Certificates Insurance
With the tool we chose6 min median
By hand42 min median
Requests measured by us621
The founder's agency · July 1 – September 12, 2026 · the tool is a vendor's, the measurement is ours
For independent insurance agency owners
Put AI to work in your independent agency.
The founder runs an independent agency and a life insurance MGA on systems he built and measured there first. We help agency owners do the same: less re-keying, calls actually reviewed, leads followed up, inside the agency system and CRM you already use.
45 minutes on Zoom with Vytas Skapcevicius, the founder. Free, and no pitch deck.
Call review Insurance
Recorded agent calls, a minute or longer2,208
Scored onfive parts
Each onewith a coaching note
The founder's agency · April to August 2026 · since late September, a manager's weekly audit per person
Intake into the agency system Insurance
Forms on the first live run7 of 7
Re-keyed by handnone of the 7
Read back from the systemevery one
The founder's agency · August 2026 · built and verified once, end to end; paused since September
Problems we solve
Where an agency’s time actually goes.
Most agencies do not need more software. They need the systems they already pay for to stop making people do the same work twice.
The same data, typed twice
Quote requests and intake forms re-keyed into the agency system. Certificates and changes handled one by one.
A service inbox that eats the morning
Requests arrive by email, phone and portal, and nobody can see what is waiting or for how long.
Calls nobody hears
Coaching by anecdote. A manager hears a handful of calls a week, so the same problems repeat.
Leads and renewals that slip
Dial lists that never get shorter. Leads called twice or never. Renewals reviewed too late to re-shop.
No view of the week
The owner cannot see where the team’s time goes, or where business leaks out of the book.
AI solutions
Four things we build. Status honest.
Each card shows its status in our own agency: measured, running, built, paused or pilot, or “we build this” for work we would do for you. We never present that last group as results.
Repetitive service work and re-keying
Service and admin automation
Certificate turnaround and rework measured around the tool you use, intake written into your agency system and read back to confirm, and documents prepared for a person to file.
Certificates: MeasuredIntake: Built · paused
Coaching by anecdote
Call review and coaching
Recorded calls transcribed and scored against each person’s own script, with a coaching note drafted for the manager to edit and send. A manager’s weekly audit, not a robot’s verdict.
Weekly audit: Running
Leads that wait, or get called twice
Lead capture and follow-up
Web and ad leads land in your CRM with an owner, a task and an instant alert to the person who calls back. Logged call outcomes move each contact on, so lists get shorter.
Lead routing: RunningOutcome follow-up: Running
A book that shrinks quietly
Renewals, retention and cross-sell
Renewals sorted into re-shop, review, accept or chase before they renew; a weekly board of the book moving by carrier, line and client; a household view with the next policy to offer.
Book board: RunningRenewal worklist: Built · pausedHousehold cross-sell: We build this
See AI in action
Measured in a working agency. Not a demo.
Both come from the founder’s own agency and its own systems. Sourced and dated, never a projection.
We rebuilt 12,108 emails into 621 certificate requests, so we could finally time them.
Median time, request to certificate
Tool we chose6 minutes · n 261
Manual path42 minutes · n 146
Needed rework: 1.1% with the tool, 5.8% manual
The certificate tool is a vendor's; what we built is the measurement. 407 requests had a clear start and finish, and we left the rest out rather than guess. The manual path includes the unusual requests the tool does not handle, so this is not a like-for-like speed test. Read the case study.
2,208 recorded calls scored on five parts, each with a coaching note.
- Recorded call
From the phone system the agency already uses
- Transcript
Calls of a minute or longer, not a hand-picked sample
- Scored
Needs questions, value, objections, the referral ask and initiative, plus one coaching moment
- The manager
Edits the drafted coaching and decides what to send
That was a nightly job, April to August. Since late September a manager runs a weekly audit per person instead, scored against that person’s own script with quotes checked against the transcript, and decides what reaches each agent. We publish the volume, not an improvement: one person’s four weeks is not proof. Read the case study.
Book a free AI discovery callWe look for the problem worth fixing first in your agency.
Real agency experience
Built in our own agency before anyone else’s.
The founder runs an independent P&C agency and a life insurance MGA in Elmhurst, Illinois. Our insurance systems were built there first. Some run every day, some are paused, and the case studies say which.
Call outcomes move contacts through the CRM, so worked leads leave the lists. Intake forms were written into the agency system end to end, 7 of 7 on the first live run; that job is paused while the operations machine is replaced. A weekly board shows the book moving by carrier, line and client. Some of it did not work the first time, and the case studies say so.
We understand agency management systems, carrier portals, producers and service teams, because we work in them every day. See our work.
Our agency also uses an outside AI vendor for its main phone line and its certificates. That vendor publishes its own figures for our agency; we report only what we measured ourselves.
170written playbooks across the founder’s businesses, counted October 2
19cloud services we built and run
70scheduled jobs enabled on October 2; more are paused while the operations machine is replaced
That machine runs our agency. For clients, our standard is that production systems run in accounts they own.
How it works
Discover. Prioritize. Implement. Optimize.
One expensive problem first, measured before and after. Nothing moves to the next step until you say so.
Discover
A free 45-minute call. Where your agency loses time or money, in your words.
You leave knowing whether anything is worth measuring.
Prioritize
We pick the one process worth fixing first, and how we will measure it from your own systems.
A fixed fee for the first fix, agreed before any work starts.
Implement
The baseline measured first, then the fix built inside the software you run and made production-grade.
Your team trained, with written runbooks.
Optimize
Monitoring, maintenance and a monthly report against the baseline.
A monthly plan. The next problem ranked by what it is worth.
For P&C agencies
Add life insurance for the households you already insure.
If your agency does not offer life insurance today: the founder also runs a life insurance agency, and in the states where it is licensed it partners with P&C agencies that want to offer it, including policies with living benefits.
That is a different business, licensed separately. If it interests you, start with a short owner-to-owner call; it can cover both the AI work and the life side.
Questions agency owners ask
Straight answers.
What does Independent Agency AI do?
We help independent insurance agencies put AI to work. We find where your agency loses time or money, measure it from your own systems, and build the AI or automation aimed at the biggest problem, inside the software you already run. A person approves anything that commits your agency, or approves the exact wording of a routine message in advance.
Do I have to replace my agency management system or CRM?
No. We build around the systems you run today, through whatever they allow, from a proper API to a checked export.
What does it cost to start?
The discovery call is free. After it, the first fix is a fixed fee, agreed before any work starts, once we both know what the problem is worth. No part of the fee depends on a promised result.
How fast will I see something?
Each pilot is scoped to show a measurable result in 30–90 days, measured against the baseline taken before the work starts. That is the window we plan for, not a promise.
Do you train our people?
Yes. Every deployment includes training for the people who use it and written runbooks. Ongoing support and improvements are a monthly plan.
Will AI quote, bind or give coverage advice?
AI can run a rate for a licensed person to review. It never presents a price to a client, binds, changes coverage or gives advice: a licensed person does.
Who owns the systems and the data?
You do. Our standard is that production systems run in accounts you own. If something runs on our side while we build it, we tell you, and it moves to your accounts at hand-over.
What if AI is not the answer?
Then we say so. Sometimes a report, a script or a process change does the job for less, and that is what we recommend.
Find the one problem worth fixing first in your agency.
A free 45-minute discovery call on Zoom with Vytas Skapcevicius, the founder. Four questions about your agency, no pitch deck. You leave knowing whether there is something worth measuring.