The first Wahl-O-Mat started in 2002, several weeks prior to the federal elections. Due to effective marketing, e.g. in a German late night show and in magazines like “Stern” and “Spiegel”, the Wahl-O-Mat became famous right from its first implementation; the 2002 version was played about 3.6 million times. Since its first implementation the Wahl-O-Mat has become one of the most popular and famous online tools in Germany. Its initial success spurred new versions for the federal election in 2005 and 2009, the European Elections in 2004 and 2009, and for almost all elections on the sub-national level. The best score of the Wahl-o-mat has been achieved prior to the German federal election in 2009 when the tool was played more than 6.7 million times.

The Wahl-O-Mat presents between 30 to 40 propositions to its users, for example, ‘All public buildings should by law become non-smoking areas’ or ‘The VAT must be raised’. People visiting the website are asked to take a stand on the propositions by clicking one of three buttons: ‘agree’, ‘disagree’ or ‘neutral’. Additionally, users have the option of skipping several theses. After voting on all items on the list, the users can mark those propositions, that they consider important to them, to give them a special weight in the final calculation. Finally, the Wahl-O-Mat calculates the distances between the voter’s and the parties’ positions and displays, as the result, the party with the smallest distance - the best match.

Additionally, the Wahl-O-Mat calculates the extent of agreement between the voter’s position and all parties by displaying the respective summed distances for all parties selected. Furthermore, for each proposition the users have the option to take a closer look at the relation between the parties’ positions and their own point of view. Finally, Wahl-O-Mat players can have a look at any explanation the parties may have provided concerning the positions they have taken on these propositions.

The Wahl-O-Mat versions are produced by the “Bundeszentrale für politische Bildung” (bpb) (Federal Agency for Civic Education), partly in cooperation with other agencies of civic education. About three months prior to the elections, a group of first- or second- time voters (the “Wahl-O-Mat editorial staff”), all under the age of 27, convene in order to generate the policy propositions or “theses”. Based on the party platforms and other information provided at the time of the first staff meeting, the participants identify relevant issues that might be controversial within the election campaign. Journalists and political scientists advise them in the process, and there are numerous other feedback loops to ensure quality and accuracy.

The propositions must meet several criteria.  They should: 1) be easy to understand; 2) address questions relevant to citizens’ lives; and 3) present different party points of view in respect to them. Before launching the tool (approximately five weeks before the election day), parties are asked to position themselves in respect to about 60 to 80 theses, which meet the mentioned criteria and upon which the Wahl-O-Mat editorial staff agreed. On the basis of the parties’ responses, a final set of around 30 to 40 propositions is selected. This final selection is guided by, firstly, statistical calculations, ensuring that the theses are able to distinguish appropriately between the Wahl-O-Mat-parties, and secondly, by the question of relevance: What are the vital issues for the voters?

The final list is then turned into an online Wahl-O-Mat, becoming part of the website of the Federal Agency of Civic Education. Download versions for personal computers and mobile phones are also available. In addition, didactical material for schools, colleges, and universities is provided for each version.

The results are calculated on the basis of the city block method. The party receives one point for every block difference between the party and the user, so the party with the lowest score is the one nearest to the user. The difference between “agree” and “don´t agree” is translated into two points. The difference between “agree” or “don´t agree” and “neutral” counts one point. By giving a proposition a special weight, the users can double the score of the respective differences to the parties.