AI FOR LOBBYING FIRMS

AI for lobbying firms: more capacity for every client, and room to grow the book.

For partners and managing directors: what AI changes in the economics of running a lobbying firm, a capacity model you can run with your own numbers, and the measures that tell you whether it is working.

IN SHORT

AI changes how many hours each client costs the firm to serve, not what the firm sells. When monitoring, meeting prep, research, first drafts, reports and follow-up are prepared for a person to review, every lobbyist gets hours back each week. The firm can spend them on deeper service for the clients it has or on more clients at the same standard, and partners can see the whole book without asking for status.

THE CONSTRAINT

A lobbying firm sells judgment and access. It is limited by hours.

The economics of a lobbying firm are simple to describe and hard to change. Revenue comes from client engagements, usually retainers. The work is delivered by senior people whose time is the scarce input. And every client carries a load of work that is not the meeting, the call or the strategy session the client is paying for.

That load is the back office: reading what moved overnight and deciding which clients it touches, assembling a brief before a meeting, researching a member’s record, drafting the one-pager and the request, keeping submission deadlines, writing the monthly report, sending the follow-up, and recording what happened so the next person knows. It grows with the number of clients, not with the number of meetings that went well.

That is why AI for lobbying firms is worth judging at the level of the firm. It does not change what clients buy. It changes how many back-office hours a person spends assembling instead of reviewing, and so how much attention each lobbyist has left for the work only they can do.

MonitoringMeeting prepResearchDeliverablesSubmissionsClient reportsFollow-upRecords
THE CAPACITY MODEL

Run your firm’s numbers.

Enter your own team and book. The model assumes prepared work still needs a person to read, check and edit it, so review time is never free. It shows two ways to use what comes back: more client-facing time for every client you have, or room to take on more clients at the same standard.

An interactive model. You set the number of lobbyists, active clients, weekly working hours, back-office hours per client, the share of back-office work AI prepares for review and the time review takes. It calculates the hours returned to the team each week, client-facing hours per client, and how many clients the team could serve at today’s standard. Two charts show one client’s week before and after, and client capacity as the prepared share rises.

ILLUSTRATIVE DEFAULTSThese starting numbers are an example, not industry data. Change every one to match your firm.
30.2hours returned to the team each week
5hours per lobbyist, per week
9 → 10.7client-facing hours per client, if the time goes back to the book you have
18 → 20clients the team could serve at today’s standard, if the time goes to growth

How the model works: team hours are lobbyists times weekly hours. Each client today gets team hours divided by clients, split into client-facing time and back-office time. Prepared work still costs review time, so the hours returned per client are back-office hours times the prepared share times what review does not use. Capacity is team hours divided by the new hours each client needs at the same client-facing standard.

THE LEDGER

Where AI shows up in a lobbying firm’s economics.

The model counts hours. The bigger effects sit on lines a spreadsheet does not show well: how many clients the team can serve well, what partners can see, how consistent service is across the book, and what the firm keeps when someone leaves.

01CAPACITY PER LOBBYIST

Capacity is attention per client, not clients per person.

Firms often ask how many clients a lobbyist can carry. The more useful question is how much attention each client gets, and how much of a lobbyist’s week goes to assembling things rather than deciding things. Two lobbyists with the same book can deliver very different service depending on how much of the week disappears into prep and paperwork.

AI changes the mix rather than the job. When the meeting brief is already assembled from the client record, the bill text is already compared against the client’s request, and the follow-up is drafted from the meeting notes, the lobbyist’s time shifts toward reading, correcting and deciding. That is still work, which is why the model charges for review. It is the right work for a senior person.

The test for how many clients a team can serve well is not a ratio. It is whether every client still gets a brief before every meeting, a response when something moves, and a report on the rhythm you promised, in a markup week as well as a quiet one.

02PARTNER VISIBILITY

Partners should see the whole book without a status meeting.

In most firms, the partner view of the book is assembled by asking. A weekly meeting goes around the table, each account lead reports on their clients, and whatever does not come up is assumed to be fine. The clients that need attention are often the ones nobody mentions: quiet for weeks, with commitments coming due.

When meetings, commitments and deadlines are recorded against each client as the work happens, the partner view becomes a reading of the record instead of a round of questions. Partners can see which clients have gone quiet, what is due, who owns it and where the team is stretched, then spend the meeting on decisions. Switch the view below to see the difference.

A chart of twelve sample clients plotted by days since the last substantive touch and commitments due in the next fourteen days. In the status meeting view only six clients appear. In the shared record view all twelve appear, including two clients that have gone quiet while commitments come due.

03SERVICE CONSISTENCY

Every client gets the standard, not just the loudest ones.

Service in a firm tends to track the busiest person on the account. When a lobbyist has a heavy week, the smaller clients get thinner briefs, later follow-ups and shorter reports. Clients notice, and renewal conversations often start there.

AI makes a written service standard cheaper to keep, because the baseline work is prepared the same way for every client. A firm can decide what every client gets and hold to it:

A brief before every meetingWho you are meeting, what was said last time, and what the client needs from this one.
A read on what movedGovernment activity checked against each client’s priorities, with the source attached.
Follow-up while it is freshCommitments from each meeting captured and drafted for a person to send.
Reports on the promised rhythmMonthly or quarterly updates drafted from the record and reviewed before they go out.

Nothing on that list leaves the firm without a person approving it. The point is that the first draft is never the reason a client waited.

04INSTITUTIONAL MEMORY

When someone leaves, the client knowledge should stay.

Turnover is one of the most expensive events in a firm’s year, and much of the cost is invisible. When an account lead leaves, the history of the account often leaves with them: which offices have been helpful, what was promised to whom, why the client stopped pushing a provision two sessions ago. The successor rebuilds it by asking the client, the one person the firm least wants doing that work.

Keeping meetings, commitments, drafts and decisions with the client record turns a departure into a handoff. The successor reads the history, sees open commitments and walks into the next meeting with context. It also lowers a risk partners rarely say out loud: a relationship that lives in the firm’s record is less likely to walk out the door with one person.

A conceptual chart, not a measurement. When an account lead leaves in week eight, client context kept in inboxes drops sharply and recovers slowly, while context kept with the client record dips briefly and returns.

WHAT TO MEASURE

What to measure after a lobbying firm adopts AI.

Measure outcomes the firm already cares about, from information it already has. Take a month of baseline before changing anything, then compare the same measures each quarter. If a number does not move, the workflow behind it needs attention, not more software.

Measure
How to read it
Direction
Back-office hours per client, per week
Estimate it honestly at the start, then track it the same way. It is the number the model above runs on.
Down
Client-facing hours per client
Meetings, calls and strategy time logged against each client.
Up, or steady while the book grows
Meetings with a brief prepared
The share of client meetings that started from a written brief.
Toward every meeting
Time from meeting to follow-up
Days between a meeting and the note or materials promised in it.
Shorter
Deadlines met without a scramble
Submissions, request deadlines and reports delivered on time, without late nights to get there.
Toward all of them
Clients quiet for too long
Clients with no substantive touch inside a window the firm chooses.
Fewer
Edits to prepared drafts
How much reviewers change. Heavy edits usually mean the client context is thin, not that review should stop.
Falling over time
Ramp time for a new account lead
How long until a new teammate runs a client meeting without a senior person in the room.
Shorter
WHERE CAPIRO FITS

Priced the way a firm earns.

Capiro is the AI back office for lobbyists. It keeps one record per client, reads government activity against each client’s priorities, and prepares briefs, drafts, reports and follow-ups for your team to review. Meri, the AI Chief of Staff inside the back office, keeps the work moving across the book, and every finding carries its source.

The pricing follows the economics on this page. Capiro is priced by client profile, so cost tracks the client book rather than headcount. Seats are unlimited, so the whole team works from the same record, and there is no separate AI usage fee to ration. AI should not replace the relationship. It should carry the work around it.

QUESTIONS PEOPLE ASK

Frequently asked questions

How does AI change the economics of a lobbying firm?

It lowers the hours each client costs to serve without changing what the firm sells. Monitoring, meeting prep, research, first drafts, reports and follow-up are prepared for a person to review, so lobbyists spend more of the week on judgment and relationships. The firm can spend the time on deeper service for the clients it has, or on more clients at the same standard.

Will AI mean a lobbying firm needs fewer lobbyists?

That is the wrong frame. Clients pay for judgment, relationships and access, and AI provides none of those. The practical effect is more capacity per lobbyist: more attention for every client, more room to grow the book, and less senior time spent assembling materials.

How many clients can one lobbyist handle?

There is no reliable single number. It depends on the depth of each engagement, the issue calendar and how much back-office work each client carries. A better test is whether every client still gets a brief before every meeting, timely follow-up and reports on schedule in a busy week. The model on this page lets you try it with your own numbers.

How can partners see what is happening across every client?

Record meetings, commitments and deadlines against each client as the work happens, then read the book from that record: which clients have gone quiet, what is due, and who owns it. That replaces most of the weekly status round and leaves partner time for decisions.

What should we measure to know whether AI is working in our firm?

Back-office hours per client, client-facing hours per client, the share of meetings with a brief prepared, time from meeting to follow-up, deadlines met without a scramble, clients quiet for too long, how much reviewers edit prepared drafts, and ramp time for a new account lead. Take a month of baseline before changing anything.

Should a lobbying firm tell clients it uses AI?

That is a decision for the firm and its engagement terms. What clients care about is that the work is accurate and the relationship is still with the people they hired. In Capiro, nothing leaves the firm without a person approving it, and findings carry their sources, so the firm can stand behind every deliverable.

More capacity for every client.

Capiro is the AI back office for lobbyists. It prepares the monitoring, briefs, drafts, reports and follow-up for your team to review, so every lobbyist has more time for the clients and relationships the firm is built on.

See the product

The category, explained: What is an AI back office for lobbyists?