The Challenge

Most organisations have a multitude of data sources, including

  • historic data
  • observations that come in through messages from neighbours or informers
  • data from third-party data sources that do not have an API but which can export their data as csv-file


By making these data sources available through the Sensing Clues Upload Service, you can include them in your analytics, maps and reports in the same way as you can with Cluey observations.



How to

The upload service can handle .csv, .tsv and skv files.

To avoid any issues with conversions etc. we recommend to user tsv files (tab delimited). If you want to use other types, please contact us if you encounter any issues.


General instructions

  • The filename and the headings of the columns are case-sensitive! 
  • The Id-column values need to be unique within the file
  • Multiple date formats can be handled, as long as you are consistent within one file. 
    The preferred notation is yyyy-mm-dd hh:mm:ss
    If you just have dates, you can also use yyyy-mm-dd, dd-mm-yyyy or dd/mm/yyyy
  • Coordinates are preferably provided in WGS84 format (for example: Lat 22.36353 Long 77.63536. Other formats are possible, but require manual configuration on our side. Contact us if you use e.g. UTM coordinates
  • The order or the columns is not important
  • If you have an updated version of the file, you can simply overwrite it in the Upload Service. Make sure the file name is identical, otherwise it will be interpreted as a new one and all data will be doubled.
  • If you delete a file from the Upload Service, its data is also deleted in the rest of the platform. 


Mandatory columns

All CSV-files need at least the following fields.

  • ID                 the values in the ID column need to be unique within the file
  • Date             see above for the preferred notation
  • Lat                WGS84 is the preferred format
  • Long            WGS84 is the preferred format


Please note that the Lat & Long should not contain more than 4 or 5 decimals:

  • 4 decimals (± 10m accuracy) is closest to the quality of the collected data and it's usage.
  • 5 decimals  (± 1m accuracy) is an overestimate of your accuracy, unless you work with Garmin in ideal situations
  • 6 decimals (± 10cm accuracy) represents a fake accuracy.

The impact of more decimals is that search Indices become significant larger, and hence, query response times much longer.


Optional columns

Anything you like, but make sure the headings do not contain characters like * ^ % $ # @ + " ; '
Also check the content of the columns. The results in Focus are better if values are consistent.