Insurance · Upsell Call
Upsell Script for Insurance Customers: From Claims Account Review to a Recovery Module Expansion
You sold this carrier a claims triage and routing product fourteen months ago. Personal auto FNOL-to-close came down, the VP of Claims Operations quoted the number in her last board deck, and the account has been quiet ever since. Now you want to sell the recovery module — subrogation identification and salvage flagging — and you have thirty minutes on a Tuesday in April, three weeks before hail season, with a claims org that is mid-conversion on a twenty-two-year-old policy admin system.
This is not a discovery call and it should not sound like one. The buyer picked up expecting a check-in. Everything you need to earn the second half of this call lives in your own dashboard: which regions are weekly active, which claims teams took 40 seats and never onboarded, what her cycle time did by line, whether her SIU Director ever logged in. If you open with "I wanted to show you something we just launched," you get a polite ten minutes and a referral to the innovation team, which is where claims vendors go to die.
What follows is a block-by-block script for the expansion conversation with a Chief Claims Officer, VP of Claims Operations, or Director of Claims Transformation at a regional carrier — plus the objections you will actually hear (core conversion, a fraud analytics vendor that got switched off, a mutual's board expense-ratio target, and a chief actuary who does not accept vendor assumptions) and the sample dialogue to rehearse before you dial.
The upsell call script
Say it in your own words. The structure is the part that matters.
- 1
0. Pre-Call Usage Audit — do not dial without these six lines
Fill these in from your own systems, not from memory. If you can't, reschedule. 1. ADOPTION: "Personal auto has [180] licensed seats and [118] weekly active across [three] regions. Property took [60] seats in the November amendment and has [11] weekly active — the [Midwest] property team never completed onboarding after the [derecho]." 2. DEPTH: "They use FNOL triage and severity routing daily. They have never opened the reserve-variance view. The exec sponsor last logged in [February 4]." 3. OUTCOME IN THEIR NUMBERS: "Personal auto FNOL-to-close went from [21.4] days at baseline to [15.8] days last quarter. ALAE per auto claim is down [$61]. Both numbers came out of their own monthly ops pack, not our benchmark." 4. COMMERCIALS: "[$412K] annual, renewal [March 31], signed by [the Chief Claims Officer, who is still in seat]. [12%] multi-year discount. Property seats co-termed." 5. SUPPORT HISTORY: "One P2 in January on the FNOL feed lag, closed in nine days. No open escalations. Last CSAT [8]." 6. CHAMPION STATUS: "[Dana] was promoted from Director of Claims Transformation to VP of Claims Operations in October — she now owns cycle time, LAE, and the DOI complaint count. That is a bigger surface than she had when she bought." Test: can you say one thing about their account that would mildly surprise them? "Your property team has eleven weekly actives out of sixty seats" passes. "How's everyone finding the platform?" fails.
- 2
1. The 30-Second Honest Frame
"Thanks for the time. Two things, and I'll be straight about both. First — I pulled your usage across the last two quarters and there's a pattern in the property book I want to check with you, because I don't think it means what it looks like it means. Second, depending on your answer, there's a piece of our platform I think is relevant to where your recovery numbers sit, and if it isn't relevant I'll say so and we'll spend the time on something else. Fair?" Do not say "new module." Do not say "quick check-in" and then ambush her at minute twenty. Pre-committing to walking away lowers the guard faster than any rapport.
- 3
2. Anchor on the Result She Already Owns — in her numbers, out loud
"Before anything else — when we started, personal auto was averaging 21.4 days FNOL to final payment and your ALAE per claim was running about $[X]. Last quarter I've got you at 15.8 days and ALAE down about $61 a file. Does that match how it feels on your side, or am I reading my own dashboard optimistically?" Then shut up. Her correction is worth more than your number. If she corrects you — "it's really 16.9 because we exclude litigated files" — write that sentence down verbatim. That is the metric she defends internally, and it becomes line one of the business case she takes to her Chief Claims Officer. HARD RULE: if she can't point to a result, stop the upsell. "Then let's not talk about anything new today. Let's fix the fact that you're paying $412,000 a year for something you can't put in front of your actuary." That call preserves the March renewal. The other version loses it.
- 4
3. Name the Dark Corner Before She Does
She is sitting there thinking *we barely use what we've already bought.* Say it first so you're diagnosing instead of defending. "Here's what I'd push back on if I were you. Property has 60 seats and eleven weekly actives. You'd be entirely right to ask why you'd buy anything else before you've used that. So let me tell you what I think those other 49 adjusters are doing — they're not idle. They're working the same triage decisions in a spreadsheet and a Teams channel because onboarding got interrupted by the derecho and never restarted. That's an adoption problem I own, not a product gap, and I've got a fix that doesn't cost you anything: two 45-minute sessions with your property supervisors before hail season starts, run by our CSM." Then — and only then — the pivot: "That's separate from what I want to ask you about, and I'd do it whether or not the second conversation goes anywhere."
- 5
4. Find the Seam — five diagnostic questions before you name the module
Lead with the leak, never the feature. - "When an adjuster closes a file with a clearly at-fault third party, who decides whether it goes to subro, and how does it get there?" - "What's your subrogation recovery rate running on personal auto right now, and is that number in front of your Chief Claims Officer monthly or quarterly?" - "What happens to total loss files after the check goes out — who's watching salvage?" - "Where does this still break? What's the Monday morning fire drill in your recovery unit?" - "What did we deliberately leave out of scope in the original deal because it was too big a bite?" That last one is the highest-yield question on this call. Recovery was almost certainly scoped and cut in the first negotiation. You are not selling something new — you are reopening a decision she already made once, with fourteen months of her own cycle time data as the reason to revisit it.
- 6
5. Build the Case in Her Arithmetic — unit, volume, capture, net
"Let me do this in your numbers and you tell me where I'm wrong. THE UNIT: your recovery unit's average net subro recovery on a referred auto file is about $1,900 after fees, based on what your team told us in the January working session. THE VOLUME: last quarter you closed 412 personal auto files that had a coded at-fault third party and no subro referral before the file went to closed status. That's from your own claim extract, not my model. THE CAPTURE RATE: I'm not going to tell you we recover all 412. Statutes have run on some, some have no collectible carrier behind them, some your team looked at and correctly passed on. Say we surface 20% as genuinely recoverable and your unit works them. That's 82 files, roughly $156,000 a quarter, against a subrogation recovery rate that's currently sitting at [her number]. THE NET: the module is $7,500 a month on top of your current agreement. Annualized, that's $90,000 against roughly $620,000 — call it seven times, and I'd rather under-promise than have your chief actuary find air in my numbers. If you think 20% is generous, tell me the number you believe and we'll rerun it right now." Say the price out loud in the same breath as the value. Never make her ask for it. And hand her the pencil on the capture rate — that's what separates a business case from a pitch.
- 7
6. Kill the ROI Argument Before Actuarial Does — offer the back-test
"Here's what I'd rather do than argue about my assumptions. Give me last year's closed personal auto files — historical extract, flat file, no production connection, nothing that touches the core. We score them and hand you back two lists: the files where we'd have flagged recoverable subro that never got referred, and the false positives, so your recovery manager can see exactly how much noise comes with the signal. That way the savings are calculated against actual settlements you already booked, not against a projection. If your chief actuary wants to strike the number, he can strike it off real closed files. And if it comes back thin, I'd rather find that out in a back-test than in month four of a rollout."
- 8
7. Cost the Implementation in Hours and Names — because bandwidth is the real objection
"On effort, because I know what your April looks like. The back-test costs your team one extract and one hour with whoever owns the claim data mart. That's it — we do the rest. If it converts to live: one 90-minute config session with your recovery manager, then about two hours a week from that one person for three weeks while we tune the referral threshold to whatever volume your unit can actually work. No IT project team. Nothing written back into the PAS — we read the claim file and we write to a queue, so nothing your core conversion team owns changes and statutory extracts and check issuance never see us. If it needs more than that, I've mis-scoped it and I'll come back and tell you." If you can absorb work — templates, your services team doing config, your CSM running enablement — put it on the table here, not later as a concession.
- 9
8. The Ask — one region, one quarter, reversible, co-termed
"So here's what I'm actually asking for. Not the whole book. Give me personal auto in [the Central region] for one quarter, starting after hail season if that's cleaner — say June 1. You set the success criteria and we write them down today. My proposal: subrogation recovery rate on that region's auto book moves by [X] points over the quarter, and the false-positive volume stays inside whatever your recovery manager says she can work. If we miss either one, we pull it out at the end of the quarter and you owe nothing further. If we hit it, we roll it to the other two regions at your March renewal and co-term it into the existing agreement, so you're managing one contract and one negotiation instead of two." Co-terming is the most underused unlock on this call. It removes a procurement cycle and makes the add-on feel like an amendment rather than a new purchase.
- 10
9. Buying Committee Check — ask directly, on this call
"One practical thing. Last time this went through [name] and the Chief Claims Officer signed. Is that still the path at this dollar level, or has the threshold changed since the conversion project started? And does your Chief Underwriting Officer or your CFO see recovery numbers in a way that means they'd want to be in this conversation? I'd rather bring them in at the front than surprise them in week six." If a new procurement policy, a new CFO, or a board expense-ratio target has landed since the original deal, you are running a fresh evaluation, not an expansion. Find that out today.
- 11
10. Same-Day Recap Email — write it for the forward, not the CRM
Send within four hours. Contents, in this order: 1. The result she confirmed, in her words: "personal auto FNOL-to-close at 16.9 days excluding litigated files, down from 21.4." 2. The gap she described: 412 closed auto files last quarter with a coded at-fault third party and no subro referral. 3. The math with her assumptions, including the capture rate she chose, not the one you proposed. 4. The price: $7,500/month, co-termed to March 31. 5. The implementation ask in hours and names. 6. The pilot scope, the success metric she set, and the decision date. 7. The free adoption fix for the property seats, with the two session dates. This email is the artifact she forwards to the Chief Claims Officer. Write it for that reader.
- 12
PARK SIGNALS — stop and don't push
Do not run the expansion if any of these are true: an open P1 or unresolved escalation on the core product; your champion changed roles in the last 60 days; the March renewal is inside 60 days with an unanswered value question; weekly actives flat or declining and you can't explain why; a cat event in their footprint in the last six weeks; or a board-level expense review announced this quarter. Pushing through any of these converts a renewable regional carrier into an at-risk one to book an incremental $90K. Say instead: "You've got 3,000 hail claims in the queue. I'm going to stop talking about this and come back in July — can I hold thirty minutes on the 14th?"
How the call actually sounds
Prospect on the left, the rep on the right.
Rep
Thanks for making the time, Dana. Two things, and I'll be straight about both. First, I pulled your usage across the last two quarters and there's a pattern in the property book I want to check with you. Second, depending on your answer, there's a piece of the platform I think is relevant to your recovery numbers — and if it isn't, I'll say so and we'll spend the time elsewhere. Fair?
Buyer
Fair, but I'll set expectations. We're eleven months into the core conversion, I've got 60 IA contracts sitting unsigned for hail season, and my budget for anything with the word platform in it closed in November. So if this is a sales call, it's going to be a short one.
Rep
Understood. Then let's start with something that isn't a sales call. When we signed, personal auto was averaging 21.4 days FNOL to final payment and ALAE per auto claim was running around $840. My data says last quarter you were at 15.8 days and ALAE is down about $61 a file. Does that match your side, or am I reading my own dashboard optimistically?
Buyer
It's 16.9, not 15.8. We pull litigated files out — you can't put a rep-represented BI file in the same average as a comp glass claim and expect anyone in this building to take the number seriously. But yes. 16.9 from 21.4. My Chief Claims Officer put it in the March board pack and the DOI complaint count is down with it, which is the part that actually mattered.
Rep
16.9 excluding litigated. I'm writing that down because that's the number I should have been quoting. Second thing, and I'd rather say it than have you say it: property has 60 seats and eleven weekly actives. If I were you I'd be asking why you'd buy anything else before you've used that.
Buyer
I have asked that. Twice. My property manager says onboarding got blown up by the derecho in June and never restarted, which is true, but at some point that becomes an excuse for a line item I'm paying for.
Rep
It's my problem, not yours. Two 45-minute sessions with your property supervisors, run by our CSM, before hail season starts — no charge, and I'll do it whether or not the rest of this conversation goes anywhere. Separately: can I ask you about recovery? When an adjuster closes an auto file with a coded at-fault third party, who decides whether it goes to subro?
Buyer
The adjuster does, in theory. In practice they're trying to close files before month-end and subro is a checkbox nobody's measured on. We have a four-person recovery unit for the whole auto book. Our subrogation recovery rate is around 2.1% of paid losses and my Chief Claims Officer wants it at three, which is a number he got from somewhere and won't tell me where.
Rep
Then here's the specific thing. Last quarter you closed 412 personal auto files with a coded at-fault third party and no subro referral before closure. That's from your own claim extract in our system. At your average net recovery of about $1,900, if only a fifth of those are genuinely recoverable, that's roughly $156,000 a quarter sitting in closed files.
Buyer
Stop. I've heard this pitch. Two years ago we bought a fraud analytics tool that was going to transform SIU. It threw 900 alerts a month at a six-person unit, the precision was somewhere south of terrible, my SIU Director stopped opening the queue, and we switched it off eight months in and ate the contract. I'm not doing that again with subro.
Rep
That's the right scar to have. What was the threshold set at, and who set it — you or the vendor? Because most of those failures are threshold and workflow, not the model.
Buyer
The vendor set it. They wanted the recall number for their case study. We were the case study.
Rep
Then invert it here. Before you buy anything, give me last year's closed personal auto files as a historical extract. No production connection, nothing that touches the core, no write-back to the PAS. We score them and hand you two lists: the files where we'd have flagged recoverable subro that never got referred, and every false positive, so your recovery manager can count the noise herself. If it comes back thin, you've spent one hour of a data analyst's time and I go away.
Buyer
And when it comes back looking wonderful, you'll want money I don't have. We're a mutual. There is no innovation budget — surplus is policyholder money, and the board set an expense ratio target for next year that my Chief Claims Officer has already told me is not negotiable.
Rep
Then don't frame it as tech spend, frame it as loss ratio. Subro recovery lands against paid losses, not expense. It's $7,500 a month — $90,000 a year — against a number your own extract says is somewhere north of half a million. And I'd ask for one region, personal auto only, one quarter starting June 1 so it's after hail. You set the success criteria today. If we miss, we pull it and you owe nothing further.
Buyer
Our chief actuary will pull that apart. He doesn't accept vendor capture rates, and honestly neither do I.
Rep
He shouldn't. That's why the back-test runs on closed files — the recovery is calculated against settlements you already booked, and he can strike the capture rate to whatever he believes. Tell me the number you'd use instead of 20% and I'll rerun the arithmetic on this call.
Buyer
Ten. And I want my recovery manager setting the alert volume, not you. She can work maybe forty referrals a week on top of what she's got.
Rep
Ten percent and forty a week — that's the deployment size, and I'll write it into the pilot scope. At ten it's roughly $78,000 a quarter against $22,500 of cost for the quarter. Last question and then I'll get out of your way: last time the Chief Claims Officer signed. At $7,500 a month co-termed into the March renewal, is that still the path, or has the conversion project changed the threshold?
Buyer
Under a hundred thousand annual he can still sign it, but if it's a new agreement it goes to procurement and that's three bids and a quarter of my life. Co-term it as an amendment and send me the back-test proposal. Not to innovation. To me.
Objections you will hear
What they say, and what you say back.
| Objection | How to answer it |
|---|---|
| “"We're eleven months into the core conversion. There's no IT bandwidth for anything until 2027."” | Agree, then remove yourself from the path. "That's exactly why I'm scoping this so it never touches your conversion team. Read-only against the claim file, flat file extract for the back-test, and if it goes live we write to a queue — no schema change, nothing written back into the PAS, no change to statutory extracts or check issuance." Then ask the diagnostic question: "What's your auto cycle time done since the conversion started?" It has almost certainly slipped, and that's the argument for acting now rather than after go-live. If it genuinely requires the VP of IT, say so plainly and ask to be on the roadmap conversation for the quarter after cutover instead of pretending otherwise. |
| “"We bought a fraud analytics tool two years ago. SIU drowned in false positives and we switched it off."” | Don't defend your model — interrogate theirs. "What was the precision, and who set the referral threshold, you or the vendor?" Most of these failures are threshold and workflow. Then invert the pilot: score last year's closed files, show both the hits and every false positive, and let the recovery manager or SIU Director declare the alert volume they can actually work — forty a week, sixty a week, whatever it is. Size the deployment to their capacity, not to your model's output, and put that number in the pilot scope in writing. |
| “"Every dollar goes through actuarial scrutiny. Your ROI won't survive our chief actuary."” | Bring the model empty. "I'm not going to hand you a calculator with my assumptions baked in. Give me your paid losses on the auto book, your current subrogation recovery rate, and your average net recovery per referred file, and we'll build it in your numbers." Then move the proof onto ground the actuary trusts: cycle time is hard to book, but a documented subrogation recovery and a measurable move in ALAE per claim land in a reserve review. Offer the back-test on closed claims so the number is calculated against actual settlements rather than a projection. |
| “"Budget for your category closed in November. There's nothing left this year."” | Do not discount — discounting tells them the original $412K was padded and teaches them to wait every time. Ask instead: "Committed to what, and when does it reset?" Then explore three doors: is there an underperforming line in the same budget — the IA spend, the contractor doing recovery triage manually, the analytics tool nobody opens? Can it come out of the recovery unit's budget rather than claims ops? Can you sign the order form now and start it at the March renewal, co-termed? Budget objections at carriers are usually timing and authority objections wearing a costume. |
| “"We're a mutual. Surplus is policyholder money — we don't have a technology line like a stock carrier."” | Move it out of expense and into loss ratio. "Subrogation recovery lands against paid losses. This isn't an expense-ratio conversation, it's a combined ratio conversation, and every point of leakage you don't recover is a point you have to claw back in a rate filing that needs a regulator's signature." Then get practical: ask when the planning cycle closes, what the board's expense ratio target is for next year, and what dollar threshold the Chief Claims Officer can approve without going to the board. Size phase one to fit under that threshold. |
| “"My property team never onboarded on what we already bought. Why would I sell more of this to my Chief Claims Officer?"” | Concede it completely and fix it for free. "Sixty seats, eleven weekly actives — you're right and it's mine to fix. Two CSM-led sessions with your property supervisors before hail season, no charge, and I'll run them whether or not you buy anything." Then be honest about whether the new module touches that gap. It doesn't — this is an auto recovery problem, that's a property adoption problem — and saying so is what earns you the right to keep talking. Never let unused seats and a new line item ride in the same sentence unless you can show the unused seats are doing the work somewhere else. |
| “"Send me something and I'll route it to our innovation team."” | The innovation team is where claims vendors go to die. Acknowledge it and anchor back to the operational number: "Happy to — before I do, is subrogation recovery rate actually something you're being measured on this year, or is the pressure coming from somewhere else, like severity on the property book or agent retention?" If she names a real pressure, book time with her and whoever owns that number. If she can't, you've learned this is a stall, and the right next question is: "Who owns recovery rate in the org, and can you introduce me?" |
Questions reps ask about this call
- How is an upsell call to an insurance carrier different from a normal expansion call?
Three things change it. First, the buyer's calendar is seasonal — an expansion pitch three weeks before hail season or during an active named storm will be resented no matter how good the math is. Second, the money is judged against a combined ratio, not a tech budget, so recovery and leakage arguments land where seat-count arguments don't. Third, the buying committee is wider than it looks: a VP of Claims Operations may own the number but the Chief Claims Officer signs, the chief actuary strikes your assumptions, and at a mutual the board's expense ratio target is a real constraint, not a stalling tactic.
- What usage data do I need before I dial a Chief Claims Officer or VP of Claims Operations?
Licensed seats versus weekly active by line and by region, so you can name the dark corner yourself; which features are untouched and when the exec sponsor last logged in; one outcome number sourced from their data — cycle time from FNOL to close, ALAE per claim, DOI complaint count — not your benchmark; contract value, renewal date, discount and who signed; open tickets and the last escalation; and whether your champion is still in the role. If you can't say something about their account they'd be mildly surprised you knew, don't dial.
- Should I lead the upsell with cycle time or with claims leakage?
Lead with cycle time only to anchor the result they already own from your existing product — get them to say the number out loud in their own words. Then pivot to the leakage side for the expansion, because cycle time reduction is notoriously hard to book in a reserve review while a documented subrogation recovery or a measurable drop in ALAE per claim shows up in numbers the chief actuary already trusts. Anchor on cycle time, expand on recovery.
- How do I handle the objection that they're mid-migration on their policy admin system?
Don't fight it and don't wait for it. Scope the expansion so it never competes for the conversion team: read-only against the claim file, flat file or message queue, no schema change, no write-back to the PAS, no impact on statutory extracts or check issuance. Then ask what their cycle time has done since the project started — claims service usually degrades during a 24–36 month conversion, which is the argument for acting now. If it genuinely needs the VP of IT, say so and ask for the roadmap conversation the quarter after go-live.
- What's the right size of ask on the first expansion call?
One line of business, one region, one quarter, with success criteria the buyer writes down on the call — a target subrogation recovery rate, a cycle time threshold, a false-positive volume their recovery manager or SIU Director says they can work. Make it reversible: if you miss, it comes out and they owe nothing further. Then co-term the add-on into the existing renewal so it lands as an amendment rather than a new purchase requiring three bids and a procurement cycle.
- When should I not run the upsell at all?
Park it if there's an open P1 or unresolved escalation, if your champion changed roles in the last 60 days, if the renewal is inside 60 days with unanswered value questions, if weekly actives are flat or declining and you can't explain why, if a cat event has hit their footprint in the last six weeks, or if the board has announced an expense review this quarter. Pushing through any of those turns a renewable carrier account into a churn risk to book an incremental $90K. Say you'll come back in July, and hold a specific date.