A digital will looks like a cheap product to run. Someone signs up, fills in a form, gets a document, and the software does the rest.

Then you look at the actual costs and find a lawyer reviewing the draft, an identity check, encrypted storage you have to keep for decades, support for people who log in twice a year and forget everything in between, and a payout to the partner firm that did the legal work.

This is a walkthrough of what one customer really costs and really earns. No spreadsheet required. The numbers below are illustrative, chosen to show the shape of the model rather than to report any company's actual figures.

The four numbers that matter

Everything else is detail.

What it costs to get a customer. Usually called customer acquisition cost, or CAC. All your sales and marketing spend for a period, divided by the customers you gained in that period.

What a customer pays you over their whole relationship. Lifetime value, or LTV. Not their first payment. Everything they will ever pay.

What is left after the cost of serving them. Gross margin. Revenue minus the direct costs of delivering the product.

How long until you get your acquisition cost back. Payback period. The one people ignore and then run out of cash.

If you know these four for your product, you know whether the business works. If you know only revenue, you know nothing.

Where the money comes from

Digital will products usually earn in one of three shapes, and the shape drives everything else.

One time fee. The customer pays once for a will and owns it. Simple to sell, hard to build a business on, because you have to find a new customer for every rupee of revenue and you still carry the storage cost for years.

Subscription. The customer pays yearly to keep the will current, keep documents stored, and keep beneficiaries updated. Harder to sell, much better economics, because the same customer pays repeatedly and the costs of serving them fall over time.

Fee plus subscription. A larger payment when the will is created, covering the legal review, and a smaller yearly fee for storage and updates. This matches the actual cost shape best. The expensive work happens once, at the start.

There is usually a fourth line: upgrades. A second will for a spouse, a more complex estate, cross border assets, or professional executor services. These matter more than people expect, because the customer is already acquired and already verified. The margin on an upgrade is far better than on a first sale.

What one customer actually costs you

Split the costs into two buckets, because they behave completely differently.

Costs that happen once, at the start

Legal review or partner fee. If a qualified professional reviews the will, this is your largest single cost by a wide margin. Whether it is a share of revenue paid to a partner firm or an in house cost, it dominates the first year.

Identity verification. A per check charge from a vendor. Small individually, real at volume, and it can double if a check fails and has to be repeated.

Payment processing. A percentage of the transaction. Bigger than people think on a larger one time fee.

Onboarding support. The messages and calls a new customer needs while setting up. Some customers cost nothing. Some cost an hour of a person's time. Use the average, not the ideal.

Costs that repeat every year

Storage and infrastructure. Encrypted document storage, backups in more than one place, and the servers. Per customer this is genuinely small. The catch is duration, which we will come back to.

Ongoing support. Password resets, updating a beneficiary, adding an asset, explaining what changed. Low per customer, but it never reaches zero.

Compliance and audit. Security audits, penetration tests, insurance, legal review of your terms. This is a fixed company cost, but you should spread it across customers to see the truth about your margin.

Renewal handling. Failed cards, expiry reminders, and the small share of renewals that need a human.

The important insight: the first year of a customer is expensive and the following years are cheap. That single fact drives every decision below.

A worked example

Illustrative numbers, in a generic currency unit, to show the shape.

A customer pays 100 to create a will and 20 a year after that to keep it stored and current.

Year one costs. Partner or legal review 40. Identity check 3. Payment processing 3. Onboarding support 5. Storage and infrastructure 2. Total 53.

Year one revenue is 100. Gross profit is 47, so a gross margin of 47 percent.

Every year after. Revenue 20. Costs: storage 2, support 3, renewal handling 1. Total 6. Gross profit 14, a margin of 70 percent.

Now say the average customer stays five years.

Total revenue: 100 plus four years at 20, so 180.

Total costs: 53 plus four years at 6, so 77.

Lifetime gross profit: 103.

If acquisition cost is 40, you make 63 per customer and your LTV to CAC ratio is roughly 2.6 to 1 on a gross profit basis. Workable, not exciting. Andreessen Horowitz notes that investors often use 3x LTV to CAC within five years as a rough benchmark of financial health, so this example sits just under the line.

If acquisition cost is 90, you make 13 per customer over five years and the business does not work, no matter how good the product is.

Two things fall out of this immediately.

The partner or legal cost is the biggest lever in year one. A shift from 40 to 30 adds 10 straight to profit on every single customer.

Retention is the biggest lever overall. The same customer staying eight years instead of five adds 42 of gross profit with no extra acquisition cost.

Why partner revenue share is the hardest number

If a law firm reviews the will, they take a share. Setting that share is the most consequential commercial decision in the model, and there is no clean answer.

Too low and firms will not do the work properly or will not do it at all. Estate work has real professional liability attached. A firm accepting a fee that does not cover its risk will deprioritise your cases, and you will find out through slow turnaround times.

Too high and you have no margin left to acquire customers with, which means you cannot grow, which means the partner gets no volume either.

Three things make the number easier to live with.

Pay for outcomes, not attempts. A share on completed matters aligns everyone. Paying for started matters funds abandonment.

Tier by complexity. A simple single asset will and a cross border estate with a trust are not the same work. One flat share means you are overpaying on the simple cases and underpaying on the hard ones, and firms will notice which is which.

Value the volume honestly. Partners accept a lower share per matter when the flow is steady and the admin is low. That is a real benefit you provide and it is fair to price it. Just do not promise volume you cannot deliver, because that trust is very hard to rebuild.

The cost everyone forgets: the long tail

A will is not a product you deliver and forget. You may hold that document for forty years.

That creates costs no normal software business carries.

Storage for decades, including backups in multiple locations, most of which will never be read.

Format survival. The file formats and encryption methods used today will need to still be readable in thirty years. That means periodic migration work.

Dormant accounts. Customers who paid once and stopped, whose documents you still hold because the whole promise is that you will hold them.

The claim itself. The moment a will is actually needed is the most support heavy interaction in the entire relationship, and it happens after the customer has stopped paying you.

Per customer per year, these are small. Across a large base over a long period, they are the difference between a profitable book and a slowly bleeding one.

Two consequences follow. First, one time pricing is dangerous for a product with a forty year obligation. Second, this is a strong argument for the subscription model, and worth explaining honestly to customers, because "we charge yearly because we hold this for your whole life" is a reason people accept.

Payback period, or how you run out of cash while growing

You can have healthy lifetime economics and still fail.

Spend 40 to acquire a customer who returns 47 of gross profit in year one, and you recover your money within the first year. That is fine. You can reinvest and grow.

Spend 40 to acquire a customer who pays 20 a year with a 70 percent margin and no upfront fee, and you wait roughly three years to break even on them. Every new customer is now a hole in your bank account that fills up slowly. Grow fast enough and you run out of money while every metric looks good.

This is the single strongest argument for charging something meaningful at the start rather than putting everything into a low annual fee. The upfront payment is not just revenue. It is what funds your next customer.

The levers, ranked

If you want the model to work better, this is the order of impact.

1. Retention. Every extra year is nearly pure margin. Reminders that are useful rather than annoying, a genuine reason to log in each year, and easy updates when life changes.

2. Partner cost per matter. Tiering by complexity and reducing the admin load per case moves the biggest cost line.

3. Acquisition cost. Content and referrals compound. Paid acquisition does not. In a category where people search when a life event happens, being the answer that already exists is cheaper than buying the click.

4. Upgrade rate. Selling a second product to an existing verified customer has no acquisition cost and no verification cost. This is the most underworked lever in most estate products.

5. Support cost per customer. Better empty states, clearer copy, and self service for the five things people always ask.

Note what is missing from that list: raising the price. It works, and it is usually the first idea, but in a category built on trust it is also the fastest way to increase churn. Try it after the other five.

The number to put on the wall

If you track one thing, track gross profit per customer over their expected lifetime, next to what you paid to acquire them.

Not signups. Not revenue. Those two numbers, side by side, updated monthly.

A digital will business fails in one of two ways. It pays too much for customers who do not stay, or it earns too little per customer to survive the decades it promised to be around for. Both are visible in that comparison long before they show up in the bank account.

Frequently Asked Questions

What is unit economics for a digital will product?

It is what one customer costs to acquire and serve, against what they pay you over their whole relationship. The four core numbers are acquisition cost, lifetime value, gross margin, and payback period.

Is a one time fee or a subscription better for digital wills?

A subscription, or a larger upfront fee with a smaller annual fee, fits better. The product carries storage and support obligations for decades, which a single payment does not fund.

What is the largest cost in a digital will?

The legal review or partner fee in the first year, by a wide margin. Everything else is small next to it.

What is a reasonable partner revenue share?

There is no single number. Tier it by matter complexity, pay on completed matters rather than started ones, and price the value of steady volume and low admin honestly.

Why does payback period matter more than lifetime value?

Lifetime value tells you if the business works eventually. Payback period tells you whether you can afford to grow now. Companies with good lifetime value and slow payback run out of cash while growing.

Which lever improves digital will economics the most?

Retention. Each additional year a customer stays is almost pure gross profit, because the expensive work all happened in year one.

Sources

Further reading: