Most startup metrics advice assumes your customer uses the product often. Daily active users, session length, weekly engagement. Those numbers work for a messaging app.

Estate planning does not behave like that. A good customer might log in three times in the first month, then twice a year, then not at all for a decade, and the product still worked exactly as intended. Someone who logs in every day is probably confused, not delighted.

So the standard dashboard is not just unhelpful here, it actively points the team in the wrong direction. Here is a framework that fits the category instead: what to look at weekly, what to look at monthly, what to look at quarterly, and which popular metrics to ignore.

The one question every metric should answer

Before any list, a filter. Every metric on your dashboard should answer one question:

Will this plan work when the family needs it?

That is the product. Not signups, not sessions. A will that is incomplete, out of date, unfindable, or attached to a beneficiary nobody can contact is a failed product, even if every growth number looked good.

Metrics that connect to that question stay. Metrics that do not come off the board.

Weekly: the operating numbers

Weekly metrics exist to catch problems while they are still small. Keep the list short enough that the whole team can hold it in their head.

New plans started. Your top of funnel. Watch the trend, not the number.

Plans completed. A started plan is worth nothing. This is the number that matters.

Completion rate. Completed divided by started, for the same cohort. When this moves, something in the product or the messaging changed.

Where people stop. Not a single number but a short list: the three steps where the most people drop out this week. This is the most actionable thing on the weekly board.

Partner turnaround time. For anything sent to a legal partner, how long from submission to return. Slow partners are invisible in revenue for weeks and very visible in customer trust immediately.

Support tickets by type. Volume matters less than the mix. Ten tickets about the same screen is a product bug wearing a support costume.

Failed identity verifications. These are customers who tried to pay you and could not finish. High failure rates are usually your instructions, not your users.

Seven numbers. Ten minutes to review. If your weekly review takes an hour, it is not a weekly review, it is a monthly one happening too often.

Monthly: the health numbers

Monthly is where you look at whether the business works rather than whether last week went well.

Plans completed this month, and the total live. The cumulative number is the real measure of the company.

Activation rate. Of the people who signed up this month, how many reached the point where the product actually protects them? Define that point precisely and do not move the definition to make the number look better.

Acquisition cost by channel. Total spend divided by customers gained, split by where they came from. Blended acquisition cost hides the channel that is quietly wasting money.

Renewal rate. For subscription products, the share of customers due to renew who did. In a category built on trust, this is your quality score.

Revenue and gross margin. Revenue alone tells you very little when your largest cost sits in the first year of each customer.

Plans updated. How many existing customers changed something. This is the closest thing to an engagement metric that means anything here, because a plan that is never updated slowly stops matching the person's life.

Beneficiary reachability. Of live plans, the share where every named beneficiary has current contact details that have been confirmed. This is the metric most estate products do not track and should. An unreachable beneficiary is a plan that will fail at the exact moment it is needed.

Partner load. Matters per active partner. Uneven distribution means some partners are drowning and others are going cold.

Security and compliance items. Open findings, overdue access reviews, incidents. Small numbers that get ugly fast when ignored.

Quarterly: the direction numbers

Quarterly is for the slow signals that would be noise week to week.

Retention by cohort. Take the customers who joined in a given quarter and track how many are still active four, eight, and twelve quarters later. This is the single most honest number about your business.

Lifetime value against acquisition cost. Updated with real retention data rather than an assumption made a year ago.

Payback period. How many months to earn back what you spent acquiring a customer. This governs how fast you can grow safely.

Partner retention. How many partners onboarded a year ago are still taking cases. Partner churn is slow and quiet and expensive.

Support cost per customer. Total support cost divided by live plans. It should fall as you grow. If it rises, your product is getting harder to use.

Time to first payout for new partners. From signature to first money received. A long lag here predicts partner churn better than almost anything else.

Recovery drill result. Once a quarter, restore from backup and confirm a plan can be produced end to end. Record whether it worked and how long it took. For a product that must work decades from now, this is a business metric, not an engineering one.

The metrics that mislead

Some numbers are worse than useless because they look meaningful.

Daily and weekly active users. In estate planning, low usage after completion is success. Chasing this number pushes teams to build engagement features that annoy people who are done.

Time in product. More time usually means more confusion. Falling time to completion is the good direction.

Signups. The easiest number to move and the least connected to value. A signup that never completes has cost you money and delivered nothing.

Total documents stored. Impressive on a slide. Says nothing about whether the documents are current, complete, or usable.

Net promoter score. People rating a product built around their own death produce noisy scores that reflect their mood about the topic more than your product. Ask specific questions instead: was anything confusing, did anything take longer than expected.

Average resolution time on support, when bereavement cases are included. Those cases correctly take weeks. Track them separately or the average will push the team to close cases that should stay open.

A dashboard that fits on one screen

Structure beats completeness. Four blocks.

Block 1: This week. New plans, completed plans, completion rate, top three drop off steps, partner turnaround, support ticket mix, failed verifications.

Block 2: This month. Completed plans, live plans, activation rate, acquisition cost by channel, renewal rate, revenue, gross margin, plans updated, beneficiary reachability.

Block 3: Trust and safety. Open security findings, overdue access reviews, incidents this quarter, days since last recovery drill, cases with no update for over seven days.

Block 4: This quarter. Cohort retention curve, lifetime value against acquisition cost, payback period, partner retention, support cost per customer.

Block 3 is the one people leave out and it is the one that belongs on the same screen as revenue. In this category, a trust failure is not a support problem. It is the end of the company.

How to run the review

The numbers are the easy part. The habit is what fails.

Same day, same time, every week. Thirty minutes. If it moves, it dies.

One owner per metric. Not one owner for the dashboard. Each number has a name next to it, and that person explains movement.

Compare to last week and to four weeks ago. A single week is noise. Two comparison points are enough to see a trend without building a forecasting model.

Write one line per metric that moved. Not an essay. What moved, why you think it moved, what you are doing. Three months of those lines is the most valuable operating document you will have.

Look at what stayed flat too. Metrics that never move are usually not being worked on, or are not really being measured.

Change the dashboard rarely. Every change resets your ability to compare. Add a metric only when you know what decision it will drive, and remove one at the same time.

The number to protect

If you can only defend one metric from the pressure to optimise everything else, defend this one: the share of live plans that are complete, current, and usable.

Complete means nothing critical is missing. Current means it has been reviewed or confirmed within a reasonable window. Usable means the documents open, the beneficiaries are reachable, and the release process has been tested.

That number is the actual product. Every other metric on the board describes how you get there or how you pay for it.

Growth numbers tell you the company is getting bigger. That one tells you it deserves to.

Frequently Asked Questions

What KPIs matter most for an estate planning startup?

Completed plans, completion rate, activation rate, renewal rate, cohort retention, acquisition cost against lifetime value, and the share of live plans that are complete, current, and usable.

Why are daily active users a bad metric here?

Because customers are supposed to finish and stop using the product. Frequent logins usually signal confusion, not engagement. Chasing this metric leads to features that annoy customers who are already done.

How often should an estate tech startup review metrics?

Weekly for operating numbers, monthly for business health, quarterly for retention and unit economics. Weekly reviews should take about thirty minutes.

What is beneficiary reachability and why track it?

It is the share of live plans where every named beneficiary has confirmed, current contact details. An unreachable beneficiary means the plan fails at the exact moment it is needed.

Should security metrics be on the main dashboard?

Yes. Open findings, overdue access reviews, incidents, and time since the last recovery drill belong next to revenue. In this category a trust failure is an existential event, not an operational one.

What is the single best measure of an estate tech business?

Cohort retention. It reflects whether customers trust you enough to keep the plan with you, which is what the whole product depends on.

Sources

Further reading: