What happened to the demand for your work.

Every kind of work we measure. Find yours.

The number of advertisements we observe in this dataset grew ×2.20 over the two years to June 2026. That is a figure about our data, not a measurement of the labour market. Measured against it, 13 of 23 kinds of work grew faster and 6 more slowly. For 4 we publish no direction at all: reasonable aggregation methods produce different classifications, and choosing one would be choosing the answer. Open a page to see which way yours went — and where in the world it went differently.

The seven groups are ours, and the grouping is the part we chose: each kind of work sits where most of its job titles fall under ISCO-08, the international standard the ILO publishes, and we aggregate its major groups into the seven plain-English names above. The methodology sets out the rule.

The 24 markets we counted.

Ordered by how many postings we could see in the 12 months to June 2026. Coverage is not the same everywhere, and the count is printed next to each one so you can judge it yourself — Argentina at 5,928 postings does not carry the same weight as United States at 4.8M.

United States4.8M

Brazil677k

France529k

United Kingdom505k

Canada431k

Germany426k

Mexico368k

Spain347k

India344k

Italy219k

Netherlands150k

Australia142k

United Arab Emirates81k

Poland41k

Portugal36k

Colombia34k

Switzerland30k

Chile27k⚠

Pakistan26k

South Africa25k

Austria20k⚠

Nigeria17k⚠

Indonesia11k

Argentina5,928⚠

2 more were collected but are not published: Bangladesh and Ethiopia — our provider has no coverage of them at all under the published recipe, so there is no base window to compare against. 4 published markets carry a caution: Argentina, Austria, Chile and Nigeria. In those, at least a quarter of what we counted in the base window sits in occupation cells too small to publish on their own, so read the country trend as a direction rather than a figure. How we decide that.

Flag artwork: flag-icons by Panayiotis Lipiridis, MIT licence.