Why I Built a Legislative Tracking Platform After Five Sessions Under the Gold Dome

By Buddy Costley, Founder of Lobb-E

Five sessions at the desk where the work happens

From 2020 through 2024, I spent five Georgia legislative sessions under the Gold Dome as a policy advocate and lobbyist. Five years of session mornings, crossover-day crunches, hearings stacked on top of each other, and the question every policy professional carries around all session: what changed in that bill overnight, and who does it affect now?

If you’ve done this work, you know the rhythm. A bill moves. An amendment drops. A substitute shows up with struck-through language and new provisions woven into the same document. Somewhere between your tracking spreadsheet, three browser tabs, and the PDF you printed at 7 a.m., you have to figure out what the bill actually says. Then explain it to the people who depend on you, usually by end of day.

The problems I kept running into

1. Tracking was one job. Understanding was another.

Many legislative tracking tools tell you something happened. A bill moved, a status changed. Fine. But status updates and abstract summaries aren’t the work. The work is reading what the bill actually says now, figuring out which parts matter to the people you represent, and being ready to act.

And here’s the part that doesn’t show up on any product page: doing all of that for one bill is a real chunk of an afternoon. Now multiply it. Over a session you might be tracking 30, 40, 50, 60 bills. Some weeks five of them move at once. The tracking tools told me which ones moved. The hardest, most time-consuming part, times sixty, was entirely on me.

2. General-purpose AI wasn’t built for professional policy work.

Like many others, I started to experiment with AI to aid my work and inform my understanding of the bills and policies I cared about. Not to replace me, but to help me get ahead and make more informed closer readings of the actual bills themselves.

Then I discovered, probably like you have, that as amazing as AI can be as an aid for policy work, the average chatbot can lie to you.

It often doesn’t notice that language has been deleted with strikethroughs. Or it will just plain hallucinate and make up facts or apply provisions that shouldn’t be applied. They’ll hand you confident-sounding conclusions about a bill that the text simply doesn’t back up: a fiscal effect that isn’t in the bill, an obligation the bill never imposes.

In the end, I discovered that my chatbot subscription was a pretty good communication tool for recipes and travel itineraries, but it wasn’t up to snuff for the level of grounded accuracy I needed for policy advocacy work. As a small non-profit executive director, we couldn’t afford the big-money legislative power tools, so I just did the best I could.

3. The serious tools came with unnecessary enterprise weight…with huge cost to you and your organization.

There are heavy-duty legislative platforms out there, but most are priced and structured for large government-affairs departments, not the small non-profit organization, the government affairs team, the elected official, or the concerned citizen.

These big-money software platforms have bells and whistles you just don’t need. It’s like buying a spaceship, when all you need is a trusty pickup truck to carry the load of legislative advocacy and policy work.

What I built next

Lobb-E is a legislative tracking platform and legislative workspace built around the workflow I lived for five sessions, at the volume I lived it. It’s a new company and I won’t pretend otherwise. I built it because the tool I wanted didn’t exist at a price a working policy professional could justify.

How it works, step by step:

1. Bring a bill in. Find a bill or upload a legislative PDF straight into the workspace.
2. Clean the text first. Lobb-E’s Redline Cleaner is designed to remove the struck-through, deleted language before AI analysis runs, so your summary is built on what the bill actually says, not on language that’s already been cut.
3. Look at it through a stakeholder lens. This is the part I wish I’d had under the Gold Dome. Pick the group you represent and Lobb-E generates an editable briefing filtered and prioritized for that group. Then switch the lens and run the same bill again from a different perspective. Start with your own people, then look at how the bill may affect other groups that could support it, oppose it, or push changes. Walk in another stakeholder’s shoes. Anticipate their concerns. Prepare your answers, all tied to what the text actually supports.
4. Track it through the session. Follow the bills on your list, bring your team into one shared workspace, and pull tracked bills into multi-bill briefing documents when you need the whole picture in one place.
5. Turn analysis into communication materials. A briefing becomes the stuff your work runs on: member alerts, hearing testimony, talking points, coalition communications, sponsor emails, amendment language, press releases, policy highlights, implementation timelines, slide-deck outlines. All editable so you can correct, guide, and add your own VOICE to the content if you choose to share it.

Being honest about AI and accuracy

Lobb-E includes legislative-specific safeguards designed to reduce hallucinations, unsupported claims, and deleted-language errors. The workflow strips out the deleted language before analysis. These safeguards reduce risk. They do not make AI infallible and they do not replace professional review.

Every briefing is editable, and you should read it before you send anything. That’s not a disclaimer bolted on at the end. It’s how I’d use any tool in this field. Lobb-E supports your judgment; it doesn’t substitute for it.

Try it on a bill you already know

Buddy Costley
Founder, Lobb-E
Former Georgia policy advocate and lobbyist