Most agencies do not decide to add service capacity. They discover they needed it about six months earlier, usually when a renewal goes out late, a certificate holder escalates to the insured, or a good account leaves without a conversation.
The problem is that "we are busy" is not a measurement. Every agency is busy. What separates a busy team from a short-staffed one is whether the work that protects revenue, renewal prep, policy checking, response time, still holds schedule as volume rises.
Below are seven signs you can measure in your agency management system this week, with the operational fix for each. None require a consultant to diagnose, only someone willing to pull the numbers.
Sign 1: Renewal prep is landing inside 30 days
Pull your expiration report out of AMS360, Applied Epic, EZLynx, HawkSoft, QQCatalyst, or NowCerts, then look at the date of the first logged activity on each account. If the first renewal touch on commercial accounts is consistently 30 days or fewer before the effective date, you do not have a renewal process. You have a scramble with a calendar attached.
The reason it matters is sequencing, not effort. Loss runs frequently take carriers five to ten business days to return, and an exposure update from the insured takes a week of chasing. Open the file at day 30 and you present one option under deadline pressure instead of a real marketing effort.
The fix: build a dated renewal pipeline and assign the early stages to a service role rather than the producer.
- Day 120: pull the expiring list, verify named insured, entity type, and contacts.
- Day 105: order loss runs, request payroll, sales, vehicle, and property schedule updates.
- Day 90: complete ACORD applications and supplementals from the updated exposures.
- Day 75: submit through carrier portals, log the submission and underwriter contact.
- Day 45: chase quotes, resolve underwriter questions, build the comparison.
- Day 30: proposal ready for the producer to present.
Days 120 through 45 are almost entirely data gathering, documentation, and follow-up, exactly the block a trained remote insurance CSR can own end to end, leaving the producer the presentation and the relationship.
Sign 2: Certificate turnaround is measured in days
Certificates of insurance are the clearest diagnostic in the agency because the clock is unambiguous: request received, certificate delivered. Separate the volume by type first, because the categories carry different realistic targets.
- Standard wording, existing holder: reissue off a stored template. Minutes of work.
- Standard wording, new holder: verify holder name and address, issue, log to the account.
- Contract-driven wording: additional insured, waiver of subrogation, primary and non-contributory. Needs a carrier endorsement, so the calendar is partly outside your control.
- Master lists and renewal reissues: bulk work to schedule, not handle reactively at renewal.
A reasonable standard is standard certificates delivered inside four business hours, and endorsement-driven requests acknowledged in the same window with a stated expected date. If standard certificates take a day or more, the delay is intake and queueing, not underwriting.
The fix: one intake channel, a template library keyed to common contract language, and a named queue owner with an escalation path for anything needing a carrier endorsement. Certificate work rewards process discipline more than seniority, which is why it is usually the first workflow agencies hand off.
Sign 3: Your producers are doing service work
Pull a week of logged AMS activity for your two best producers and classify it. Service work landing on a producer is the most expensive labor mismatch in an agency: commission-level compensation for tasks a trained service professional does better, because they do them fifty times a day.
The table below is illustrative, a template for your own numbers.
| Service task pulled onto a producer | Illustrative hours per week | Who should own it |
|---|---|---|
| Certificate requests and holder updates | 2.0 | Service role |
| Endorsement requests and carrier follow-up | 1.5 | Service role |
| Renewal data gathering, loss runs, applications | 3.0 | Service role |
| Billing questions and cancellation notices | 1.5 | Service role |
| Claims intake and FNOL handoff | 1.0 | Service role, exceptions escalated |
| Total | 9.0 |
On a 45-hour week, nine hours is roughly 20 percent of a producer's capacity spent on work that does not require a licensed sales professional. Run the arithmetic with your own logged hours and close rate before drawing conclusions about what it costs.
Sign 4: Calls and emails sit past one business day
Take one week of the service inbox and voicemail log and count items with no outbound response within one business day. Above roughly 5 percent of weekly volume, you have a queue problem, and queue problems are structural rather than motivational.
The usual root cause is that requests arrive in personal inboxes and personal voicemail. When one person is out, at a carrier meeting, or simply underwater, those requests are invisible to everyone else. Nobody drops the ball on purpose; the ball is somewhere no one else can see it.
The fix: route service requests to a shared address and shared queue, set a same-day acknowledgment standard separate from your resolution standard, and require every request to create an activity or suspense record on the account so anyone can pick it up cold. A dedicated remote CSR helps here for a simple reason: their whole day is the queue, so the queue has an owner during business hours.
Sign 5: Retention is drifting and no one can explain why
A retention slide of one or two points a year is easy to rationalize: rates moved, appetite changed, one large account left. Before accepting that, run cancellations for the trailing 12 months and check each lost account for a logged proactive service touch in the prior year. Silence is a leading indicator of departure.
A worked example, assumptions stated plainly: an 800-account book averaging 1,000 dollars of annual commission per account. A two-point slip is 16 accounts and roughly 16,000 dollars of recurring revenue, plus the new-business capacity spent replacing it. Those figures are placeholders showing the shape of the math, not a measured result. Substitute your own numbers.
Retention is not a sales metric that service happens to influence. It is a service metric that sales gets credited for. If nobody owns the mid-term touch, nobody owns retention.
The fix: a scheduled mid-term contact on every account above a revenue threshold you set, logged in the AMS with a documented outcome. It is calendar work, not talent work, which is why it is the first thing to disappear when a service team is short.
Sign 6: Policy checking has quietly stopped
Ask what share of new business policies were checked against the application, quote, and binder last quarter. If the answer is a shrug, this is the sign with the sharpest tail risk: an unchecked discrepancy does not surface at issuance. It surfaces at a claim.
A workable policy check covers, at minimum:
- Named insured spelling, entity type, and additional named insureds.
- Mailing address and every location or vehicle schedule entry.
- Limits, sublimits, deductibles, and coinsurance against the quote.
- Forms and endorsement list, including anything the carrier added unrequested.
- Additional insureds, waivers, mortgagees, and loss payees.
- Effective and expiration dates, and premium against the quoted figure.
- Confirmation that mid-term endorsements were issued as requested.
The fix: check 100 percent of new business and every renewal with a material change, then sample clean renewals. Attach the checklist to the account and log the check as an activity so the work is provable later. This is repetitive, documentation-heavy work with clear right answers, well suited to a trained service professional following a written SOP.
Sign 7: The owner is clearing the service queue at night
Sort AMS activity by user and timestamp. If the owner is logging service activity after 6 p.m. and on weekends, week after week, the agency is not short on hustle. It is running on a subsidy with one person's name on it.
The real cost is not the hours. It is what those hours displace: carrier relationships, hiring and training, acquisition conversations, the work only an owner can do. Service work expands to fill whoever will absorb it, and an owner absorbing it looks like resilience right up to the point it becomes the ceiling on growth.
Where to start
Diagnose before you staff. The first four steps cost nothing but half a day of an operations manager's time.
- Pull 30 days of data: expiration report with first-activity dates, certificate request-to-delivery timestamps, service inbox and voicemail response times, and AMS activity by user and hour.
- Score yourself against thresholds you set in advance, so you are not grading on a curve after seeing the numbers.
- Rank the workflows consuming the most hours against the judgment they require. High volume plus low judgment is where you hand off first: certificates, renewal data gathering, policy checking.
- Write the top three as real SOPs, covering system access, naming conventions, documentation standards, and escalation rules. A workflow that cannot be written down cannot be delegated to anyone, remote or in-house.
- Staff the smallest version that tests the theory: one dedicated, trained person owning those workflows inside your own systems, with least-privilege access, a weekly sample quality review, and the same numbers re-measured at 60 and 90 days.
Use this scorecard for step two:
| Signal | How to measure it | Threshold worth acting on |
|---|---|---|
| Renewal lead time | First logged renewal activity to effective date, by line | Median under 45 days, commercial |
| Standard COI turnaround | Request to delivery, standard wording only | Median over 4 business hours |
| Producer service load | Service-coded activity as a share of logged hours | Above 15 percent |
| First response time | Requests with no reply within one business day | Above 5 percent of weekly volume |
| Retention | Policy and account retention, trailing 12 months | Two straight years of decline |
| Policy checking | Share of new business with a completed, logged check | Under 100 percent |
| After-hours owner activity | Owner activity logged after 6 p.m. or weekends | Recurring every week |
One sign on this list is a workload spike. Three or more, sustained across a quarter, is a capacity problem, and capacity problems do not resolve by asking the team to try harder. Start with one dedicated service professional owning documented workflows, re-measure in 90 days, and scale from evidence rather than from the feeling of being busy.