Customer acquisition cost by channel and line of business

Two carriers buy the same auto insurance lead within minutes of each other. One pays $40, the other $400. Same driver, same ZIP code, same coverage request. The difference is the channel it arrived through, and what each buyer already knows about how that channel converts and how long the resulting policy stays on the books.

CACCACCustomer Acquisition Cost (CAC) is the total sales and marketing spend divided by the number of new customers gained in a period. It measures how efficiently you grow.View full definition → (customer acquisition cost) is the founding number of any insurance marketing P&L, and it moves by an order of magnitude across channels and across lines of business (LOB: auto, home, life, commercial). Held as a single blended figure, it hides both the profit and the waste.

What CAC actually includes

CAC is more than media spend. A fully loaded insurance CAC carries:

• Media and ad spend (search bids, comparison-site fees, affiliate commissions)
• Agent or broker commission tied to the sale
• Call centre and quoting-platform cost per bound policy
• Marketing technologyMarketing technologyThe connected set of software tools a marketing team uses to plan, run, measure and automate campaigns across channels.View full definition → and attributionattributionA framework for assigning credit to the touchpoints that contributed to a conversion, so you can measure which channels and interactions actually drive results.View full definition → tooling, amortised per lead
• Creative production and campaign management overhead

Formula:

CAC = Total acquisition-related spend (channel + fulfillment) / Number of new policies bound

Note "bound," not "quoted" or "clicked." Plenty of insurers report CAC on leads or quote starts, which flatters the number by whatever the drop-off happened to be that month. The P&L unit is cost per *bound policy*, because that is what starts earning premium.

Two accounting traps are worth catching before they contaminate a year of reporting. Rewrites and reinstatements come first: a policy cancelled for non-payment and reinstated three weeks later often re-enters the system as new business, padding the denominator and making CAC look better than it is. Then early cancellation. A policy that lapses before the second premium payment has consumed the full acquisition cost and returned a sliver of premium. Careful teams compute CAC on policies still in force at 60 or 90 days, and the gap between that number and the raw one is usually widest on the channels that looked cheapest.

Why the same lead costs 10x more in one channel

Paid search (auto insurance quote)

A consumer searches "cheap car insurance [city]" and clicks a paid ad. Insurance has sat at the top of Google's most expensive verticals for well over a decade, because carriers, aggregators and lead brokers all bid on the same handful of high-intent phrases. Reported CPCs (cost per clickcost per clickCost Per Click (CPC) is the average amount you pay each time someone clicks your ad. It is a core pricing metric for paid search and social advertising.View full definition →) for competitive US auto terms run in the $20 to $50+ range (estimate, varies heavily by state and season; see benchmark reports such as WordStream's advertising benchmarks, from a firm that sells search advertising tooling).

Run the arithmetic before the optimism sets in. At $35 a click and a 2% click-to-bind rate (the funnelfunnelThe customer journey from awareness to purchase, typically Awareness, Interest, Consideration, Decision, Action, with prospects narrowing at each stage.View full definition → lesson maps where those clicks go missing), media alone is $1,750 per bound policy. Nobody survives that, which is why most of the real work in paid search is subtraction: excluding states and ZIPs where the carrier is not filed or does not want the risk, negative keywords for searches it cannot serve, device and daypart caps. Filtered tightly, fully loaded CAC lands in the $150 to $400+ band. Left unfiltered, paid search is the fastest way to buy quotes a carrier has no legal ability to convert.

Comparison-site referral

Price-comparison platforms aggregate demand and charge per click, per lead or per bound policy, with commercial terms that vary by carrier and volume commitment. Intent is already high, so conversion beats cold search and estimated CAC often comes in around $80 to $200 (estimate). The cost shows up later: shoppers acquired mid-comparison renew worse than the rest of the book, so a channel that wins on CAC can lose on the multi-year math. How to compete there without buying your own churn is the aggregator lesson's subject.

Affiliate partnership

An affiliate (a personal finance site, a fintech app, a content publisher) embeds a quote widget and earns commission per bound policy, often $20 to $60 for auto (estimate) and considerably more for life. On a per-policy basis this is frequently the cheapest channel, because the audience trust is pre-built and the insurer pays on success rather than on clicks. Volume is the constraint, not price.

Captive agent lead

A captive agent (selling exclusively for one carrier) generates business through local marketing, referrals and their own book. The media line looks small; the fully loaded number has to carry base compensation, commission override, office support and whatever local marketing subsidy the carrier funds. That routinely puts CAC per new auto policy at $300 to $500+ in year one (estimate), against materially better retention and bundling rates.

Geico built the opposite structure: near-total direct distribution, no captive field force, so its acquisition cost is overwhelmingly media plus contact centre. It spends over a billion dollars a year on advertising in normal years, and when underwriting results deteriorated in 2022 it cut that spend hard, with policies in force falling for several consecutive quarters while margin recovered. The acquisition budget is the fastest lever an insurer has on its own growth curve, and it works in both directions inside two or three quarters.

A worked example

Say a carrier runs three channels for personal auto in the same quarter:

Blended CAC: ($180,000 + $120,000 + $40,000) / (600 + 800 + 1,000) = $340,000 / 2,400 = $141.67

The blended $142 hides that paid search costs seven times affiliate for the same product. The naive response, move everything to affiliate, fails on supply. Partner traffic is finite, and the marginal cost of the next 200 policies from any channel exceeds its average: affiliate CAC of $40 at 1,000 policies may be $120 at the 1,400th. Budget decisions belong to marginal CAC, not the average printed in the dashboard.

Root Insurance shows the shape of that shift. After years of buying performance media direct to consumer, it cut sales and marketing spend sharply in 2022, put weight on embedded partnerships (its Carvana tie-up) and independent agents, and returned to net profitability in 2024. Cheaper distribution was part of the path, not a substitute for the loss-ratio work alongside it.

CAC varies by line of business too, not just channel

• Auto: highest competitive intensity, shortest cycle, lowest average premium, so CAC as a share of first-year premium gets scrutinised hardest.
• Home: usually sold next to auto, which raises an allocation question with no clean answer. Charge the whole bundle cost to auto and home looks free, so it never earns a budget. Split it evenly and neither channel reconciles to the P&L. Pick a rule, document it, and keep it stable for years, because changing allocation mid-stream makes every trend line unreadable.
• Life: highest CAC in absolute dollars (many hundreds per policy, estimate) given long cycles, medical underwriting steps and low online conversion. Carriers tolerate it because of the multi-year value the lifetime valuelifetime valueLifetime Value: the total revenue (or profit) a customer generates throughout their entire relationship with your business.View full definition → lesson models.
• Commercial and SME: measured per bound account rather than per policy, often four figures, with broker-mediated distribution and multi-year relationships behind it.

The fast sanity check is CAC against first-year premium, not against long-run value. Take a $1,200 annual auto premium at a 65% loss ratio and 25% non-acquisition expense ratio: roughly $120 of margin in year one. A $300 CAC on that policy pays back somewhere in year three, so anything that shortens the expected tenure below that destroys value no matter how good the CAC looked at bind.

European context

Structure is similar in the EU and UK, the rules are not. The UK's FCA (Financial Conduct Authority) general insurance pricing practices reforms, in force since January 2022, ban price walking (charging loyal renewal customers more than equivalent new ones). The second-order effect on acquisition is direct: carriers can no longer fund loss-making new-business pricing out of an overpriced renewal book, so each acquisition channel has to stand closer to its own economics. Price-comparison websites also dominate UK personal lines distribution far more than in the US, which drags national CAC benchmarks toward comparison-site fee structures and makes cross-market comparisons unreliable unless mix is disclosed.

A simple attribution snippet

Even non-technical marketers should recognise this logic, since it underpins most CAC dashboards:

# Simplified fully-loaded CAC calculation by channel
channels = {
    "paid_search": {"spend": 180000, "bound_policies": 600},
    "comparison_site": {"spend": 120000, "bound_policies": 800},
    "affiliate": {"spend": 40000, "bound_policies": 1000},
}

for name, data in channels.items():
    cac = data["spend"] / data["bound_policies"]
    print(f"{name}: CAC = ${cac:.2f}")

Deliberately simple. Real systems also allocate shared overhead (contact centre, tech stack) proportionally, which is where most CAC arguments between marketing and finance actually start.

🎬 [VIDEO: "Customer Acquisition Cost Explained" - https://www.youtube.com/results?search_query=customer+acquisition+cost+explained - a primer on CAC calculation logic applicable across industries, useful for building the baseline before layering insurance-specific channel nuances]

Key Takeaways

• CAC must be fully loaded (media, commissions, fulfilment, tech) and counted per bound policy that survives its first 60 to 90 days, or reinstatements and early lapses will flatter it.
• The same lead type can cost 3 to 10x more depending on channel, and unfiltered paid search on insurance keywords can price out entirely once you multiply a $35 click by a 2% bind rate.
• Blended CAC hides channel-level inefficiency, and average CAC hides the marginal cost of the next tranche of volume. Break it out by channel and by line of business before moving budget.
• Bundle allocation between auto and home is a judgement call: choose a rule and keep it, because changing it rewrites your own history.
• Regulatory change reshapes acquisition economics, not only underwriting: the FCA's 2022 reforms removed the renewal-book subsidy that once paid for cheap new business.