The insurance value tree
Insurance value is underwriting result plus investment return. The tree separates the two, then splits underwriting into the price you charged for a risk and what that risk actually cost.
01 The same tree, this industry
Where the money is made and lost here
The structure does not change: value is profit plus how well that profit becomes cash, profit is revenue minus cost, and revenue is price times quantity. What changes is which drivers sit underneath each branch, and which system holds them. If you have not read the general version, start with the Enterprise Value Tree and come back.
The distinguishing feature here is time. You set the price before you know the cost, sometimes years before, so every branch of this tree is really a question about the quality of an estimate — and about how quickly you learn that the estimate was wrong.
Every driver below names the metric it lands on. Follow one and you get its formula, the system the number lives in, and the ways it is commonly misread.
02 Price
What decides the price you actually get
Rate adequacy
Whether the premium charged covers the expected loss plus expenses. The single question the whole sector turns on.
Price realisation net_revenue ÷ list_revenuePriceRisk selection
What you chose to write. Better selection at the same rate is the cheapest margin improvement available.
Contribution margin (net_revenue − variable_cost) ÷ net_revenuePriceBroker commission and discount
The gap between technical price and written price, granted through a channel that has its own incentives.
Average net price net_revenue ÷ unitsPriceRenewal pricing
Where most of the book is repriced, usually with less analysis than new business.
Net revenue retention (starting_revenue + expansion − contraction − churn) ÷ starting_revenuePriceDeductibles and cover structure
The shape of the policy moves loss cost more than the rate does, and is decided transactionally.
Contribution margin (net_revenue − variable_cost) ÷ net_revenuePriceCross-sell and bundling
Multi-policy customers have materially different retention and loss economics.
Attach rate orders_containing_the_attached_item ÷ eligible_orders03 Quantity
What decides how much you sell
New business volume
Written premium by channel and segment, against the appetite that was supposed to govern it.
Active customers count(distinct customers with at least one purchase in the window)VolumeRetention
Renewal is where the book is really made, and where a rate increase meets a customer decision.
Customer retention rate customers_retained ÷ customers_at_startVolumeQuote to bind conversion
Where price, appetite and journey friction all show up in one ratio.
Lead to opportunity conversion leads_accepted_as_opportunities ÷ qualified_leadsVolumeBroker and channel productivity
A small number of intermediaries usually produce most of the book, and most of the concentration risk.
Win rate opportunities_won ÷ opportunities_closedVolumePolicies per customer
The cheapest growth in the sector, and the strongest retention lever.
Attach rate orders_containing_the_attached_item ÷ eligible_ordersVolumeMarket share by segment
Growth against a market that is repricing at a different speed from you.
Market share your_units (or value) ÷ total_market_units04 Cost
What it takes to operate
Claims cost and loss ratio
The dominant cost line, known late and estimated in the meantime.
Contribution margin (net_revenue − variable_cost) ÷ net_revenueCostClaims handling expense
The cost of settling, which trades directly against settlement quality and customer retention.
Cost to serve (delivery + service + returns + order_handling) ÷ customers or ordersCostFraud and leakage
Paid claims that should not have been paid at that amount — detectable in patterns across policy and claim data.
Data quality score weighted pass rate across completeness, validity, uniqueness, timeliness and consistency rulesCostAcquisition cost
Commission plus marketing, recovered over a policy lifetime that retention determines.
Customer acquisition cost (sales_cost + marketing_cost) ÷ new_customers_wonCostExpense ratio
Everything that is not a claim, measured against premium and stubbornly fixed.
Operating margin operating_profit ÷ net_revenueCostReinsurance cost
The price of transferring volatility, which is a capital decision priced as an expense.
Contribution margin (net_revenue − variable_cost) ÷ net_revenue05 Cash
Where the cash actually sits
Reserves and IBNR
Money set aside for claims not yet reported. An estimate, revised, that moves reported profit directly.
Forecast accuracy 1 − ( Σ|actual − forecast| ÷ Σactual )CashPremium receivables
Collected through intermediaries, with a settlement lag the insurer does not control.
Days sales outstanding (accounts_receivable ÷ net_revenue) × days_in_periodCashClaims payment timing
When cash leaves, which is a customer experience decision as much as a treasury one.
Days payables outstanding (accounts_payable ÷ cogs) × days_in_periodCashInvestment float
Premium held between collection and payment — the asset side of the underwriting business.
Free cash flow operating_cash_flow − capital_expenditureCashCapital and solvency
The regulatory constraint that decides how much risk the balance sheet can carry.
Return on capital employed EBIT ÷ (total_assets − current_liabilities)CashReserve development
Whether last year's estimates are proving right, which is the honest audit of the whole pricing branch.
Forecast accuracy 1 − ( Σ|actual − forecast| ÷ Σactual )06 Where the numbers live
The systems behind the branches
A value tree is only usable once each box maps to a system and a record. These are the six that matter most in insurance: what each one is actually for, the records inside it the tree depends on, and which branch it feeds.
| System | What it holds | Key records | Feeds |
|---|---|---|---|
| Policy administration | The contract of record: cover, terms, endorsements and the premium actually charged. | Policy, Risk, Endorsement, Premium, Cover section | Price, quantity, cash |
| Claims | Notification through settlement, with the reserve movements that drive reported result. | Claim, Reserve movement, Payment, Recovery, Peril | Cost, cash |
| Rating and underwriting engine | The technical price and the appetite rules, against which every written price is a deviation. | Rating factor, Technical price, Referral, Decline | Price |
| Broker / distribution | Who placed what, on which terms, and the commission and bordereaux behind it. | Intermediary, Agreement, Bordereau, Commission | Price, quantity, cash |
| Actuarial and reserving | Development triangles, IBNR and the assumptions the balance sheet rests on. | Triangle, Cohort, Assumption set, Model run | Cost, cash |
| Finance and regulatory | The ledger and the solvency returns, which have to reconcile to all of the above. | GL account, Solvency return, Capital requirement | Cash |
Almost every hard question in insurance needs two of these joined. That join — not the calculation — is the work.
07 Where it leaks
Value lost between two systems
These are the losses that no single system can see, because the evidence is split across two of them. Each one is a real number that stays invisible until the join exists — which is why the tree is an integration exercise before it is an analysis one.
| Where value leaks | Why it happens | The join that finds it |
|---|---|---|
| The policyholder who is also the claimant, twice | Policy and claims systems hold the same person under different identities, so cross-policy fraud patterns and true customer profitability are both invisible. | Resolve party identity across policy, claim and intermediary records |
| Technical price versus written price | The rating engine produces a price; the front line writes another. The deviation is the margin story and it is rarely reported as a distribution. | Join technical price to written premium at policy level, every renewal |
| Broker performance measured on volume | Commission is paid on premium written while profitability depends on the claims that follow, two systems and eighteen months later. | Join intermediary agreements to developed loss ratio by cohort |
| Claims leakage invisible between handlers | Settlement variation by handler, peril and region is a large, recoverable cost that no single claim reveals. | Join settlement amounts to reserve history, handler and claim characteristics |
| Reserving learns slower than pricing needs | Pricing is refreshed annually from data the actuarial cycle has not yet developed, so last year's error is repeated before it is known. | Join emerging claim experience back to the rating factors that priced it |
08 Worked example
Growth that arrived as loss ratio
Gross written premium grew 11% and the combined ratio worsened by two points. The tree separates it: the growth came through one channel writing consistently below technical price, retention was bought with discount on the profitable segment, and the reserve development on the prior year turned adverse in the same peril.
| Component | Effect | What sits behind it |
|---|---|---|
| Premium growth | +€8.9m | 11% growth, concentrated in one channel |
| Rate against technical | −€3.6m | Written consistently below the rating engine |
| Renewal discounting | −€1.8m | Applied to the profitable segment |
| Claims frequency | −€2.9m | Segment mix, not weather |
| Prior year development | −€2.2m | Adverse on the same peril |
| Net | −€1.6m | Growth that cost two points of combined ratio |
Illustrative figures, shown to demonstrate the split. The point is the shape of the walk, not the numbers — on your own data the same bridge is built from your ledger.
09 Diagnostics
Six questions to ask in insurance
Ask them of your own team before anyone asks them of you. In most organisations at least two of these cannot be answered without a manual exercise, and those two are the plan.
- What is the distribution of written price against technical price, by channel?
- Can you measure profitability by intermediary on developed losses rather than on premium written?
- Do policy and claims systems agree on who the customer is?
- How quickly does emerging claims experience reach the pricing team — in weeks or in annual cycles?
- What proportion of claims cost is leakage, and could you evidence that number?
- Which segments are growing, and are they the segments your reserve development says are profitable?
10 The metrics behind it
Definitions for every box
11 Questions
Frequently asked
Why does insurance need entity resolution so badly?
Because the same party appears as a policyholder, a claimant, a driver and sometimes a beneficiary, across systems that were never designed to agree. Fraud detection, customer profitability and cross-sell all depend on that resolution being right.
Is the pricing branch really about data quality?
Largely, yes. Rate adequacy depends on knowing the risk you wrote, and that knowledge lives in the completeness and accuracy of the risk data captured at quote. Missing or defaulted rating factors are priced as if they were average, and they are not.
How does the cash branch differ from other sectors?
The cash arrives before the cost is known, which inverts the usual working capital question. The scarce resource is capital held against uncertainty, so reserve accuracy is a cash metric as much as an accounting one.
Where do most insurers start?
With the party record, then with technical price against written price. The first makes every other analysis possible; the second usually pays for the programme within a renewal cycle.
See this tree on your own data
Connect the systems above, define each box once, and the tree stops being a slide.