You Can't Manage What You Don't Measure
This saying is often attributed to Peter Drucker, the father of modern management, and like a lot of sayings, it's absolutely true. If you're running a startup and you can't pull up your key metrics in under sixty seconds, you're flying blind. You might feel like you know how the business is doing. You don't. You're guessing.
The companies I work with that execute best have one thing in common: they measure relentlessly. Not vanity metrics. Not whatever their investor asked about last. The metrics that actually tell you whether the business is healthy, whether it's improving, and where it's about to break. Getting this right — choosing what to measure, building the infrastructure to capture it, forecasting it, and displaying it in near real-time — is one of the highest-leverage things a startup can do. And almost nobody does it well at the seed stage, which is exactly when it matters most.
Universal Metrics: What Every Startup Must Track
Regardless of whether you're building software, hardware, a marketplace, or a biotech product, there is a set of metrics that every founder, every board member, and every investor cares about. These are non-negotiable.
Start with cash. Your current cash balance, your gross burn rate, your net burn rate, and your cash-zero date. If you don't know exactly when you run out of money, nothing else matters. "We have about eighteen months" is not an answer. "At current burn, we hit zero on November 14th" is an answer.
Next: gross margin and contribution margin. These tell you whether your unit economics work — whether every dollar of revenue actually contributes to covering your fixed costs, or whether you're losing money on every sale and hoping to make it up on volume. Then headcount and revenue per employee, which together tell you whether you're scaling efficiently or just adding bodies.
These metrics are table stakes. They apply to every startup, at every stage, in every industry. If you're not tracking all of them today, that's your first action item.
Metrics by Startup Type
Beyond the universals, the metrics that matter depend heavily on what kind of company you're building. Here's where the specificity matters.
SaaS Startups live and die by recurring revenue metrics. MRR, ARR, and MRR growth rate tell you the trajectory. Churn rate and net revenue retention tell you whether customers are sticking around and expanding. LTV and CAC tell you whether your customer economics work, and the LTV:CAC ratio — healthy is 3:1 or better — tells you whether growth is sustainable. CAC payback period tells you how long it takes to recoup acquisition cost. At the growth stage, the Magic Number and Rule of 40 become your north stars for efficiency.
Hardware Startups have a fundamentally different set of operational metrics layered on top of the financials. First pass yield tells you what percentage of units come off the line without rework — target 95% or higher, because every point below that erodes your margin fast. BOM cost and its trend over time tells you whether your cost structure is improving as you scale. Inventory turns tell you how efficiently you're managing working capital. Demand forecast accuracy is critical — get it wrong and you either can't fulfill orders or you have cash tied up in unsold inventory. Warranty and return rates tell you about product reliability in the field. And DPMO — defects per million opportunities — gives you a quality benchmark that manufacturing partners understand.
Marketplace and Platform Startups need to track both sides of the network. GMV — gross merchandise value — tells you the total transaction volume flowing through the platform. Take rate tells you what percentage you keep. Liquidity — the percentage of listings that actually transact — tells you whether your marketplace is working. Search-to-fill rate, supply/demand balance, and buyer and seller CAC round out the picture. A marketplace with high GMV but low liquidity is a vanity story.
D2C and E-commerce Startups need to obsess over customer acquisition and retention. CAC, LTV, and the ratio between them are foundational. Repeat purchase rate is your leading indicator of product-market fit — anything above 25% is solid, above 40% within twelve months is excellent. Average order value, conversion rate, return on ad spend, and cart abandonment rate are your operational levers.
Fintech Startups layer risk metrics on top of the standard financials. Fraud rate, payment success rate, loan default rate, and chargeback rate all measure the health of your core operations. Operating expense ratio — operating costs divided by revenue — should compress from around 50% in year one to 30% by year three. And regulatory compliance metrics are essential, because in fintech a compliance failure can shut you down.
Biotech and Life Sciences Startups are unique because you may be years from revenue. Your primary metrics are R&D burn rate, cash runway, and regulatory milestone progress. Clinical trial metrics — patient enrollment rate, trial completion rate, adverse event occurrence — track the execution of your core value-creation process. Patent filings and citations measure the strength of your IP moat. In biotech, every metric is ultimately in service of the question: are we de-risking the science fast enough before we run out of cash?
Your company may also have metrics peculiar to your specific business — ones that don't fit neatly into any standard category. That's fine. What matters is that you've thought carefully about what actually drives your business and you're measuring it.
Leading vs. Lagging Indicators
Here's a distinction most founders don't think about carefully enough. Lagging indicators tell you what already happened: revenue, churn, retained customers, profitability. They're the rear-view mirror. By the time a lagging indicator moves, the damage is done or the win is already locked in.
Leading indicators tell you what's about to happen: signups, activation rates, demos booked, feature adoption, trial conversions, onboarding completion. They're the windshield. They give you time to react.
Early-stage companies should weight leading indicators heavily, because you need to know what's coming — not just what already arrived. If your demo-to-close rate drops from 30% to 15%, you need to know that now, not three months from now when revenue falls off a cliff. If your activation rate after signup is declining, that's a product problem showing up before it becomes a churn problem.
The disciplined approach is to track both, understand the causal chain between them, and use leading indicators to drive action while using lagging indicators to confirm results.
Your Metrics Evolve With Your Stage
At the earliest stage, when you're taking technology risk off the table, your metrics might be a yield percentage or a performance benchmark. Can the thing actually work? Does the reaction produce the expected output? Does the prototype hit the target specification? These aren't business metrics yet — they're science and engineering metrics. But they're the ones that matter, and they're the ones your pre-seed investors want to see trending in the right direction.
Once you begin developing the product, your metrics shift to efficiency and quality. Bugs per line of code. Time to resolve critical issues. Test coverage. Build success rate. For hardware, it's prototype iteration speed, component qualification rates, and DFM compliance scores. You're proving that the team can execute, not just invent.
As you acquire your first customers, the business metrics kick in: CAC, conversion rate, early retention signals, and of course burn. By the time you're raising your Series A, investors expect to see real unit economics — LTV, CAC payback, gross margin trajectory — alongside evidence that the metrics are improving over time.
At every stage, you're taking a category of risk off the table, and your metrics need to prove it. Technology risk, then product risk, then product-market fit risk, then execution risk, then scale-up risk. The metrics that matter shift at each stage, but the discipline of measuring them should be constant from day one.
Time-Series Is Where the Insight Lives
A snapshot of your CAC is interesting. CAC trending down over six months while conversion rate trends up is compelling. That's a story investors want to hear, and it's a story you can only tell if you've been measuring consistently over time.
This is why I push every company I work with to think in time-series. Not "what is our churn rate?" but "how has our churn rate moved over the last eight quarters, and where do we forecast it going?" Not "what is our gross margin?" but "how has gross margin improved since we switched contract manufacturers, and what does the model say it will be at twice our current volume?"
Your Financial Model needs to forecast every one of your KPIs and metrics. And you need to be in the habit of producing what most people call a BVA — Budget vs. Actual — although I think the better framing is AVF: Actual vs. Forecast. The forecast came first. Actuals follow. The variance between them is where you learn the most about your business. A positive variance means something is working better than you expected — understand why. A negative variance means something is underperforming — debug it before it compounds.
Companies that produce AVFs monthly and discuss them rigorously have a massive advantage. They catch problems early. They understand their own dynamics. And when they sit in front of investors, they can explain not just where the numbers are, but why they moved and what they're doing about it.
Collect the Underlying Data — Now
Many KPIs are calculated from raw data that you need to be capturing from the very beginning. LTV requires transaction history and retention data over time. CAC requires granular marketing and sales spend data tied to customer acquisition dates. Net revenue retention requires expansion, contraction, and churn data by cohort. If you're not collecting the underlying data today, you won't be able to compute these metrics later — and you certainly won't have the time-series history that makes them meaningful.
Here's a concrete example that illustrates why this matters, especially for hardware startups. You want to understand what's happening with your sales pipeline. Most CRM tools don't retain a history of the pipeline — they show you the current state. Yesterday's pipeline is gone. The solution is to take a snapshot of your pipeline on a daily basis and store it in a data warehouse, so you can analyze the evolution of the pipeline over time. How long do deals stay in each stage? What's the conversion rate from stage to stage? How has the pipeline's total value trended over the last six months?
This is especially critical in a hardware startup, because of utmost importance is minimizing working capital. Among the ways to minimize working capital is an accurate demand forecast. That forecast will be informed by your sales pipeline and your understanding of pipeline dynamics — the velocity, the conversion rates, the seasonality. Because if you can't forecast demand properly when you're a manufacturing company, you will commit one of two sins: either you'll be unable to fulfill in a timely manner because you manufactured too few units, or you'll have cash tied up in inventory sitting on the shelf because you manufactured too many. The demand forecast is one of the most important exercises in a hardware startup, and you probably need a data warehouse to master your pipeline dynamics.
Dashboards: Near Real-Time or You're Flying Blind
Once you're measuring the right things and collecting the underlying data, you need to display it — not in a monthly report that someone spends two days assembling, but in a dashboard that updates in near real-time. The whole point of measuring is to notice changes as they happen, not to discover at the end of the quarter that something went sideways two months ago.
This means automation. Nobody should be spending time copying and pasting data into spreadsheets. It's error-prone, it's time-consuming, and it guarantees your data is stale by the time anyone sees it. Data needs to flow automatically from your single sources of truth — your CRM, your accounting system, your manufacturing execution system, your support ticketing tool — into your FP&A and BI platform.
Invest in a decent tool. For FP&A, platforms like Pigment give you multi-dimensional financial modeling with real-time data updates and AI-assisted scenario planning. Runway is excellent for early-stage startups that need lightweight burn and runway tracking. Mosaic is purpose-built for SaaS metrics. For dashboards and BI, Metabase is a strong open-source option that's easy to set up, Tableau handles complex enterprise visualization, and Power BI integrates well if you're already in the Microsoft ecosystem. The specific tool matters less than the commitment to using one. Otherwise someone on your team is doomed to spend hours, days, or weeks in spreadsheets — time that produces no value beyond what a properly configured tool would give you automatically.
Alignment: Model, Metrics, and Board Reporting
Here's where it all comes together. Your Financial Model forecasts your KPIs. Your dashboards show the actuals. Your AVF analysis explains the variance. And your board slides should display the same data you're already looking at every day.
I've written elsewhere about board reporting, and this is the key connection: your board slides should be the same structure every meeting. The data in those slides shouldn't be materially different from the data you're visualizing daily, weekly, and monthly to manage the company. If board prep is a fire drill, it's because your measurement infrastructure isn't working. When it is working, board prep becomes straightforward — you're just packaging data you already have into a format your board expects.
The alignment chain is: Financial Model forecasts the metric → your systems capture the actual value → your dashboard displays both → your AVF explains the delta → your board slide presents the story. When this chain is unbroken, you have a company that understands itself. When any link is missing, you have guesswork.
This Is How You De-Risk Execution
Investors are always evaluating execution risk. It's the risk that keeps them up at night, because unlike technology risk or market risk, execution risk is entirely about you and your team. Can you actually run this company?
You make them comfortable when you know your important metrics and KPIs. When you dashboard them. When you forecast them. When your AVF is tight. And especially when you can explain the variance and show that you're already acting on it. That's the signal investors are looking for. Not that everything is perfect — nothing is perfect at the seed or Series A stage. But that you're paying attention, that you understand your own business, and that you have the systems in place to catch problems before they become crises.
Building this measurement infrastructure — choosing the right metrics, setting up the data collection, configuring the dashboards, integrating the FP&A tool, establishing the AVF discipline — is exactly the kind of work a fractional COO does. It's operational architecture. It requires experience to know which metrics matter at your stage, judgment to avoid drowning in data, and discipline to maintain it month after month. The founder's job is to build the product and close customers. My job is to make sure the company knows how it's performing while they do it.