Discover unique insights and generate custom reports on public policy in any jurisdiction. Customers use us to analyze changes in US state education policy, generate reports on UK financial regulations, monitor vaccine policies across the EU, and much more!
Parsagon
Hi Makers!
I'm excited and a little bit nervous to introduce you to Parsagon, a new way to track and analyze policy developments.
🌟 What is Parsagon?
Parsagon is an AI that can collect and analyze publications from any government body in any jurisdiction. It currently parses and analyzes legislation, press releases, speeches, meeting minutes, executive orders, and much more in 64 countries and counting.
Fortune/Global 500 companies, policy firms, non-profits, and other organizations are using Parsagon to uncover policy insights in hundreds of jurisdictions that they would otherwise miss.
🧠 Key features:
✅Automatically monitors any government body in any jurisdiction
✅Automatically translates documents in different languages
✅Search for relevant policy developments using natural language
✅Generate polished, custom reports, monitoring alerts, briefings, and more
🤔 Why Parsagon?
Parsagon started out as an AI data collection tool for developers, but it turned out that many sign-ups were from government affairs professionals who were trying to scrape and monitor government websites. This seemed odd, since there are platforms that track government activity and would presumably be better suited to their use case than a developer tool.
But as these users told me their stories, I realized why they were trying to use Parsagon. Current political intelligence platforms are outdated. They track a limited set of sources, and they rely on clunky keyword searches that are too noisy to adequately track the complex issues organizations care about. This means government affairs professionals have to spend countless hours sifting through bills, press releases, etc., which are spread across multiple platforms and dozens or hundreds of government websites.
We felt confident we could make something much better for these users, and thus Parsagon was born.
🚀 Join Us in Making Governments More Accessible
I'd greatly appreciate any feedback, so comment with your feature requests and questions!
This is awesome! I remember when I was working on my e-commerce business, I had to search so many websites to make sure the product I was selling could be imported there and that my ads complied with their policies. This would definitely streamline the process a lot!
Congrats on the launch!
Parsagon
@cywdev Thanks Peter! Much appreciated!
This is going to do so much to improve government transparency/accessibility of government data. Congrats on the launch, Parsagon!
Parsagon
@dxu33 Thanks Doreen! That definitely is a big goal in making this!
AltPage.ai
Dang, being able to track policy changes in *any* country is such a gamechanger—no more digging through 50 sites for one answer, fr. Love what you’ve built!
Parsagon
@joey_zhu_seopage_ai Thank you, much appreciated!
NextDrop
This is awesome and extremely helpful, especially for those working in highly regulated industries. With policies changing so often at a global scale, tracking that can be a nightmare.
Parsagon
@0xdavide Thanks Davide! And yeah, especially with all the global chaos going on right now, it seems something like Parsagon is becoming more and more necessary
Tough Tongue AI
Impressive scope—how easy is it to add a custom government source or jurisdiction not yet covered?
Parsagon
@aj_123 Very easy! A Global 500 company asked us earlier this year to add about 40
countries on short notice, and we added everything they asked for in about 2 weeks
Policy tracking at scale is impressive! But how do you ensure:
1. Source integrity (e.g., distinguishing draft bills vs enacted laws)?
2. Real-time monitoring of local government portals?
3. Version control for evolving policies?
Show me the data pipeline architecture! 🏗️
Parsagon
@erliza_p Great questions! So the pipeline architecture basically looks like many many thousands of pipelines, each one handling a specific collection of documents (e.g., there might be a pipeline to collect all press releases from the Maryland Dept of Health, and that pipeline just goes to the Maryland Dept of Health's website periodically to check for new press releases. There may be multiple pipelines for the same government body to collect each of the different kinds of publications it releases.) All pipelines write the data they collect to a relational database, with large bodies of text stored in S3.
Distinguishing different kinds of sources (e.g., draft bills from enacted laws) is simply a matter of marking what kind of sources each pipeline collects. Each pipeline labels what kind of source it collects in the database.
We don't actually do too much local government at the moment (though we don't see any reason we can't). We do plenty of stuff involving state/provincial and national governments though and would be curious how local governments differ.
Version control is done by simply re-scraping existing bills and similar documents from time to time and adding a new version to the database (while still keeping old versions ofc). It's often pretty easy because most government sites record when a bill has changed, advanced to a different stage, etc., making it easy to know when a new version has been published. The database schema for that looks pretty similar to any standard version control. That being said, we actually don't do super meticulous version control for every jurisdiction's legislation. Some customers care about this, some don't, and we generally only add this feature for smaller countries/jurisdictions when someone asks for it.
@sand1929 Got it! How do you avoid duplicate docs when re-scraping?
> - SHA-256 hashing? Or rely on URL + timestamp?