How Insurance Agencies Increase Sales With CRM Follow-Up and Call Coaching
Agencies grow phone sales when every lead gets worked on a fixed schedule and every agent gets coached on the numbers where their calls break down. The CRM supplies the schedule and the record. Call scoring and manager feedback supply the improvement. Neither works alone: follow-up without coaching repeats the same weak calls more often, and coaching without follow-up polishes conversations with people you never reach.
This guide is for US agency owners, sales managers and licensed agents running a call floor. It covers how to set up lead stages and follow-up in your CRM, how to build a coaching loop that agents will use, which ratios to watch, and which calling rules shape how you automate any of it.
TL;DR
- Set up CRM stages and a written follow-up cadence so no lead sits untouched. Speed to first call matters most, and every lead needs a next step with a date.
- Segment by lead source (live transfer, purchased web lead, aged lead, referral) and compare close rates by source before you shift budget.
- Coach the ratio where each agent breaks: contact rate, quote rate or close rate. Do not coach only on premium totals.
- Score a sample of calls every week against a short rubric and turn recurring objections into the next training session.
- Keep consent, Do Not Call and calling-hours checks in the workflow before the dial, not after.
Table of Contents
- Why follow-up and coaching have to work together
- Build the CRM follow-up system
- Build the call coaching loop
- Metrics worth watching
- Calling rules that shape your automation
- Perspective
- What it costs to run this on RevRing.com
- A 30-day rollout plan
- Common mistakes
- FAQ
- Sources
Why follow-up and coaching have to work together
Most agencies have two leaks. The first is leads that never get a real conversation: a new lead arrives, one agent tries once, and the record goes quiet. The second is conversations that happen but do not convert, because the agent skips needs analysis, rushes the quote or folds on the first objection.
A CRM fixes the first leak. It holds the lead, the attempt history, the notes and the next task. Coaching fixes the second. It gives managers evidence from actual calls instead of impressions.
The link between them is data. If your CRM records attempts, contacts, quotes and sales by agent and by lead source, you can tell which leak is bigger on any given week. That tells you whether to spend the next Friday on cadence discipline or on call review.
Build the CRM follow-up system
Define pipeline stages that match the phone workflow
Keep stages few and unambiguous. A workable set for a phone team:
- New lead: nothing attempted yet. This stage should be short, and the CRM should show how long each lead has sat in it.
- Contact attempted: at least one call or text, no conversation yet.
- Contacted: a live conversation happened, with a documented next step.
- Quoted: a quote went out or was presented.
- Application submitted: the client signed or the app is in underwriting.
- Policy issued: the sale is done and the record moves to service and renewal.
- Closed lost: with a required reason code.
- Nurture: not now, with a future date to reopen.
The rule that makes stages useful is that every lead in an active stage has a dated next action. A lead with no task is a lead you have lost without knowing it. Add stage-time alerts, for example when a quote has sat untouched for a week, and route them to the agent and the manager.
Route leads to the right agent immediately
Speed collapses when a lead waits in a shared queue. Decide routing rules once and let the system apply them:
- Licensing and state: a lead in a state where the agent is not licensed or appointed should never reach that agent.
- Product specialty: Medicare, health, auto, home, commercial. Send each to the agent who sells it best.
- Availability: skip agents who are on a call or out of the office.
- Distribution method: round robin for fairness, weighted distribution to favor top closers, or skill-based routing for complex products.
If you buy live transfers or shared leads from multiple vendors, the same logic applies at the buying side. Our guide on setting up a ping tree walks through how multi-buyer routing is structured.
Tag every lead by source and type
Put source, vendor, cost and lead type on the record at the moment of entry. Live transfers, shared web leads, exclusive web leads, aged leads and referrals behave differently, and averaging them together hides which ones pay. If an agent edits the source later, you lose your return-on-spend data. Make the field required and lock it after creation.
Set a written follow-up cadence
The first call is the most valuable touch, so it needs a target time. Pick a number your team can actually hit, put it in the CRM as an alert, and review misses weekly. After that, a cadence gives the team permission to stop guessing.
Here is a starting cadence for a new web lead. Treat it as a template to test against your own contact rates, not a law of nature:
| Touch | Timing | Channel | Purpose |
|---|---|---|---|
| 1 | Within minutes of lead arrival | Call | Reach while the request is fresh |
| 2 | Same hour if no answer | Call | Second attempt, different line or time |
| 3 | Same day | Text (with consent) | Short intro and a question that is easy to answer |
| 4 | Next morning | Call | Different time of day than touch 1 |
| 5 | Day 2 | Quote summary or what you need to finish | |
| 6 | Day 4 | Call | Reference earlier attempt |
| 7 | Day 7 | Text (with consent) | Check-in with a specific offer to help |
| 8 | Days 10 to 14 | Call or email | Close-out message, then move to nurture |
A few principles keep it from feeling like spam:
- Every touch has a reason. “Following up” is not a reason. “I have your quote ready, and I need your current deductible” is.
- Vary channels and call times. A prospect who never answers at 10 a.m. may pick up at 5:30 p.m.
- Pause the sequence the moment the prospect replies, and move them to a booking, quoting or close-out path.
- Write templates in short, plain sentences, then let agents edit the first line so it sounds like them.
An automation that sends texts and emails on schedule and creates call tasks for the rest keeps the cadence consistent without making agents memorize it. For the broader setup, see our post on insurance sales playbooks.
Keep the CRM you have, or run one platform
A new CRM is not always the answer. If your team already lives in a CRM and the issue is that dialing, texting and call data sit in separate tools, connecting them can solve the follow-up problem faster than a migration. We cover what to look for in our write-up on insurance CRM integration. The test is simple: when an agent finishes a call, does the outcome land on the lead record without typing? If not, your cadence data will be unreliable.
Build the call coaching loop
Start with a scorecard, not a feeling
Managers often coach on whichever call they happened to overhear. A scorecard makes review consistent. Keep it to a handful of items an agent can fix:
- Opening: identified self and agency, stated the reason for the call.
- Discovery: asked about current coverage, needs and timeline before quoting.
- Quote presentation: explained the options and why one fits.
- Objection handling: acknowledged, asked a question, then responded.
- Next step: ended with a booked time, an application or a dated follow-up.
- Compliance items: required disclosures made, consent and opt-out requests handled.
Score each item as yes, partial or no. Resist rubrics with twenty lines. Agents remember three things from a coaching session, not fifteen.
Coach the ratio where performance breaks
Premium and sales totals tell you who is winning, not why anyone is losing. The more useful approach, laid out in The Insurance Dudes’ agent scorecard guidance, is to track contact attempts, contact rate, quote rate, close rate and average premium for each producer weekly, then coach the stage where the numbers slip. Their breakdown is a good rule of thumb:
| Weak ratio | Likely cause | Coaching focus |
|---|---|---|
| Contact rate | Timing, list quality, single-channel outreach | Call windows, cadence adherence, adding text and email |
| Quote rate | Weak needs analysis or value framing | Discovery questions, building the reason to quote |
| Close rate | Objection handling or closing | Role-play on top objections, asking for the decision |
This keeps the conversation honest. An agent with a low contact rate does not need a closing seminar. An agent with a strong quote rate and weak close rate does not need a lecture on dial discipline.
Turn objections into the curriculum
Have agents log the objection and the next follow-up date on every quote. After a month, sort by frequency. The three most common objections become your next three team trainings, built from your own calls rather than a generic script. Pull two or three real recordings for each, one that handled it well and one that did not, and let the team compare.
Use AI scoring to widen the sample
Managers realistically review a small fraction of calls. AI call scoring can grade every call against your rubric and flag the ones worth a human listen: missed disclosures, long silences, no next step, a strong objection response worth sharing. It does not replace the manager. It decides where the manager’s thirty minutes go. For a closer look at how to design the rubric, see our post on AI QA scorecards, and for the wider case that automation pays back in agency work, this overview of AI productivity gains for agencies is worth a read.
Run a weekly coaching rhythm
- Monday: pull the scorecard numbers per agent and per lead source.
- Tuesday to Thursday: one 20-minute one-on-one per agent, with two calls reviewed and one skill to practice.
- Friday: a 30-minute team session built from the week’s most common objection.
- Following Monday: check whether the targeted ratio moved.
Small, repeated, specific feedback beats a quarterly training day.
Metrics worth watching
A short list beats a dashboard no one reads:
- Speed to first attempt, by lead source and by agent.
- Attempts per lead before a lead is closed out or moved to nurture.
- Contact rate (live conversations divided by leads worked).
- Quote rate (quotes divided by contacts).
- Close rate (policies divided by quotes), by agent and by source.
- Cost per lead and cost per sale, by source and vendor.
- Average premium, so a high close rate on small policies does not mislead you.
- Leads sitting in a stage past your limit.
Compare close rates by source before you move spend. A live transfer, a shared web lead and an aged list should each be judged against their own cost and their own history, not against each other. Use your own numbers as the benchmark. Published close-rate ranges vary widely by vendor, product and state, so they are a poor target for your floor.
Calling rules that shape your automation
Follow-up automation touches regulated activity. What follows is general information, not legal advice. Have your compliance counsel or E&O carrier review your workflow, and check your state’s rules, which can be stricter than federal ones.
Key points the workflow should account for:
- Calling hours. Federal telemarketing rules limit sales calls to between 8 a.m. and 9 p.m. in the called party’s local time, as summarized from the FCC’s TCPA rules. Several states set narrower windows. Build time-zone logic into your dial lists.
- Consent for autodialed and prerecorded marketing. The same summary describes the FCC’s requirement for prior express written consent for telemarketing robocalls, and consent for autodialed calls to cell phones. Keep the consent record attached to the lead so you can produce it later.
- Do Not Call. The FTC’s Telemarketing Sales Rule guidance covers the National Do Not Call Registry, limits on when telemarketers may call, caller ID and abandoned calls. Scrub lists against the registry and keep an internal do-not-call list that your texting and email tools also respect.
- Texting. A text sequence is only as safe as the consent behind it. Our post on consent management for texting shows how to store and check opt-ins.
- Recording. Recording laws differ by state, and some require all-party consent. Use a recording notice at the start of calls and confirm your states’ rules with counsel.
Our telesales compliance checklist is a practical starting point for a review. Tools can run the checks before each dial, but your agency still owns the policy and the results.
Perspective
We see the same pattern in agencies that stall. The leads are fine and the agents are capable, but the follow-up lives in someone’s head and the coaching lives in a once-a-month meeting. The fix is not another tool for each task. It is one place where the lead record, the dial, the text, the call score and the compliance check all update the same timeline.
That is why we built RevRing.com around a single platform: CRM, preview, progressive, power and predictive dialing, two-way texting, lead routing, AI call scoring and coaching, and compliance checks before every call. Agencies can run on RevRing.com’s CRM or connect HubSpot, Pipedrive or GoHighLevel at no extra cost. When the data sits together, the manager can see that an agent’s contact rate is fine and the close rate is not, and coach that one thing on Tuesday.
Our advice is to start with the cadence and the scorecard. Get those written down and enforced, then let software carry the load.
Marc
What it costs to run this on RevRing.com
If you want to see how the pieces fit in one platform, RevRing.com publishes seat-based plans: Basic Dialer at $49.99 per seat per month, Field Agent at $50 per seat per month, Scale at $139.99 per seat per month and Pro at $189.99 per seat per month. Enterprise is quoted separately. Usage is billed on top. Outbound calls cost $0.03 per minute, or $0.05 per minute through the Power Dialer. Outbound SMS is $0.02 per segment and inbound is $0.01. AI call scoring costs $0.20 per scored call, which lets you decide how many calls to score.
If your floor needs higher contact volume, look at the Power Dialer or the predictive dialer. If you buy leads, the Lead Marketplace sits in the same system as your routing and follow-up.
For a walkthrough of the dialer, CRM and call scoring with your own lead flow in mind, start at RevRing.com and schedule a live demo.
A 30-day rollout plan
Week 1: Define. Write your stages, your required fields (source, cost, lead type) and your first-attempt target. Decide who owns each routing rule.
Week 2: Configure. Build the pipeline, the routing rules, the cadence tasks and the text and email templates. Make sure outcomes from calls land on the lead record automatically. Confirm consent capture and Do Not Call checks are part of the flow.
Week 3: Baseline. Run the cadence and the scorecard without changing anything else. Record speed to first attempt, contact rate, quote rate and close rate by agent and source.
Week 4: Coach. Pick the weakest ratio on the floor, hold the first Friday session on it and set a target for next month. Review lead sources and cut or renegotiate the weakest vendor.
Repeat monthly. Gains come from small corrections held in place, not from a single launch.
Common mistakes
- Too many stages. If agents cannot explain the difference between two stages, merge them.
- Cadence without a reason per touch. Prospects can tell when a message exists only to hit a quota.
- Coaching on outcomes only. Revenue totals hide whether the problem is contact, quote or close.
- Averaging lead sources. One strong vendor can mask two weak ones.
- Treating compliance as a one-time setup. Rules, lists and consent records change, so review them on a schedule.
- Scoring calls and never acting on the scores. If no one reads the results, stop paying for them.
FAQ
How fast should an agent call a new insurance lead?
As fast as your team can sustain, measured in minutes rather than hours. Set a specific target, alert on misses and review the miss rate by agent each week. Speed matters most for web leads, where the prospect is likely shopping several agencies at once.
How many follow-up attempts are enough?
Most agencies work a new lead across roughly two weeks, mixing calls, texts where consent exists, and email, then move the lead to nurture with a future reopen date. Test your own count: compare contact rates on attempts five through eight against the cost of the labor. Stop when additional attempts stop producing conversations.
What should a call scorecard include?
Keep it to five or six items: opening, discovery, quote presentation, objection handling, next step and required disclosures. Score each as yes, partial or no. A short rubric gets used, and it gives agents one or two clear things to change.
Do I need a new CRM to coach calls and track follow-up?
Not necessarily. What you need is call outcomes, attempts and source data landing on the lead record automatically. You can get there with a connected setup or an all-in-one platform. Choose based on how much manual entry your agents still do.
Does automation make follow-up compliant?
No tool guarantees compliance. Automation can run consent, Do Not Call and calling-hours checks consistently, but your agency is responsible for its policies, consent records and state-level rules. Review the workflow with counsel.