Page tree
Skip to end of metadata
Go to start of metadata

This page aims to facilitate the reuse of the TED (Tenders Electronic Daily) open data by collecting links to already existing reuse examples and serving as a repository for sharing ideas. The target audience for this wiki includes researchers, journalists, and data analysts, as well as procurement practitioners and policy makers. Your experience with TED data does not matter: the page should be useful for novices (e.g. students looking for a dataset and a topic for their thesis) as well as experienced professionals (e.g. procurement practitioners who want to submit an idea).

Please note that Digital Whistleblower (DIGIWHIST), a large EU funded research project on procurement data, provides a very comprehensive overview of existing procurement networks, research, legislation, data sets, and government agencies. This wiki does not duplicate this information, thus please refer to the DIGIWHIST website as a primary source of information, especially for an overview of existing articles. However, please note that the scope of the DIGIWHIST project is broader than the research done on the TED dataset.   

This page is conceived as a wiki, i.e. editing by any user is possible. Please contribute below!

Reuse examples

Reuse / research proposals

  • Some contract notices indicate that tenders are accepted in multiple languages. Does this increase the number of bids? Does this increase cross-border procurement? How does the effect differ depending on the language (e.g. is the effect strongest for English; does it depend on geographical proximity (e.g. German in Czech Republic); on mutual intelligibility (e.g. Portuguese in Spain, Slovak in Czech Republic)).
  • Is there any impact on cross border procurement by proximity (neighboring regions from different countries), shared currency (eurozone vs. non-eurozone), shared language (e.g. UK / Ireland; regions in Switzerland and EU countries speaking the respective language).

  • When contract notices indicate that tenders may be submitted after a longer period than is the legal minimum, does this influence the number of received bids?
  • Is their a correlation between groups of companies which win and the election cycles in countries? I.e. bigger changes in winning companies around election years than in other years?

  • What increases the probability of a notice getting cancelled?

  • Does repeated awarding of contracts to the same suppliers correlate with lower savings, or any indicators of bad practice? Is is possible to predict vendor lock-in?
  • In the USA, "no top 25 supplier of the government has, reportedly, been founded after 1975". What about the EU? How much is the supplier structure changing? Does this follow the same patens as with the private sector, or is it different? (See https://www.youtube.com/watch?v=z3CwGA2SWlc for inspiration.)
  • How has the quality of data submitted to TED changed with changes to procurement legislation?
  • What is the impact of missing and incorrect information in contract notices such as lack of cpv code on bidding outcomes such as number of bidders or the success of non-local bidders?
  • No labels

14 Comments

  1. Good day.
    Tell me, can this text

    Reuse / research proposals

    be considered as a task for research?
    Are the results of research published in dataset?

  2. Hello, thank you for your question. Regrettably, we're not 100% sure what you mean by "tasks for research". The list is intended as "ideas for research" though, so probably yes, you could take it as "tasks for research"? 

    As far as we know, the questions have not been replied to anywhere. Concerning the dataset on the EU open data portal page, they contain just the TED data. Other examples of research on procurement data, besides those above, are available for example at http://digiwhist.eu/resources/research-and-policy-papers/

  3. Are there any other datasets to help? Who is the contact with the research results?

  4. There are many potential datasets with which procurement data can be linked, but we have no obvious recommendation beyond the data available through Digiwhist (https://opentender.eu/download). 

    Concerning the contact point for reporting your research, please see point 1.1. of the TED notesGROW-G4@ec.europa.eu

  5. Jachym HERCHER 

    I'm from Ukraine. Here for the first time.
    I think that any idea for a solution becomes a condition of the problem. For accurate wording you need more numbers, percentages and fewer words.
    Tell me, what data files do you need help with? What are the csv file names and sizes?
    To solve on a computer or in the cloud?
    thank you
    Dr A. Szpinkowski


  6. I am currently writing a dissertation about defence equipment procurement (Directive 2009/81/EC) and would like to know if there are any CSV datasets on defence equipment procurement? I've only come across these (https://data.europa.eu/euodp/en/data/dataset/ted-csv) which do not seem to include defence equipment procurement under 2009/81/EC ..

    Any help would be highly appreciated! Thanks a lot in advance!

    1. Dear Max,

      The CSVs contain the contract award notices published under the defence directive but the problem is that there isn't any variable in the file which identifies them. Luckily, we have a list of notice id-s of these contracts because we are working on a defence procurement related research, and we scraped this info from TED's html website. Please write me an email to the contact email address which you can find on our institute's website: http://www.govtransparency.eu/ and I'll send you the list in private.

      Best, Agnes Czibik from Government Transparency Institute

      1. Hello,

        You can find the defence directive by using the variable ID_TYPE. To find them, you need to filter out ID_TYPE = 18 (in case of CANs) or ID_TYPE = 17 (in case of CNs). The overview of all values for the forms is available at http://simap.ted.europa.eu/standard-forms-for-public-procurement. (To make it clearer in the future, we'll emphasize in the next version of the documentation that form type also corresponds to the directive type. )

        1. Thank you for your reply - that is indeed very helpful. But how do I identify defence related notices in the CSV datasets for earlier years? Because the standard forms used now weren't used back then right? I'm looking at a period from 2008 to 2017 and the first few years are very hard to get by..

          Thank you very much in advance!

          1. The defence directive (2009/81/EC) was adopted in 2009 and then it takes several years for EU countries to transpose it into national law, so that's why they aren't there. Before the defence directive, publication for defence was essentially voluntary, so there is likely to be very little (and whatever there is will be published under the classical procurement directive, with no clear way to single it out). 

            1. Yes, that's exactly the issue.  Do you know in which CSV datasets I can find the prior information notices (standard form 16)? They are not included in the CN or CAN datasets as they are another type of notice.

              1. Regrettably, we don't currently publish prior information notices in CSV format (unless they are used as a call for competition, which is not the case for the defence directive). We are planning it for the future, but it will take quite some time.

                1. Thanks for getting back to me so quickly - your comments are very helpful for my understanding.

                  That is what I assumed, I was just hoping I hadn't missed anything. 

                  Another question of understanding: If the CSV datasets have the exact same entry multiple times, does it mean that the contract is awarded multiple times or is it possible that there are double entries by mistake? I just want to make sure I count and analyse the correct number of notices. 

                  1. No problem. The topic of repeated lines is a bit complex, please have a look in section 3 of the open data documentation, it explains it in quite some detail. Of course, beyond the questions of overall structure, errors are always possible.