A Call Quality Assurance Checklist for Insurance Agencies
What to score on every recorded sales call, how to weight it, and how to turn a handful of flagged calls a week into a real coaching habit.
A call quality assurance checklist is a short, written list of what a manager listens for on a recorded sales call and how each item is scored, so that two different reviewers reach the same conclusion about the same call. For an insurance agency, that list needs to cover two different things at once: whether the call met the compliance rules that apply to it, and whether the agent actually sold well. Most agencies have an informal version of this in someone's head. Writing it down, weighting it, and running it consistently is what turns "spot-checking calls" into an actual quality assurance program. Why a written checklist, not just "listen to some calls" For Medicare sales specifically, this is not optional busywork. CMS requires that plans and the third-party marketing organizations (TPMOs) acting on their behalf record all marketing, sales and enrollment calls, including the audio portion of calls made over web-based technology, in their entirety — a requirement spelled out at 42 CFR §§ 422.2274(g)(2)(ii) and 423.2274(g)(2)(ii) and restated in CMS's own 2026 Agent and Broker Training & Testing Guidelines . CMS also expects those recordings, along with other records tied to the sales and enrollment process, to be retained for ten years . ACA and life sales don't carry that specific CMS mandate, but if your agency is already recording those calls for coaching purposes, the same review discipline applies. A recording requirement by itself doesn't improve anything — it just creates an archive nobody opens. A checklist is what turns the archive into a feedback loop: a consistent set of things every call is checked against, a score that means the same thing from one reviewer to the next, and a record of what was found so a pattern across an agent, a product line, or a lead source is visible instead of anecdotal. What a call quality assurance checklist should score Split the checklist into two halves and don't let them blend together: compliance items are pass/fail, because a missed required disclosure isn't a style note, and skill items are scored on a scale, because "good rapport" is a matter of degree. A workable starting set of skill categories, drawn from what most call-coaching platforms (AgencyView's included) already score automatically on every transcribed call, looks like this: Compliance — required disclosures given, consent obtained before moving into plan details, no unapproved claims about coverage or price. Customer sentiment — how the client's tone shifted over the call, not just how the agent sounded. Professionalism — pacing, interruptions, whether the agent stayed in control of a rambling call. Empathy — whether the agent acknowledged what the client said before moving to the next question. Confidence — hesitation or hedging on plan details that should be second nature. Clarity — whether a client with no insurance background could follow the explanation. Those six hold up across Medicare, ACA and life calls because they're about how the call was run, not what was sold. The part that has to change by product line and by agency is everything compliance-specific: a Scope of Appointment captured before a Medicare needs assessment, the required TPMO disclaimer in the first minute of the call, a documented Permission to Contact, an ACA attestation read correctly. We've written up the specific rules for those elsewhere — see our posts on CMS call recording requirements and HIPAA-compliant call recording — so the checklist itself can stay short: name the three or four required items for the call type in front of you, and mark each one met or missed. Building your own scorecard sections The six categories above are a floor, not a ceiling. Most agencies add sections specific to what they coach on — needs discovery, objection handling, cross-sell mentions, whatever a sales manager actually walks the floor correcting. If you're using a platform that supports custom scorecards, each section is really three things: a name, one sentence describing what "good" looks like for that section (specific enough that two reviewers would score the same call the same way), and a weight relative to the other sections. In AgencyView, that description is what the AI scoring pass reads verbatim when it grades the section, which is exactly why a vague description like "good discovery" produces vague, inconsistent scores — write it the way you'd explain it to a new hire: "asked about current coverage, household and budget before recommending a plan." Weight the sections the way you'd weight anything that matters to the business: if plan-fit discovery is the single biggest driver of retention and persistency on your book, it should carry more of the total score than tone. Review the weights after a quarter of scored calls, not before — you won't know which sections actually predict a good outcome until you have data to look at. Using talk time and flagged calls instead of reviewing everything No sales manager has time to fully review every recorded call, and a QA program that requires that collapses within a month. The more sustainable approach is to let the data narrow down which calls get a full listen. Talk-time ratio — how much of the call the agent spoke versus the client — is a cheap, automatically-calculated signal: a call where the agent talked 80% of the time is a call where discovery almost certainly didn't happen, regardless of how it scored elsewhere. We've written separately about why talk time, not dial count, is the metric worth coaching to ; it's also the fastest filter for deciding which calls a manager actually needs to hear. Pair that with an AI-flagged review queue if your platform has one: calls where the scoring pass catches a likely missed disclosure, a compliance keyword that should have appeared and didn't, or a sentiment score that dropped sharply mid-call. That queue is where a manager's limited review time should go first — not a random sample, but the calls most likely to have actually gone wrong. Running the review, not just scoring it A checklist that only produces a score and no action is a reporting exercise. Build the workflow around three habits: Calibrate across reviewers. If more than one person scores calls, have them independently score the same handful of calls monthly and compare. Disagreement on a compliance item means the checklist wording is ambiguous, not that someone is wrong. Close the loop with the agent within days, not weeks. A scorecard an agent sees a month after the call is a historical document. One they see within a few days is still coaching. Track the trend per agent, not just the score per call. One low score is a bad call. The same section scoring low across an agent's last ten calls is a training gap, and it's the pattern — not any single call — that should drive what you coach to next. A sample call quality assurance checklist Category What to check Pass criteria Required disclosures TPMO disclaimer, SOA, Permission to Contact, as applicable to the call type Given verbatim, before plan specifics are discussed Compliance score No unapproved claims, no pressure tactics, consent obtained No flagged compliance moments on the call Needs discovery Current coverage, household, budget, priorities asked before a recommendation All four covered before plan talk begins Clarity Plan details explained in plain language No unexplained jargon a first-time client wouldn't follow Talk-time ratio Share of the call the agent spoke versus the client Agent at or under roughly 60% on a needs-based call Objection handling Client hesitation addressed directly, not talked over Objection acknowledged before the agent responds Next steps Clear statement of what happens next and when Client can repeat back the next step unprompted FAQ How often should an insurance agency review sales calls? Full manual review of…
Where AgencyView fits
Keep reading
Share: Facebook · LinkedIn · X · Email
All articles · AgencyView