Template18 pages20 min read

Startup KPI Framework

The 15 metrics that matter at each stage — and the operating rhythms that turn data into decisions.

Most startup dashboards are graveyards of good intentions. Forty metrics nobody acts on. Vanity numbers that make board decks look impressive but don't drive a single weekly decision. This framework is the antidote. It's the exact metric architecture we install with every client — a three-tier system that connects daily operational signals to weekly decisions to monthly outcomes. No more than 15 numbers at any given time, each with a clear owner, a defined review cadence, and an explicit action protocol when the number moves outside its expected range. We've refined this across 60+ startups from pre-seed to Series C. The structure is universal; the specific metrics are calibrated to your stage, business model, and current constraints.

Who This Is For

Founders drowning in dashboards but starving for actionable insight
Heads of Operations or Chiefs of Staff building their first reporting infrastructure
Series A/B companies preparing for board-level metric rigour
Data and analytics leads who need a framework before they build the pipeline

01The Problem with How Startups Track Metrics

There are three failure modes we see in almost every startup that comes to us: 1. The Metric Graveyard — 30+ KPIs tracked in a sprawling dashboard that nobody opens except before board meetings. Data is collected but never reviewed in a cadence that allows action. By the time you notice a number is off, the underlying problem has been compounding for weeks. 2. The Vanity Trap — Metrics that feel good (total users, page views, gross revenue) but don't expose the levers you can actually pull. A founder who celebrates '10,000 users' without knowing that only 400 are active and only 50 are paying is flying blind with a smile. 3. The Metric-Action Gap — Even when the right numbers are tracked, there's no protocol for what happens when they change. A 15% drop in activation rate is noticed in a Monday meeting, discussed for 10 minutes, and then forgotten until the next Monday when it's dropped another 8%. The KPI Framework solves all three by structuring metrics into tiers with explicit review cadences and action protocols.

02Tier 1: Health Metrics (Monitored Daily)

Health Metrics are the vital signs of your business. They tell you whether the engine is running — not whether it's winning the race. You check them the way a pilot checks instruments: quickly, frequently, and with clear red-line thresholds that trigger immediate action. You should have 4–5 Health Metrics, monitored via automated alerts (Slack notifications, email digests, dashboard widgets). The goal is zero manual effort to monitor — your systems should tell you when something's wrong.

Cash Position & Burn Rate — Your runway in weeks, updated daily. Not monthly. Not quarterly. Daily. Include committed expenses and expected receivables. Set a 'concern threshold' (typically <6 months runway) and a 'critical threshold' (<3 months) with pre-defined responses for each.
Daily Active Users / Daily Active Revenue — Depending on your model. For SaaS, DAU or daily sessions. For transactional businesses, daily GMV or transactions. The point is a daily pulse that shows whether the core business is functioning normally.
Pipeline Coverage Ratio — Total qualified pipeline ÷ quarterly target. This should be 3× or higher for healthy businesses. Below 2.5× is a leading indicator of a miss that's 60–90 days away — early enough to act, late enough to be credible.
System Uptime / Error Rate — For product-led businesses, this is existential. Track error rates, API response times, and critical path availability. Set pager-level alerts for anything that affects user experience.
Customer Support Queue Depth — A proxy for product quality and customer satisfaction. A growing queue is often the first signal of a product issue, a scaling problem, or a support process breakdown.

KEY DELIVERABLERule: If a Health Metric crosses its red-line threshold, the response happens within 4 hours — not the next standup. Define the response protocol in advance: who is notified, who owns triage, what authority they have to act.

03Tier 2: Lever Metrics (Reviewed Weekly)

Lever Metrics are the numbers you can actually move with deliberate action. They represent the mechanisms of your business — the conversion rates, velocity metrics, and efficiency ratios that respond to operational decisions. You should have 5–6 Lever Metrics, reviewed in your weekly operating review. Each metric has a named owner who is responsible for understanding its movement and proposing actions when it drifts outside its expected range.

Lead-to-Qualified Conversion Rate — What percentage of inbound leads meet your qualification criteria? This tells you whether your marketing is attracting the right audience and whether your qualification criteria are calibrated correctly. Benchmark: 15–30% for B2B SaaS.
Sales Cycle Length (by segment) — Median days from first touch to closed deal. Track by segment because enterprise and SMB cycles are fundamentally different. A lengthening cycle is one of the earliest warning signs of product-market drift.
Activation Rate — The percentage of new users/customers who reach your 'aha moment' within their first session or first week. This is the single most important metric for product-led growth companies. Even a 5% improvement here compounds dramatically.
Net Revenue Retention (NRR) — Revenue from existing customers this period ÷ revenue from those same customers last period. Includes expansion, contraction, and churn. Above 110% means you grow even if you stop acquiring new customers.
Engineering Velocity (Cycle Time) — Median time from 'in progress' to 'deployed' for engineering tickets. Not story points, not velocity — cycle time. It's the only engineering metric that correlates with business outcomes.
CAC Payback Period — Months to recover the fully-loaded cost of acquiring a customer. Include sales salaries, marketing spend, onboarding costs, and founder time. Most B2B SaaS companies should target <12 months; below 6 is exceptional.

KEY DELIVERABLEProtocol: Every Lever Metric that moves >10% week-over-week gets a 'why note' from its owner — a one-paragraph explanation attached to the metric in your review. No explanation required for normal fluctuation; mandatory for significant moves.

04Tier 3: Outcome Metrics (Assessed Monthly)

Outcome Metrics are the results that matter to your board, your investors, and your long-term strategy. They move slowly, they're influenced by many factors, and they should never be reviewed weekly because they create noise that leads to over-reaction. You should have 4–5 Outcome Metrics, reviewed in your monthly business review with leadership and quarterly with your board.

Monthly Recurring Revenue (MRR) & Growth Rate — The headline number. Track absolute MRR and month-over-month growth rate. Decompose into New MRR, Expansion MRR, Contraction MRR, and Churned MRR. The decomposition is more important than the headline.
Gross Margin — Revenue minus cost of goods sold (hosting, support, implementation) divided by revenue. This tells you whether your business model is viable at scale. Below 60% for SaaS is a structural concern; above 75% is strong.
Rule of 40 Score — Growth rate + profit margin. A combined score above 40 indicates a healthy balance between growth and efficiency. Below 20 signals that you're neither growing fast enough nor efficient enough.
Employee Engagement / eNPS — A lagging indicator that predicts future performance. Declining engagement is a 3–6 month leading indicator of attrition, which is a 6–12 month leading indicator of execution slowdown.

KEY DELIVERABLEMonthly Business Review Format: 30 minutes. First 10 minutes: Outcome Metrics walkthrough with trend lines and commentary. Next 10 minutes: Root-cause analysis of the 2–3 biggest moves. Final 10 minutes: Decision-making on any strategic adjustments.

05Stage-Specific Metric Selection

The 15 metrics you track at Pre-Seed are fundamentally different from the 15 at Series B. Here's how we calibrate: Pre-Seed / Seed (Pre-PMF): Focus almost entirely on activation, engagement, and qualitative signals. The metrics that matter are user interviews completed per week, activation rate, weekly active users, and engagement depth (sessions per user, features used). Revenue metrics are premature — you're validating, not optimising. Seed / Series A (Post-PMF, Pre-Scale): Shift to revenue mechanics. MRR growth, CAC payback, sales cycle length, and NRR become your Lever Metrics. You're proving that the business model works before you pour fuel on it. Series A / B (Scaling): Add efficiency and predictability metrics. Forecast accuracy, pipeline coverage, gross margin, and Rule of 40 join the framework. The board expects rigour; the business needs guardrails against inefficient scaling. Series B+ (Optimisation): Layer in market metrics — market share, competitive win rate, and category leadership indicators. The business is large enough that external positioning matters as much as internal execution.

06Building the Dashboard

The framework includes three dashboard templates: 1. The Daily Pulse — A single-screen view of all Tier 1 Health Metrics with automated red/amber/green indicators. This should be visible on a wall screen or pinned in your primary communication channel. No interaction required — it pushes alerts when thresholds are crossed. 2. The Weekly Operating Review Dashboard — All Tier 2 Lever Metrics with 8-week trend lines, owner names, and 'why note' fields. This is the centrepiece of your weekly operating review. The template includes the agenda structure and facilitation guide. 3. The Monthly Business Review Deck — Tier 3 Outcome Metrics with 6-month trend lines, decomposition charts, and narrative commentary fields. Designed to be exported to PDF for board distribution with minimal formatting.

Key Takeaways

01Never track more than 15 metrics — excess data creates the illusion of insight while preventing actual decision-making
02Three tiers with three cadences: Health (daily/automated), Levers (weekly/owned), Outcomes (monthly/strategic)
03Every metric needs an owner, a threshold, and a pre-defined action protocol — tracking without action is just record-keeping
04Stage-specific calibration is essential: the metrics that matter at Pre-Seed actively mislead at Series B
05The dashboard is not the system — the review cadence and action protocols are where value is created

Want us to build your KPI architecture with you?

Book a free 30-minute diagnostic to see how these frameworks apply to your specific situation.