Multi-source Company Data

impacted dataset multi source company data impacted formats all available formats impacted delivery starting with may 2025 delivery impact new added fields, minor logic changes new fields added inferred employees counts to improve the quality of the counts, we are adding a new derived field and a full array with a historical breakdown of collected employee data we are excluding low quality profiles from headcount calculations and improving the process of matching employees to companies by implementing a more strict company standardization process action collection name data type description added added employees count inferred integer estimated number of employees, calculated based on inferred employee data added added employees count inferred by month array of structs estimated number of employees, calculated based on inferred employee data, for a three year rolling window added added employees count inferred by month employees count inferred integer added added date string date identifier see code snippet for reference { "employees count inferred" 20, "employees count inferred by month" \[ { "employees count inferred" 20, "date" "202504" }, { "employees count inferred" 18, "date" "202503" } ] } key executives quality improvements in our efforts to remove such employees from key roles, we have used our quality score threshold to remove unfinished data entries (e g old experience records missing mandatory fields such as dates) as well as stale, inactive profiles which do not belong to key executive roles this change removes about 3 3m employee records from our key executives' collections a list of affected collections can be found below data point data type description key executives array of structs removed low score profiles key executive arrivals array of structs removed low score profiles key executive departures array of structs removed low score profiles bug fixes data point data type description funding rounds array of structs fixed a bug that caused funding rounds records duplicates in the same array employees count breakdown by department struct fixed issue with lower cased values, improving mapping results employees count breakdown by department by month array of structs fixed issue with lower cased values, improving mapping results employees count breakdown by seniority struct fixed issue with lower cased values, improving mapping results employees count breakdown by seniority by month array of structs fixed issue with lower cased values, improving mapping results employees count breakdown by region struct fixed issue with lower cased values, improving mapping results employees count breakdown by region by month array of structs fixed issue with lower cased values, improving mapping results employees count by country struct fixed issue with lower cased values, improving mapping results employees count by country by month array of structs fixed issue with lower cased values, improving mapping results a three year rolling window was implemented for multiple fields a three year rolling window is now implemented for each breakdown and by month fields this change is dedicated to improving the ease of use for both flat file receiving and api using clients note employees reviews and total website visits fields are not impacted directly, since we don’t have information for three years yet (website visits started in april of last year, while employee reviews contain data starting in june 2022) the list of directly affected data points can be found below data points data type active job postings count by month array of structs employees count breakdown by seniority by month array of structs employees count breakdown by department by month array of structs employees count breakdown by region by month array of structs employees count by country by month array of structs employees count by month array of structs professional network followers count by month array of structs product reviews score by month array of structs