The 90-Day Renewal Radar Playbook for Independent Agencies

The carrier benchmarks are unkind about this number. Vertafore's 2024 retention study put average P&C renewal retention at roughly 84%. Reagan Consulting's Organic Growth and Profitability survey puts the top quartile at 91-93% and the bottom quartile under 80%. MarshBerry's APPEX agency benchmarking shows independent agencies losing 12-18% of book annually, with the bulk of the loss concentrated in personal lines and small commercial — the segments where the producer-to-account ratio is highest and the renewal touch is thinnest.

Round it off and you get the number every principal already feels in their gut: roughly 18% of accounts walk away at renewal. Most of those exits were detectable 60 to 90 days out. The carrier saw the rate change coming. The competitor saw the X-date in the state filing. The insured saw the premium letter. The only party that did not see it was the producer of record, because their renewal queue is sorted by expiry date and their week is sorted by whoever shouted loudest that morning.

This is a playbook. Not a pitch. The sequence below is what we run inside Stanton Insurance Agency and what the Renewal Radar tool automates for agencies that do not have a CSR with eight hours a week to spend on it. Read the sequence first. The tool comes at the end.

What the carrier sees that you do not

This is the part that bothers principals the most when we walk through it. The carrier underwriting desk has data on your book that does not flow back to the agency in any usable form. They see loss-frequency drift across NAICS classes a quarter before you do. They see payment-pattern changes at the policyholder level. They see X-date competitive activity from BOR requests their underwriter teams field weekly. They see appetite shifts at the company level that get communicated to the carrier rep as "we're tightening on contractors over $2M revenue" and to the agency as silence until the non-renewal letter shows up.

Closing that gap is the entire job. You do not need parity with the carrier's data — you cannot get it. You need enough early signal that a producer can have a real conversation 60 days before the renewal instead of a reactive one seven days after the rate increase letter lands.

The 90-day sequence, milestone by milestone

Pin this to the wall. The sequence is staff-agnostic. A CSR can run it. A producer can run it. An AI workflow can run the parts that are mechanical and hand off the parts that are not.

T-90 — pull the cohort, score every account

The renewal cohort is every account with an effective date 90 days out. Most AMS systems — Applied Epic, AMS360, Vertafore, EZLynx, HawkSoft, NowCerts — will produce this report natively. The score is what most agencies skip.

Score each account on five signals.

• Premium delta. What is the carrier's filed rate change for the class? Anything above 8% is a tier-one conversation. Anything above 15% triggers a remarket evaluation.
• Claims activity. Any claim in the trailing 24 months changes the renewal posture. Any open claim makes it a producer call, not a CSR call.
• Carrier appetite signal. Has your carrier rep mentioned the class tightening? Has the underwriter desk slowed quote turnaround on the segment? Has the reinsurance treaty changed?
• Account profitability to your agency. Contingent-eligible accounts get extra protection. Sub-$2,000 personal-lines monoline accounts get a different conversation.
• Producer relationship age. Accounts under 18 months are statistically the most likely to leave on price. Accounts over five years leave on service.

The output is a tiered list. Roughly 20% of the cohort gets flagged. That 20% is where the producer's renewal-week attention goes. The other 80% gets the standard letter-and-followup workflow that most agencies are already running.

The reason agencies do not score this way is that the math is tedious. Pulling carrier filings, cross-referencing loss runs, checking contingent eligibility — for an agency with 340 renewals in a quarter, that is roughly four CSR-hours per account if it is done manually. Four hours at $45/hr fully loaded is $180 per account. On 340 accounts, that is $61,200 a quarter — which is why nobody does it.

T-60 — producer touch on flagged accounts

The 20% of flagged accounts get a producer call, not a CSR letter. The call is not a soft check-in. It has a structure.

• Exposure update. What has changed in the operation since last renewal? New vehicles, new employees, new locations, new revenue, new contractors. Most of these never make it back to the carrier without a producer prompt.
• Carrier appetite check. If the producer already knows the class is tightening, this is when they raise it. "Arbella has been adjusting on contractors this year — I want to make sure we are in front of any change before the renewal lands."
• Competitive scan. "Are you getting quoted by anyone else? Anyone calling you on this?" If the answer is yes, the producer learns now instead of at T-7.
• Service review. Anything broken? Any claim experience that left a bad taste? Any coverage question the insured has been sitting on?

The T-60 touch is the single highest-leverage hour in the renewal cycle. Big I's Best Practices study has shown for years that producer-touched renewals retain at 92-95% rates, while CSR-letter-only renewals retain at 78-83%. The delta is roughly 12 points of retention on the flagged cohort. On a $4M-revenue agency, that translates to roughly $96K of preserved annual revenue. The math survives almost any test you put against it.

The reason this does not happen at most agencies is not that producers do not know to do it. It is that producers do not know which 20% to call. They get a list of 340 renewals and either call the top 20 by premium (wrong cohort) or none of them (more common).

T-30 — renewal proposal in hand, remarket if needed

By T-30 the producer should have the renewal carrier's terms in writing. If the proposed rate change exceeds the threshold from T-90 (we use 12% as the line), the producer triggers a remarket.

The remarket is not a panic move. It is a calibration. Three things have to be true to actually pull the trigger.

• The premium delta is large enough that the insured will notice and ask.
• A meaningful alternative carrier exists in the class. If your top three appointed carriers are all tightening the same NAICS, remarketing burns goodwill and produces nothing.
• The producer has the time to walk the insured through the trade-off. A remarket that hands the insured three quotes and says "pick" produces churn, not retention.

The T-30 milestone is where most of the saves actually happen. The conversations were teed up at T-60. The proposal is in hand. The remarket option is real. The producer is having the right conversation with the right insured at the right moment.

This is also where Renewal Radar does its second-order work — automating the proposal generation, the side-by-side coverage comparison, and the carrier-appetite check across the agency's appointed carrier panel. The producer's job is the conversation. Everything else is mechanical.

T-7 — verbal confirmation, BOR risk neutralized

A week before renewal, the producer gets verbal confirmation. Not email. Not portal click-through. A phone call where the insured says "yes, we are renewing." That call closes the loop and surfaces any last-second issue (BOR request from a competitor, financing change, sale of the business) before it becomes a non-renewal.

The other T-7 task is the billing handoff. Half the lost renewals we see in the post-mortem are not coverage problems or rate problems — they are billing problems. The insured's payment method expired. The agency-bill installment plan was not communicated. The direct-bill carrier sent the wrong amount because exposures changed mid-term. A clean billing handoff at T-7 prevents the cancellation-for-non-payment that gets logged as a retention miss when it was really an operational miss.

T-0 — bound, logged, post-mortem on misses

Renewal day. Bound. Logged in the AMS. The misses — the 5-8% of the flagged cohort that walked despite the sequence — get a post-mortem entry. The post-mortem is one paragraph. Why did this one leave? Was the rate uncompetitive? Did the producer not get the T-60 call done? Was there a service issue we missed? Was the BOR a friend-of-a-friend?

The post-mortem feeds back into the scoring model at T-90 of the next cohort. This is the part agencies almost never do, because there is no time to do it during renewal season, and after renewal season nobody wants to look at it. The agencies that do it methodically — Reagan Consulting's top-quartile cohort — are the ones at 91-93% retention. The connection is not mystical.

The cost stack — manual workflow vs AI workflow

Walk the numbers honestly.

Manual workflow. For an agency with 340 renewals per quarter and the 20% flagging discipline above, you are touching 68 accounts per quarter at roughly 4 CSR-hours each. At $45/hour fully loaded that is $12,240 per quarter, $48,960 annually. That number assumes a CSR who already knows how to do the scoring — most agencies need 3-6 months of training and SOP work to get there, which is another $15-20K of opportunity cost in year one.

AI workflow. Renewal Radar runs the T-90 scoring across the full cohort, automates the carrier appetite check, generates the producer brief for each flagged account, and produces the T-30 proposal package. Pricing is $79/month for the first 50 accounts and scales linearly from there — call it $540/month for a 340-renewal-per-quarter agency, or $6,480 annually. The producer still makes the calls. The CSR still handles the operational handoffs. The mechanical work — the four-hours-per-account that nobody actually wants to do — is gone.

The annualized delta is roughly $42K saved on workflow cost. The retention delta on the flagged cohort, if the producer touches happen at the rate the Big I data suggests, is somewhere between $60K and $120K of preserved book revenue at a $4M-revenue agency. The combined annual return is on the order of $100K-$160K against a $6,480 software spend.

What the sequence does not fix

Three things, honestly.

• Rate environment. If the carriers in your appointed panel are all taking 18% in the class, you are going to lose some accounts on price. The sequence reduces the percentage. It does not eliminate it.
• Claims-driven churn. An insured who had a bad claims experience is leaving regardless of what the producer says at T-60. The fix for that is upstream — FNOL triage, claims advocacy, communication during the claim.
• Producer turnover. If a producer leaves and takes their book with them, no sequence saves you. The fix is structural — book ownership, non-compete, agency-of-record discipline.

The sequence covers the operational layer. The operational layer is where the 18% walk-away rate hides. Three to five points of recovery against that number is realistic. That is the dollar value.

Where to start this week

Pull last quarter's renewal cohort from your AMS. Sort it by retention outcome. For every account that walked, write a one-sentence post-mortem. By the time you are done you will see two or three patterns repeat — usually rate-driven on a specific class, service-driven on a specific producer's book, or billing-driven on a specific carrier's direct-bill setup. Those patterns are the scoring weights for your next cohort.

Then run the next cohort through Renewal Radar at T-90. Free tier covers the first 50 accounts. Score, flag, hand the producer the call list. If the call list produces three saves the producer would have missed, the tool has paid for itself for the year before you finish the next pricing review.

The 18% walk-away rate is not a law of nature. It is the gap between the data the carrier has at T-90 and the conversation the producer has at T-7. Close the gap.