Category Archives: Insurance & Actuarial

Client Case Study: The Salary Survey Actuaries Actually Trust

Salary survey research & actuarial recruitment marketing · Ezra Penland Actuarial Recruitment

The Client

Ezra Penland Actuarial Recruitment, a national actuarial search firm. Before I was the founder and business strategist behind 1000 Startups, I was a founder and partner there, and the salary surveys were mine – built them, ran them, defended them at conference booths against actuaries who show up to arguments with spreadsheets already open.

The Problem

Actuaries price risk for a living. Hand one of them a compensation number with no methodology behind it and watch the eyebrow go up. Some recruiting firms across industries publish “salary guides” – most are lead-gate PDFs dressed up as research, three vague bands and a form to fill out before you can see them. That doesn’t survive contact with someone who has passed four actuarial exams and reflexively distrusts round numbers.

What I Built, In Order

Real respondent data, layered on top of proprietary placement data from actual searches – not guesses, not scraped job postings.
Cut by the variable that matters. Not just years of experience – exams passed, credential earned (ASA, FSA, ACAS, FCAS), because two actuaries with the same tenure and different exam counts are not the same hire.
Split by employment type. Consulting pays differently than insurance, which pays differently than reinsurance. Blending them into one number is how you lose the room.
Four practice areas, not one. Property & Casualty, Life, Health, and Pension each got their own survey, because a Health actuary reading a P&C table just closes the tab.
Free and ungated. Eleven downloadable PDFs, no email wall. The trust was the product; the recruiting pipeline was what trust bought us.

What the Numbers Said

Reported as the middle 85% of all compensation (base plus bonus) – wide enough to be honest, narrow enough to be useful.
Eleven survey cuts across four practice areas, refreshed every year so the numbers never went stale on us.
Became the number actuaries and hiring managers actually cited – in comp committee meetings, in offer negotiations, in Society of Actuaries hallway arguments – without anyone at Ezra Penland picking up the phone first.

The Resolution

The survey became the industry’s reference point, not just Ezra Penland’s marketing asset. Candidates used it to negotiate. Employers used it to build ranges. And every single one of them arrived at the Ezra Penland website to get it, which is the whole trick: publish the number people need, and the leads follow the number home.

If This Sounds Familiar

If your best marketing asset is something true that you’re hoarding behind a form, the fix usually isn’t a bigger campaign. It’s giving the number away and letting it do the selling for you. That’s the whole method – audit, build the credible thing, then get out of its way.

The Renewal Is the Sale

Leave a reply

Insurance grew up around a book that renews whether anybody calls or not, so the whole industry is built around keeping it. Everyone else staffs the renewal with nobody.

1. Costco makes its money on the renewal, not the sale. Membership fee income has for years accounted for a sum broadly comparable to Costco’s entire net profit, and US and Canadian renewal rates run in the low nineties as a percentage. The warehouse is a mechanism for making people renew a card. Everything on the shelf is a supporting argument.

2. A point of retention outperforms a point of new business. Reichheld and Sasser put a 5% retention improvement at 25% to 95% more profit. Nothing on the acquisition side has ever produced a range like that, and yet the acquisition side owns the budget, the headcount and the meeting.

3. Brokers start the renewal ninety to a hundred and twenty days out. Not thirty. Not when the client calls. There is a calendar, it is non-negotiable, and someone owns it by name. Ask a software company when their renewal motion starts and you will usually get a philosophical answer.

4. Net revenue retention above one hundred means you grow without new logos. A business where existing customers expand faster than others leave is a business that compounds while the sales team sleeps. It is the single most valuable number in a subscription company and it is produced almost entirely by people who are not in sales.

5. Build the renewal calendar first, then staff it. Every account, every renewal date, every trigger at ninety days. It is a spreadsheet. The reason it does not exist in most companies is not difficulty; it is that nobody gets promoted for it.

6. The ninety-day conversation is not about renewing. It is about what changed at their end since they signed. Companies that only appear when the invoice is due have taught the client exactly what the relationship is, and the client behaves accordingly.

7. Churn is usually decided in month two, not month eleven. Onboarding is a retention program wearing a different badge. If the thing never got properly installed, used or understood, the renewal was lost long before anyone noticed the usage chart.

8. Ask the ones who stayed, not just the ones who left. Everybody runs churn analysis on departures. Almost nobody interviews five-year customers about why they never seriously considered leaving. That answer is your actual positioning, tested by time rather than by a workshop.

9. Publish your retention number if it is good. It is the least gameable metric a service business has. A firm that publishes retention is making a claim that a dissatisfied former client could contradict in public, and everybody reading understands that.

The bottom line. Marketing departments are organized around strangers. The money is in the people who already said yes and are quietly deciding, right now, whether to do it again. Insurance figured this out a century ago because it had no choice. You do have a choice, which is the problem.

Price the Churn Like a Reserve: Stop Guessing at Lifetime Value

Leave a reply

Marketers compute LTV by dividing one by the churn rate. Actuaries have spent a century learning why that number is a work of fiction.

1. One divided by churn is not a lifetime. It is a lifetime only if churn is constant forever, which it never is. Early customers leave fast, survivors leave slowly, and the blended monthly rate you are dividing into one is an average of two completely different populations behaving in opposite directions.

2. Build a triangle instead. Every cohort gets its own row and ages across the columns – month one, month two, month twelve. This is the chain ladder, the oldest tool in casualty reserving, and it exists precisely because the number you post early is never the number you end up paying.

3. The tail is where the money is. Casualty actuaries live in fear of adverse development – losses that keep growing years after the policy expired. Your retention curve has a tail too, and it is quietly deciding whether your acquisition spend was rational.

4. Cohorts confess. Averages lie. A flat blended retention number can hide a business where every recent cohort is worse than the one before it. Triangles make that visible in about ten minutes, which is roughly ten months before the blended number does.

5. The arithmetic of keeping people is not close. Reichheld and Sasser’s work at Bain put a 5% improvement in retention at somewhere between 25% and 95% more profit, depending on the industry. Nothing on the acquisition side has ever produced a range like that.

6. You need less data than you fear. Twenty-four months of billing exports and a spreadsheet is enough. This is not a data science project. It is arithmetic with the rows arranged properly, and most companies have never done it because nobody in the building was taught to arrange them that way.

7. Pick the point where the curve flattens. Every cohort chart has a month where departures stop and the survivors become an annuity. That month is your real payback horizon, and it is the only defensible input into what you are allowed to spend to acquire someone.

8. Then re-forecast, out loud, on a schedule. Reserves get reviewed quarterly and restated without embarrassment, because everybody understands the first estimate was an estimate. Marketing forecasts get defended to the death. Adopt the actuarial manners along with the method.

9. Show the client their own triangle. Nothing in a strategy engagement lands harder than a founder seeing their own cohorts laid out for the first time. It is their data, it takes an afternoon, and it usually ends the debate about the marketing budget without anyone having to win an argument.

The bottom line. An insurance company that guessed at its liabilities the way most software companies guess at lifetime value would be in receivership by the third year. The tools to do it properly are a hundred years old, free, and sitting in a discipline nobody in marketing has bothered to read.

Client Case Study: Insurance Services Firm Nobody Could Find

Leave a reply

Marketing & business strategy engagement · small insurance advisory services practice (anonymized at the client’s request)

The Client

An insurance advisory practice in a licensed, credential-gated field, two years old. Forty combined years at name-brand employers, signatures carrying personal liability, a hard-deadline busy season, and profit from the first invoice. Almost nobody knew it existed.

The Problem

Not capability. Distribution. The website sold “the credentialed professionals” without ever naming the credentialed professionals. Four services times nine buyer types made thirty-six pitches and no position. No listings, no reviews, no phone number, and dead links. And the question nobody asks out loud: what if the people here are unavailable in March?

What I Did, In Order

Baseline audit. Public footprint, positioning, search visibility, credibility leaks.
Five-angle evaluation. The business as operator, insider, buyer, financier, skeptic.
Twenty simulated buyer panels. 125 ranked moves; fifteen named the same first fix.
Competitor and market deep dive. Ten rivals, growth data, an empty auction.
Financial model. Forecast, seasonality, margin, ceiling, enterprise value.
The merge. Ten sections cut to two pages, then one, ending in a thirty-day list.

What the Numbers Said

45% of revenue lands in a four-month window against 45-to-75-day terms — cash-flow trouble disguised as a good year.
Significant revenue sat with 650 people who already knew them, not the website.
Automated data intake returns 20–30% capacity; a 20% rate rise beats any volume rise.

The Resolution

Forty moves, sequenced by cost and speed, many under five hundred dollars. Within thirty days, the firm had names, faces, and credentials above the fold, and a city and phone number on every page. Listings were claimed and a published fixed fee let buyers budget before calling.

A coverage agreement with peers turned the March question into visible professionalism. A 650-name list got worked fifty a week.

It also said what not to do: no discounting, no rebrand, no raising money to solve a distribution problem, no blog nobody reads. Clarity is mostly subtraction.

If This Sounds Familiar

If you are excellent at the work and invisible to the market, the gap is rarely talent. It is order of operations. Audit, pressure-test, model the money, then a short list of moves in the sequence that pays.

Let’s talk.

Translate the Jargon

Leave a reply

Why the person who explains it out-earns the person who knows it

Every industry has a word it says fifty times a day and has never once defined out loud. Ours is “social inflation.” Here is how to hand it to somebody who does not sell insurance for a living, and why that skill is worth more than being the smartest person in the room.

1. Nobody is confused. They are uncontexted.

When an American says soccer is boring, he is not stupid. He is watching without a frame. Tell him offside is just cherry-picking, that you cannot camp under the hoop waiting for the long pass, and he is arguing about it by halftime. One sentence. That is the entire job.

2. The curse of knowledge, measured.

In 1990, a Stanford researcher had people tap out famous songs on a tabletop. The tappers predicted listeners would name the tune about 50% of the time. Across 120 songs, listeners got 3 right. That is 2.5%. You hear the melody. Everyone else hears knocking.

3. Social inflation, in one breath.

The textbook version is “liability claim costs rising faster than economic inflation.” Nobody has ever repeated that sentence at a dinner table. The human version: your policy was priced for a 2010 jury and you are getting a 2026 jury. Same wreck, same injuries, different number.

4. Then hit them with the number.

The median nuclear verdict (anything over $10 million) ran about $21 million across 2013-2022, hit $44 million in 2023, and reached $51 million in 2024. That year brought 135 nuclear verdicts against corporate defendants, up 52%, totaling $31.3 billion, a 116% jump in twelve months. Verdicts above $100 million climbed 81.5%, to 49. Five of them cleared a billion dollars.

5. Stoppage time is reserve development.

Nothing enrages a new soccer fan like the fourth official holding up a board reading “4 minutes” and then playing six. The board is a minimum, not a promise. That is reserving. We post a number, the game keeps going, and the number moves; recent adverse casualty development has run around $15.8 billion. Every American who has ever screamed at a referee already understands our reserving problem.

6. Explain the engine, not the trivia.

Americans did not fall for the Premier League because someone explained the offside trap. They fell for it when someone said: imagine the three worst NFL teams get thrown out of the league. Relegation turned a meaningless February match into appointment television. So skip the actuarial triangles and explain the engine: third-party litigation funding, a business estimated near $18 billion worldwide, where investors buy a slice of a lawsuit the way they would buy a slice of a startup. That is who paid for the billboard.

7. Simple is not stupid, and that cuts both ways.

Study after study finds that denser, less readable academic abstracts collect more citations from other academics. Of course they do. They are written for people paid to decode them. You are not. Roughly 54% of American adults read below a sixth-grade level, and your smart, busy buyer is reading your email at 6:40 a.m. on a phone in a parking lot. He will forgive simple. He will never forgive confusing.

The expert gets deposed. The translator gets quoted.

Reporters, buyers, brokers and juries all repeat the person they actually understood. Say it in a sentence someone could hand to their spouse, and you have done what no white paper or acronym will ever do for you: made a stranger care.

Answer in Spanish: The Second Court Nobody Is Playing

Leave a reply

Your English authority stops at the net. In U.S. insurance distribution, the other court is empty.

1. Clay is not grass. Rafael Nadal won Roland Garros fourteen times. That bought him zero free points at Wimbledon, where he won twice. Same forehand, different surface, different bounce, different everything. Language models work the same way: each language is a partially separate knowledge space. Ranking #1 in English does not seed you in the Spanish draw. Nobody carries a ranking across the net for free. Surfaces are learned.

2. The 97-to-19 problem. Researchers taught a model new facts in English. Quizzed in English, it recalled them 97% of the time. Quizzed on the identical facts in another language, accuracy fell to 19%. Same model, same facts, one language apart. A second study estimates perfect cross-language sharing could lift accuracy up to 150% – a polite academic way of saying it does not happen today.

3. Untranslated means invisible. If your annuity explainer and licensing FAQ exist only in English, a Spanish query does not rank you low. It does not see you. You are not losing the point; you never entered the tournament. English is 49.7% of identifiable web content, Spanish just 6.0% (W3Techs, June 2026).

4. The crowd is already in the stadium. 44.9 million U.S. residents speak Spanish at home (2024 ACS) – 1 in 7 people age five and up. That group grew 21.3% from 2010 to 2024 while the population grew 11.2%. And 41% say they speak English less than “very well.” They are asking chatbots about deductibles, beneficiaries and IUL right now, in Spanish.

5. The number that should sting. Hispanic life insurance ownership fell from 51% in 2021 to 40% in 2025, the lowest of any group LIMRA tracks. Roughly 20 million Hispanic adults say they are underinsured, and 72% overestimate what term costs. That is not a demand problem – the demand is already there. It is an answer-supply problem, and the answers are sitting in the wrong language.

6. Do not practice against a ball machine. Auto-translating your English pages is hitting against a ball machine and calling it a match. “Rider,” “surrender charge,” “face amount” – rendered literally, they land as nonsense that no Spanish speaker and no model treats as authoritative. Write it natively, in the vocabulary agents actually use in Miami and Houston.

7. Serve first. Carlos Alcaraz did not inherit a grass game. He built one on purpose, then won Wimbledon in 2023 and 2024. The Spanish-language retrieval surface in insurance is wide open and lightly defended. Every month you leave it untranslated is a free point handed to whoever translates first.

AI Insurance Is Evolving Exactly Like Cyber Insurance Evolved

Leave a reply

If you sat through a cyber renewal in 2021, you have seen this movie. It is Groundhog Day, the clock radio just went off, and the same guy is insisting the risk is unknowable. Only the editing is faster this time.

1. Silence Is Not a Coverage Grant. Cyber losses were once covered mainly because nobody had excluded them. The industry called it silent cyber. Now it says silent AI, the same coin flip, except the coin talks back.

2. Act Two Is Always the Exclusion. ISO’s generative AI endorsements, CG 40 47 and CG 40 48, took effect January 1, 2026, with a definition broad enough to swallow anything producing text, images, audio, video, or code. Cyber took a decade to get here. AI took twenty-four months.

3. Cyber’s Numbers Explain Cyber’s Manners. Loss ratios at major carriers crossed 100% in 2020 and 2021. Marsh clocked premiums up 79% in 2022. AM Best put U.S. premium up 50% to $7.2 billion, with the standalone loss ratio dropping 23 points to 43%. Carriers did not find religion. They found a spreadsheet.

4. The Application Form Became the Security Program. By 2021, MFA went from best practice to no MFA, no policy. Microsoft says it blocks over 99% of automated account-compromise attacks. A two-page questionnaire hardened more networks than a thousand keynotes. Answer it wrong and you get a decline, a sublimit, or a rescission at claim time.

5. AI Is Running the Same Play. Carriers now want a model inventory, human-in-the-loop review, a tested kill switch, a named accountable executive, and provable data provenance, mapped to NIST AI RMF, ISO/IEC 42001, and AIUC-1. AI sublimits near 10% of the limit are the polite way of saying we are unsure.

6. New Carriers Arrive Before the Actuaries Do. Armilla lifted its Lloyd’s capacity to $25 million in January 2026. AIUC came out of stealth with $50 million tied to its own audit standard. Testudo began underwriting in January 2026. Munich Re’s aiSure has guaranteed model performance since 2018. That is how cyber started.

7. Actuaries Have the Hardest Job in the Building. Nobody can price a tail event on twenty months of data, so the loss triangle is being built from court dockets: Bartz v. Anthropic, reportedly $1.5 billion over training data; Moffatt v. Air Canada, where a chatbot’s improvisation became a contract; Mobley v. Workday on algorithmic hiring.

8. What to Do Before the Next Renewal. About 74% of small businesses use AI; almost none have read their endorsement schedule. Is CG 40 47 or 40 48 attached to my policy? Do I have affirmative AI coverage, or just the absence of the word no? Can I prove my governance, or only describe it? “We told everybody to be careful” is a sentiment, not a control.

Murray escaped Punxsutawney only once he used what the repetition taught him. We already have the cyber tape.