Insurance agency: where the profit really is

“It was my best year out of seven. Revenue 3 297 840 ₴, net profit 745 200 ₴, bonuses for the agents. And the portfolio ended up 92 policies smaller than it started. Back then I did not know that mattered more than the profit.”

Yurii Hnatenko runs an insurance agency in Kremenchuk. Motor comprehensive, property, a little life cover, six agents on the payroll. Thirteen years in insurance, seven of them as the owner. The agency is profitable, there are no debts, revenue has grown three years running. And the portfolio is shrinking all the while.

An agency of 1 600 policies

The average premium on a policy is 14 000 ₴ a year. The agency takes a commission out of it, and it is the commission, not the premium, that is the revenue of the business.

There are two commissions, and that is the main thing the policy ledger never shows.

The first year and the renewal are different businesses

For the first year the insurer pays 25% of the premium, which is 3 500 ₴. For every subsequent renewal of that same policy — 12%, which is 1 680 ₴. More than double the difference, and it is written into every contract with every insurer.

• new policy: 3 500 ₴ of commission
• renewal: 1 680 ₴ of commission, every year
• cost of winning a new client: 1 900 ₴
• cost of a renewal: 280 ₴

A renewal is neither a new sale nor the continuation of an old one. It is a separate event with its own commission, its own costs and its own probability — and in an agency's books it almost never stands as a line of its own.

Two commissions, not one

Commission is not profit. You have to take off what it cost to earn it.

Here is where the first distortion hides. The commission on a new policy is 2.08 times bigger, and that is visible at once. The cost of it is 6.79 times bigger, and that is never visible, because advertising and agents' pay sit in the books as general costs of the agency.

What an old client costs is not what a new one costs

Winning a new client costs 1 900 ₴: advertising, an agent's time on meetings, calls to the ones who never bought. A renewal costs 280 ₴ — one call, one meeting to sign, sometimes not even that.

A sevenfold gap in costs eats almost the entire advantage of the first year.

Yurii counted it for the first time in his seventh year. Not because he could not before — the cost of winning clients was simply never tied to a particular policy, and to split it he had to pull the year's advertising invoices and break down how the agents spent their time.

What is left after the costs

A new policy gives 1 600 ₴ of contribution. A renewal gives 1 400 ₴.

The gap between them is 200 ₴. In the policy ledger the first looks twice as interesting, because the ledger carries commissions, not contributions. But it is 1 600 ₴ once against 1 400 ₴ every year for as long as the client stays.

There is one more detail that is easy to miss. The 1 900 ₴ of acquisition is 1.36 times more than the contribution of a single renewal. So losing an old client and winning a new one in his place is an operation that loses money in itself, even if the new one eventually brings more.

A new client pays once. An old client pays every year.

The agency's year in numbers

22.6% net. For an agency with no products of its own and six agents that is a strong result, and for two years Yurii showed this report as proof that everything was being done right.

The fourth and fifth lines of the table carry one number across both columns. That is how the owner sees them too: the agency's fixed costs do not split between new policies and renewals, because the office, the CRM and the licences are needed equally for both.

Where the revenue actually comes from

Renewals give 1 827 840 ₴ out of 3 297 840 ₴ of revenue. That is 55.4%, so more than half, even though the commission there is less than half the size.

In contribution the gap is wider still. Renewals bring 1 523 200 ₴ against 672 000 ₴ from new policies — 2.27 times more.

Renewals deliver 69.4% of the whole contribution. That is close to two thirds, and no report in the agency ever showed that number.

“All my life I believed an agency lives on new sales. Our meetings were about new business, the bonuses were for new business, and the best agent was the one who brought in the most. And two thirds of the contribution came from a routine nobody ever talked about.”

A month's revenue says nothing about the portfolio

The monthly report folds new policies and renewals into one line called commission income. The line grows, and that is true. It simply does not show what exactly is making it grow, or what is happening to the client base underneath.

There is a flip side. If the agency suddenly stopped selling new policies, the revenue would not fall at once but over three years, as the portfolio lived out its term. It comes back just as slowly, and that is exactly why the consequences of a bad year only become visible two years later.

Why the month looks good

The month almost always looks good.

Policies expire evenly through the year, so renewals come in a steady flow. New sales have no sharp dips either. Together they draw a stable chart, and in that stability the owner sees steadiness and does not see that underneath it the portfolio is changing its composition.

• commission income is growing — visible
• the number of new policies hits plan — visible
• client retention — not counted
• the change in the size of the portfolio — not counted

There is another side to this that people do not like to discuss. A renewal arrives not because the client is delighted but because he cannot be bothered to look elsewhere. Part of retention rests on inertia, and the moment a competitor calls first, the inertia works against the agency.

How long a client lives

Retention in Yurii's agency is 68%. Which means that out of every ten clients almost seven are still there a year later, and three go to a competitor, straight to an insurer, or stop insuring anything at all.

Yurii knew the 68% roughly, by feel. Nobody had counted it exactly, because that means comparing the list of policies that expired with the list of those re-signed, and the two lists lived in different files.

Retention is a multiplier, not a percentage

At 68% the average client lifetime is 3.125 years. The formula is simple: one divided by the churn. But that multiplier moves in a way that is not straight, and the entire economics of an agency hides right there.

A client lives a little over three years. So after the first sale he will bring about two renewals.

Lifting retention from 68% to 75% is not a seven per cent improvement. It changes the multiplier from 3.125 to 4 — almost a whole extra year of life for every client in the portfolio.

What one client brings

A client brings the agency 4 575 ₴ over the whole relationship. Of that, 1 600 ₴ comes from the first sale and 2 975 ₴ from renewals.

Renewals against the first sale

Renewals bring in 1.86 times more than the first sale. That is 65% of everything the agency ever gets from a client.

The arithmetic here is rough but honest. A client lives 3.125 years, so after the first sale he signs 2.125 renewals. Each at 1 400 ₴ of contribution — 2 975 ₴ together, and that is more than the 1 600 ₴ the agent brought in from the first meeting.

Almost two thirds of the earnings sit in the part of the work nobody plans and nobody gets praised for.

The agent sells the first year. The agency lives off the next three.

At 68% retention the agency gets 4 575 ₴ out of a client. At 75% — 5 800 ₴. Same client, same premium, same commission from the insurer: the 1 225 ₴ of difference is made purely by how many years he stays.

Who the bonus rewards

The agent sees his commission, not the LTV

An agent gets a percentage of the commission he brought in. A new policy pays him twice what a renewal of the same policy does, because that is exactly how the insurer's commission differs. So the agent calls the new client rather than the old one — and he is being perfectly rational about it.

The bonus system reproduces the structure of the insurer's commissions. It does not reproduce the structure of the agency's profit, and those are two different things.

“I built that system myself. I paid more for new policies because they felt harder. They really are harder. But I was paying for difficulty, not for the money they bring in.”

Yurii was afraid to rework the bonuses for two years. The agents had a strong argument: a new client has to be found, convinced and walked through, while a renewal calls you itself. The argument is correct, but it is about the difficulty of the work, not about its value to the business.

The portfolio melts quietly

Here is the arithmetic Yurii did not do for seven years.

Churn of 32% on 1 600 policies is 512 clients leaving in a year. New arrivals number 420. So more leave than arrive, by 1.22 times.

• policies at the start: 1 600
• left: 512
• arrived: 420
• at the end: 1 508 — minus 92 policies, or 5.8% in a year

“I saw those 92 policies and thought I had got the formula wrong. I recounted it three times. Then I pulled the ledger for three years and saw it had been the same the year before, and the year before that. The revenue was growing, and I was not looking.”

The portfolio dries up while the revenue grows. That can happen because new policies carry twice the commission, and 420 new ones cover the commission of the 512 lost in the current year. Next year the lost ones pay nothing at all.

Over seven years that gives a picture quite unlike the one a single report suggests. The portfolio does not collapse and sends up no warning signal — it just gets 5.8% smaller every year, and you cannot hear it while new sales keep covering the commission.

The retention programme

Yurii worked out what lifting retention from 68% to 75% would give. The programme is simple: a named manager for the portfolio, a reminder two months before a policy expires, a mandatory call after a claim is paid. 240 000 ₴ a year all in — the salary of one administrator.

At 75% there would be 1 200 renewals instead of 1 088.

The programme costs the same at any level of retention

The 240 000 ₴ pays for working the whole portfolio, not for each renewal saved. So the programme adds 112 renewals, and each of them costs 2 142.86 ₴. The contribution of a renewal is 1 400 ₴.

Why it did not pay off

The programme adds 156 800 ₴ of contribution. It costs 240 000 ₴.

The net effect in the first year is minus 83 200 ₴.

“That looked to me like the end of the conversation. I counted, it came out negative, I closed the file. I came back to it a month later only because I could not explain to myself why the portfolio was still melting.”

The calculation is correct and at the same time incomplete. It measures one year, while retention works on all the years that follow, and that is the difference between a cost and an investment.

The difference between a cost and an investment here is not in the bookkeeping but in the term. A cost closes inside the same year. An investment in retention keeps working for as long as the client stays, which is at least four years — and measuring it over one year means comparing the full price with a single annual slice of the benefit.

The line where it starts to make sense

The threshold is counted in renewals, not in percentages

Divide 240 000 ₴ by 1 400 ₴ of contribution and you get 172 renewals. That is how many the programme has to add to pay for itself inside one year. That is retention of 78.75%, not the 75% Yurii was planning to reach.

The gap between 75% and 78.75% looks trivial. In renewals it is 60 policies, and it decides whether the programme pays back in one year or in two.

But there is a second question, and it matters more.

Yurii counted both options and left both on the table. You can aim for 78.75% and get your money back inside a year. Or you can stop at 75%, accept the loss in the first year and watch the portfolio instead of the report.

What Yurii changed

At 75% the portfolio loses 400 policies rather than 512. New arrivals are the same 420.

• at 68%: 512 left, 420 arrived, portfolio minus 92
• at 75%: 400 left, 420 arrived, portfolio plus 20
• client LTV: 4 575 ₴ against 5 800 ₴
• LTV gain: 1 225 ₴, or 26.8%

Seven percentage points of retention turn the portfolio from drying up to growing. In the first year's profit report that looks like minus 83 200 ₴. In the size of the portfolio three years out it looks like something else.

“I launched the programme at a loss in the first year, and for the first time in seven years I know exactly what I am paying for. I am not buying renewals at two thousand one hundred each. I am buying the portfolio stopping its decline.”

Not one of these figures was new. All of them had been sitting in the policy ledger for seven years, waiting to be split into two columns.

Along with the programme the agency changed the bonuses. A renewal now earns the agent a percentage comparable to a new policy, even though the insurer's commission there is smaller. The agency carries the difference itself — deliberately, because it knows a renewal gives 1 400 ₴ of contribution against 1 600 ₴ from a new policy, and that the two run for a different number of years.

The money does not vanish — you simply do not see it

“For thirteen years I counted commission. The commission was right, the report was right, the profit was real. I just never once counted how many clients I was selling to for the first time and how many for the second. Turns out that is the whole business.”

Next year's portfolio is 1 620 policies instead of 1 508. A difference of 112 policies does not look dramatic in one year. Over three it decides whether the agency grows.

Yurii's agency did not raise its revenue and did not raise its commissions. The premium stayed 14 000 ₴, the commissions 25% and 12%, the new policies the same 420. The only change is that retention became a figure somebody looks at every month, instead of a word used to describe good clients.

The revenue grew for seven years and it is growing still. It is just that now a second number stands next to it, and the two do not always agree.

How to see this in your own agency

Everything above rests on one split the policy ledger does not make: first-year commission and renewal commission are two different businesses with different economics. While they sit on a single line called commission income, the agency sees a good month and misses that the portfolio melts by 32% a year. When new sales, renewals and the cost of each are counted separately on their own, retention becomes a number rather than a feeling — which is exactly why owners of small and medium businesses use Finmap: to see the financial picture rather than piece it together after the fact.

Read also

• Agency Client LTV: How Much a Client Brings Over Their Lifetime
• Meal subscription: why good unit economics brings no cash

Topic foundation: Management Accounting Explained: What It Is and Why Owners Need It