Time to value: how to measure and shrink it
How to define your value moment, measure time to value with the analytics you already have, and cut the days that predict churn.
How to define your value moment, measure time to value with the analytics you already have, and cut the days that predict churn.
Time to value (TTV) in SaaS is the number of days from signup to the moment a customer first gets the outcome they bought your product for. Measure it per cohort using the analytics you already have, look at the full distribution rather than the average, then attack the single longest step in the path. The slow half of your TTV distribution is where your early churn lives.
Time to value is the elapsed time between a customer signing up and reaching their first value moment: the first time your product actually does the job they hired it for.
Not the first login. Not completing your onboarding checklist. The first real outcome. A customer can log in five times, finish your product tour, and still have received zero value.
TTV matters because customers make their keep-or-cancel judgment early, while the memory of evaluating alternatives is fresh. Every day between signup and first value is a day the customer is paying (or burning trial time) on faith. The longer that gap, the more accounts give up before the payoff arrives.
The value moment must be a specific, observable event in your product. If you cannot write a query for it, you have not defined it. Three examples from different SaaS types:
| SaaS type | Weak definition | Concrete value moment | | --- | --- | --- | | Email marketing tool | "User is onboarded" | First campaign sent to 50+ real subscribers | | Analytics platform | "Account is set up" | First dashboard viewed with live production data flowing in | | Support desk software | "Team is active" | First customer ticket resolved through the tool |
Notice the pattern: each concrete definition involves the customer's own data or the customer's own customers, not sample data or setup steps. Sending a test email to yourself is activity. Sending a campaign to your real list is value.
If you are unsure what your value moment is, pull 20 accounts that renewed and 20 that churned in their first 90 days, and look for the earliest event the renewers did that the churners did not. That event, or the one just after it, is usually your value moment.
Do this this week: write your value moment as one sentence containing an event name and a threshold, e.g. "first campaign sent to 50+ subscribers." If any word in the sentence is not queryable in your database or analytics tool, rewrite it until it is.
You do not need an activation platform. You need two timestamps per account: signup date and first-value-moment date. Both already exist somewhere in your stack.
Group by monthly signup cohort, not calendar month of activation. A cohort view answers the question that matters: "of the accounts that signed up in April, how fast did they reach value?" It also shows whether your onboarding changes are working, because each cohort experienced one version of your onboarding.
Track two numbers per cohort: median TTV and the share of accounts that never reach the value moment at all. The never-activated share is often the bigger problem, and an average silently hides it.
Average TTV is close to useless, because TTV distributions are long-tailed. A cohort with a 6-day average might be half the accounts activating in 2 days and the other half taking three weeks or never activating. The average says "fine." The distribution says "half your cohort is in trouble."
Split every cohort at the median. The fast half is your happy path: they found value quickly and will mostly retain. The slow half is where early churn concentrates: they are paying, waiting, and losing faith. Your interventions should target the slow half specifically, because the fast half did not need help.
Then find the stall point. Break the path from signup to value moment into its steps (account created, data connected, first item created, first real outcome) and measure the drop-off and time spent at each step. One step will dominate. That step is your target.
Once you know the longest step, you have four ways to attack it, roughly in order of preference:
The order matters. Removing a step beats automating it, which beats templating it, which beats manually rescuing people from it. But most teams should run levers 2 and 4 immediately while engineering works on levers 1 and 3, because human assist ships this week.
TTV connects to churn through a simple mechanism: customers cancel when the cost of waiting exceeds their belief in the payoff. Long TTV drains belief daily. This is why so much churn lands in the first 30 to 90 days: it is mostly customers who never reached the value moment at all, not customers who got value and later left.
Sales sets the clock. The moment your sales page, demo, or rep promises an outcome, the customer starts a mental timer against that promise. If the demo showed a working dashboard in 10 minutes but real setup takes three weeks, the gap between promised TTV and actual TTV becomes the customer's first experience of your company. Two fixes:
TTV shrinks through routine attention, not a one-time project. A 30-minute weekly ritual is enough:
Accounts that stall past your median TTV and receive no intervention are your most predictable churn. The review exists so that never happens silently.
Early churn is an activation failure, not a product failure. Find where signups stall and fix the cheapest leak first.