estimate. Multiplying an entry count by an assumed average consignment value produces a number that looks like research and is actually a guess. We publish our own entry counts and deliberately publish no turnover figure alongside them, because the source does not support one.

Year-on-year movement needs care

Our own record shows 344 import entries indexed for 2024 and 274 for 2025. Read carelessly that is a decline. Read properly it is a change in entry composition across a period where the 2024 figure covers only part of the year, since indexing began on 30 July 2024. Comparing a partial year with a full one is the most common error in this data.

Late indexing moves the past

Entries are indexed after the fact and the lag is not uniform. A figure for the most recent quarter will usually rise after publication. Any analysis that treats the latest period as final will systematically understate it.

What the record does support

Composition and continuity. Which tariff chapters a company works in, which origins it buys from, how many distinct counterparties appear, and whether that has been stable. Those are questions the record answers well, and they are usually the questions that matter when you are assessing whether a counterparty actually does what it says it does.