Five numbers run an early-stage company: revenue growth rate, net revenue retention, customer acquisition cost, lifetime value, and burn multiple. Every other metric is diagnostic — opened when one of the five moves without explanation. The failure mode of early dashboards is the opposite philosophy: twenty charts of everything, none owned, decisions made on vibes anyway. The discipline is one screen, five numbers, each with an owner, a target, and a review cadence — small enough that the team actually argues about it on Mondays.
This is a metrics guide, not financial advice; definitions interact with accounting choices your finance lead should bless.
What are the five, and why these?
Growth rate (month-over-month revenue, new plus expansion) is the trajectory — the number investors and morale track. Net revenue retention (NRR: revenue from a cohort today over that cohort's revenue a year ago, including expansion, downgrade, churn) is the quality — above 100%, existing customers grow you without new acquisition. CAC (fully loaded sales and marketing cost per new customer) is the engine's cost. LTV (gross-profit lifetime value, not revenue lifetime) is the engine's yield. Burn multiple (net burn divided by net new annual recurring revenue) is capital efficiency — how many dollars you burn to buy a dollar of ARR; lower is better, and in tighter markets it's the number boards scrutinize. Together they answer the only strategic questions an early company has: are we growing, is the growth durable, can we afford it, and is it getting cheaper?
How do you compute each without fooling yourself?
The standard self-deceptions are definitional. CAC computed on marketing spend only — forgetting sales salaries, the largest line — understates by half. LTV computed on revenue instead of gross profit (and using 1/churn with a churn rate too low to be stable) produces the famous LTV:CAC ratios of 11:1 that never survive diligence. NRR computed on averages instead of cohorts hides concentration: one whale's expansion can mask a base churning out. Burn multiple computed on committed rather than collected revenue. The honest versions: fully loaded CAC, gross-margin LTV with a churn assumption grounded in at least four quarters of actual cohorts, cohort-based NRR, and burn multiple on net new ARR with a footnote on how ARR is recognized. The definitions your investors will use in diligence are the ones to use in the dashboard from day one.
Related stories: SEO for Early-Stage Startups: The Long Game That Pays Compounding Rent · When to Hire a Growth Team (and What It Needs Before It Can Work).
What targets are healthy at each stage?
| Metric | Pre-seed | Seed | Series A ready |
|---|---|---|---|
| MoM growth | 10–20% on small base | 10–15% sustained | ~3× YoY |
| NRR | Directional | 90–100% | 100%+ |
| LTV:CAC | Don't optimize yet | ≥3:1 on honest math | 3–5:1 with CAC payback under 18 months |
| Burn multiple | N/A (little revenue) | <2 good, <1.5 strong | <1.5 |
Treat the pre-seed column as permission: with 30 customers, cohort statistics are noise, and the only metric that matters is qualitative retention — do the first users come back without prompting?
What diagnostic metrics sit underneath?
- Activation rate — % of sign-ups reaching the value moment; explains flat growth with healthy traffic.
- CAC by channel — explains blended CAC drift; the blend hides the one channel carrying it.
- Churn cohort curves — early-tenure drop-off (onboarding problem) versus late flatline (value problem) demand opposite fixes.
- Sales cycle length — the silent killer of growth rate at constant win rate.
Per Census Bureau business data, firm-level performance varies enormously within industries — the spread is management, not luck, and the diagnostic layer is where management happens.
How does the dashboard get used, practically?
One screen, reviewed weekly at the growth meeting: five numbers, this week's actual, the trend arrow, and the single experiment each metric's owner is running this week to move it. The rule that keeps it alive: a metric that nobody is currently running an experiment against gets questioned — either it earns an experiment or it leaves the screen. Quarterly, the five get audited against the honest definitions above, because definitional drift is gradual and compounding. Growth is a system with five gauges; read them weekly, fix what moves wrong, and stop decorating the wall with charts nobody owns.
