There are currently a number of issues in occupation-based historical comparative research. First, there are at least three occupational schemes to which occupations are coded, but seldom to all three. That means we need mapping files to make schemes comparable on an aggregate level leading to ecological fallacy: occupations end up in different categories through mapping, then wen coded directly to multiple coding schemes. Second, when coding into occupational schemes, coders sometimes feel the need for an additional code, but seldom communicate this code. That means that coders from other projects with similar needs create different codes for the same solution adding to diffusion within occupational schemes. Third, occupational classifications are specifically linked to particular social and economic outcome scales, e.g. HISCO to HISCAM and HISCLASS, and OCCHISCO to Duncan’s SEI. The use of mapping schemes severely hampers analysis of variance as the mapping as we will demonstrate. To deal with these issues we provide a Linked Open Data (LOD) solution allowing for comparison of occupational titles on the job (title) level, rather than on the occupational (code) level. ‘Job Hoard’ is a collection of about half a million unique occupational titles from nearly a dozen languages ‘donated’ by a number of national and international research projects. Through Job Hoard researchers can compare occupational stratification values across coding schemes for one and the same occupation (depending availability). Moreover we provide the provenance of occupational titles, such as the source and institute of coding. Researchers can thus select occupational titles and codes based on such characteristics. Finally, since Job Hoard is provided as LOD researchers can use Job Hoard, to code new sets of occupational titles. For the future we see an expansion of Job Hoard, with other characteristics of occupations, such as wages and labour relations.