In-force policy servicing that answers from the policy's own contract, page cited, while the caller is still on the line
Answers on decades-old policies, each cited to its contract or rider page.
See one case, screen by screen ↓In-force policy servicing is the work of answering and acting on questions about policies already issued: loans, withdrawals, surrender values, beneficiaries, premium changes, conversions and riders. In-Force Policy Servicing answers those questions from each policy's own contract, riders, endorsements and values, with every sentence cited to its page.
How to choose AI solutions for insurance →
Updated
A policyholder calls to ask whether he can borrow against a universal life policy issued in 1998, and what the loan does to it. The rep can see today's values in the admin system. She cannot see what his contract says: which form version and state variant was issued, how loan interest is charged, when a loan ends the policy. That contract is a scan, often from microfilm, and the policy may be on its third admin system.
So the rep answers from a product summary — wrong for older form versions and state variants — or escalates and promises a callback. Nobody on the phone can project what a loan with interest added does over twenty years, so the customer takes it without knowing it lapses the policy at 84. Those answers come back later as complaints in the register and in market conduct exams.
Estimated staff hours for one loan question on an older policy, by hand and with the solution. The customer's wait by hand is 2–3 business days on top.
Six specialist agents read the policy's own papers and values and project what an action does; the service rep decides and sends.
Monday 12 October 2026, 09:42. Harold Lindqvist, 71, calls Larchmere Mutual Life about his 1998 universal life policy. Here is what happens next, screen by screen, in the working solution.
Renee's queue holds 14 questions from calls, agents' emails and letters, on policies issued from 1979 to 2019. Each was answered from the policy's own documents before anyone opened it: a reduced paid-up quote on a 1979 whole life, a term conversion deadline, a beneficiary question withheld because the caller is not the owner.
“14 of 14 questions today · median time to a cited answer 52 s”
Harold is verified by 3 facts. He wants to borrow $40,000 for his granddaughter's tuition. Renee presses Ask, and the agents work through it in view: the policy finder identifies form UL-97-OH, Ohio variant, Rev. 11/96, two riders (both ended) and two endorsements; the contract reader reads §7.1–7.5 on pages 11 and 12 and §5.3 on page 8; the values agent reads the 9 October valuation.
The projection agent runs three cases of 240 months through the company's in-force projection engine: as it stands, $40,000 with interest added, $40,000 with interest paid. The answer checker then reads every sentence back against its passage and every figure against the valuation or the run.
“7 of 7 sentences cited · 12 figures matched to the page, the valuation or the projection.”
He can borrow up to $78,245.19. Interest is 6% a year, charged each 14 March; on a total loan of $54,862.40 that is $3,291.74 a year. His daughters would receive $195,137.60 instead of $235,137.60. Beside the answer, the 20-year chart shows the death benefit falling as interest is added and a red marker where the policy is projected to lapse.
“Lapses August 2039 · age 84 — loan $117,018 passes the cash value.”
Citation 5 opens page 12 of 32 of form UL-97-OH, scanned from microfilm roll 1998-114. §7.4 is highlighted: once the indebtedness equals or exceeds the cash surrender value, the policy terminates 31 days after notice is mailed. Margin notes say what each clause means for Harold today, including that loaned value is credited 4%, not the current 4.25%.
Numbers and rates are quoted from the contract, never paraphrased.
Renee switches loan interest to “Pay each March”. The projection runs again: the death benefit holds at $195,137.60 and the policy is projected in force to November 2047, age 92 — close to May 2048, age 93, where it stands without the loan. The answer reminds her the payment must be marked as loan interest, or §7.5 applies it as premium.
“Servicing projection · current assumptions · not an illustration”
“Put this in the reply” turns the case into a letter to Harold: how much he can borrow, what it costs, what his daughters would receive, the year to watch, and a table every other September. Every figure is footnoted to its page, the valuation or the projection run. Before it goes, the checks show 6 of 6 statements footnoted and a reading level of grade 7.6.
“Projected lapse within 15 years (Aug 2039): an in-force specialist reviews before it is sent.”
Tom Bellamy, her agent, asks whether Eleanor Pruitt's dividends can pay down her $6,200 policy loan. Her 1984 whole life was issued in Pennsylvania, but the scan on file is base form WL-84, whose §9.2 has no option to reduce a loan. The filed Pennsylvania variant WL-84-PA adds it as option (e). Two answers are possible, so the agents did not write one.
“The papers disagree.”
Priya sees both pages side by side and what each agent found. She can answer from the Pennsylvania variant — the application, the premium notices and 41 other WL-84 policies issued in PA in 1984 all point to it — ask Records for the copy delivered to her, or send it to Legal. Recording the decision can add a rule: WL-84 issued in PA reads from WL-84-PA.
Martin sees 1,846 questions answered this week, 98% with a cited answer, and 41 callbacks against 296 a week before go-live. He reads a sample of 120 drawn by topic — 117 agreed, 1 disagreed on wording — with the number of sentences cited on each.
The same book of business, seen by the five people who carry it — what their week looked like, and what it looks like now.
Six specialist agents read the policy's own papers and values and project what an action does; the service rep decides and sends.
Finds the policy, the form version and state variant it was issued on, its riders and endorsements, and checks the caller is someone who may be told.
Reads the policy's own scanned contract, riders and endorsements and returns the passages that answer the question, with section, page and a margin note.
Reads the current values from the admin system as of the last valuation: account value, surrender charge, loan, monthly deduction, cost basis, tax status.
Turns the request into a case for the company's in-force projection engine, runs it month by month, and returns the death benefit, net cash value and lapse month.
Reads every sentence back against its passage and every figure against the valuation or the projection run; stops and escalates when the papers disagree.
Drafts the letter and the call note in plain words, with each figure footnoted to its page, the valuation or the projection run.
Decides and sends. The agents propose; a named person decides.
Reps and specialists can ask about the policy on the call, today's questions, or what an action would do to a policy. Every answer is cited.
What if Harold pays the interest each March instead?
Paying the interest, $3,291.74 each 14 March at 6%, keeps his loan at $54,862.40, so the death benefit stays at $195,137.60. The policy is then projected in force to November 2047, age 92, instead of August 2039, age 84. Tell him to mark the payment as loan interest, or it is applied as premium.
Show me where the contract says the policy can end with a loan
It is §7.4 on page 12 of his 1998 contract, form UL-97-OH: once the loan equals or exceeds the cash surrender value, the policy ends 31 days after we mail notice to him and any assignee. The grace period in §5.3 on page 8 gives 61 days when the value left will not cover next month's deduction.
Draft the letter to Harold with the 20-year table
The letter is drafted in plain words: the $78,245.19 limit, the 6% interest, the death benefit his daughters would receive, the year to watch (August 2039) and a table every other September to the lapse, with what happens if he pays the interest. Because the lapse is within 15 years, Priya Raman reviews it before it goes.
Which of today's questions need a specialist?
Two. Eleanor Pruitt, asked by Tom Bellamy: her 1984 whole life was issued in Pennsylvania but the scan on file is the base form, and only the Pennsylvania variant lets dividends reduce a loan. Owen Castellanos: the lender's release names L-6612058, not his L-6612085.
13 screens from the working solution, on its sample data. Pick one to see it large.
What the answers may say, when a specialist reviews them, and who may be told what — all settings.
| Setting | Default | Choose from |
|---|---|---|
| Specialist review when projected lapse is within | 15 years | 10 · 15 · 20 years |
| Specialist review when a loan is larger than | $50,000 | Any amount |
| Who reviews flagged replies | Priya Raman | Priya Raman · Dale Whitford |
| Every sentence cited | Always on | Always on |
| Projection horizon | 240 months | 240 · 360 months |
| When issue state, scan and form variant differ | Stop and escalate | Stop and escalate |
| Tax wording | Status and Form 1099-R only; no advice | Locked |
| Who may be told | Owner after three facts · agent of record · anyone else, how to file a claim | Values, beneficiaries, actions · values, no beneficiaries · claim filing only |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample policy data · “target” = the design goal, measured in the live solution · “estimated” = our estimate from typical service-centre practice · “typical” = published figures (ACLI Life Insurers Fact Book 2025 — $14.1 trillion of individual life insurance in force in the US). People, companies, policies and systems named on this page are fictional — characters and sample data in the working solution.
In-force policy servicing is the work of answering and acting on questions about policies already issued: loans, withdrawals, surrender values, beneficiaries, premium changes, conversions and riders. In-Force Policy Servicing answers those questions from each policy's own contract, riders, endorsements and values, with every sentence cited to its page.
The policy finder works out which form version and state variant was actually issued, the contract reader reads that policy's scanned contract, riders and endorsements, and the values agent reads today's values from the admin system. In the working solution, a 1998 universal life loan question was answered in 36 seconds, with eight sources.
No. Values are read from the admin system as of the last valuation, read only, and the loan limit is the contract's own rule applied to them. The answer checker matches every figure to the valuation, the page or the projection run.
No. The projection agent runs the request through your in-force projection engine and labels the result a servicing projection, not an illustration. A formal in-force illustration can be ordered from the answer.
The agents stop. When the issue state, the scan and the form variant differ, or two passages give different answers, the question goes to an in-force specialist with both pages side by side. The specialist decides, and can add a variant rule for the next policy on that form.
Only an owner verified by three facts, the agent of record (values, no beneficiaries) or an assignee. Anyone else — family, friends, other callers — is told how to file a claim. These are settings.
Yes. The representative decides and sends every answer; the reply writer never sends. An in-force specialist reviews any reply whose projection lapses within 15 years or whose loan exceeds $50,000, and decides every wording escalation.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your form library, scanned contracts, letter templates and a sample of past questions — and it goes live once every quality gate has passed. We will walk you through it on your own older policies first.
Hundreds of record pages read before the underwriter opens the case.
1-min demoNew Business Case ManagementEvery open requirement chased, every case status known without a phone call.
1-min demoLife Underwriting Manual SearchImpairment ratings answered from the manual, with the page that says so.
1-min demoWe'll run In-Force Policy Servicing on a sample of your own older policies and the questions your reps escalate.