Why I built Parteiatlas

I started Parteiatlas because a simple question about German politics rarely has a simple route to an answer.

If you want to know what a party stands for, you can read its manifesto. If you want to know how many seats it holds, you can visit the Bundestag. Election results sit somewhere else. Coalition agreements, legislative records, polling and political analysis each have their own homes. The information exists, but the reader has to assemble the picture.

That work becomes harder when you are new to the system, returning to politics after some time away, or trying to understand an issue outside the daily news cycle. A headline may tell you what happened today. It rarely tells you how a party arrived there, what it promised before an election, or how its position compares with the alternatives.

I also found a gap for people who want to go deeper in English. Germany has excellent political information, but much of the most useful detail is written for readers with strong German and an existing understanding of the country’s institutions. Parteiatlas is for English speakers, international residents, learners and anyone whose German is not strong enough to comfortably work through manifestos, parliamentary material and official datasets. It should make that material more approachable without flattening it into a few slogans.

I wanted a guide that begins with the reader’s question, then keeps the evidence close enough to inspect.

More than an election tool

Many political guides are built for the weeks before an election. They help people compare answers, then become less useful once a government is formed. But politics continues after election night. Parties negotiate, compromise, govern, oppose, introduce laws and sometimes abandon the promises that brought them there.

Parteiatlas is meant to connect those stages. A reader should be able to move from a party profile to an issue comparison, see the current electoral picture, understand which coalitions are plausible, and eventually follow a promise through the policy-delivery record.

Clarity should not require false certainty

Political information mixes several different things: official facts, party claims, reported events and editorial interpretation. They should not look interchangeable. An election result is a measured fact. A manifesto position is the party’s own proposal. A point on an ideological axis is an interpretation that needs an explanation.

That distinction shapes the site. Sources sit next to material claims. Election-day results and today’s parliamentary groups are shown as different snapshots. The party overview publishes its scoring method, rationale and evidence. Where information has not been fully reviewed, the product should say so instead of borrowing authority from a polished interface.

Independent by design

Parteiatlas is not affiliated with any political party. It is not here to tell people how to vote, and it does not pretend that a chart can make that decision for them. The purpose is to make the choices easier to examine: what parties say, where they differ, what power they hold and what happens after promises meet government.

Independence also means being open to correction. A source can go stale. A political status can change. An interpretation can miss context. The right response is to expose the method, date the information and give readers a direct way to challenge an error.

Built faster with AI and GPT

I used AI, including GPT, to make both the research and the build faster. It helped me organise source material, compare large amounts of information, shape early drafts, write and review code, and move from a prototype to a working site more quickly than I could have done alone.

That speed does not make an AI output a source. The underlying political claims still need traceable evidence, and interpretive decisions still need to be labelled and explained. AI supports the work; it does not replace source checking, editorial judgement or responsibility for what appears on the site.

From a local project to a public guide

I originally built Parteiatlas with AI as something I could use locally. It was a practical way to bring the information I kept looking for into one place, organised in the way I wanted to read it.

After I showed it to a friend, they told me it was quite informative and that I should share it online because it might be helpful to other people too. That conversation is what pushed the project from a private tool towards a public guide.

Publishing it also made two things especially important to me. The site should state clearly that I am not affiliated with any political party, and readers should have a straightforward way to report errors or request corrections. Sharing the work means being open about where it came from and willing to fix it when something is wrong.

This is the beginning

The current site is a foundation: eight party profiles, issue comparisons, election and parliamentary snapshots, coalition context, and an early policy-delivery tracker. There is much more to build. The delivery record needs deeper structured data. Polling and office-holder information need clear freshness rules. German language support, history and regional detail belong in the longer journey.

I am building Parteiatlas because understanding political power should not depend on knowing which document to search for or which institutional website to open first. The guide should do that organising work while leaving the judgement with the reader.