The founders who run out of runway almost never have a data problem. They have a clarity problem — too many numbers, not enough of the right ones, and no clear line from the dashboard to the decision that actually needed to be made.
According to CB Insights analysis of startup failures, running out of cash is among the top causes of early-stage business failure. But cash doesn't run out because founders weren't watching the bank balance. It runs out because they were watching the wrong indicators, or because the ones they were watching weren't telling them the truth about what was actually happening.
There's an industry of financial planning advice aimed at founders that produces elaborate models, tracks thirty metrics, and is updated quarterly with great discipline. And yet the founders sitting across those dashboards still find themselves surprised when the business is six weeks from difficulty. The models didn't fail. The metrics did.
Early-stage FP&A is not about tracking everything that could be tracked. It's about identifying the small number of numbers that would change what you do this month, and making sure you understand them deeply enough to act on them.
Build the model around the decisions, not the other way round
The most useful reframe for early-stage financial planning is to start from the decisions you actually need to make, and work backwards to the data that would inform them. This sounds straightforward. It runs counter to how almost every financial model is built.
Models are typically built to capture everything — on the grounds that more information is better, that completeness is a proxy for rigour, and that an investor might ask about anything. The result is a model that takes three hours to update and six minutes to explain, and that doesn't change any decisions because nobody actually interrogates it between board meetings.
The decisions an early-stage founder genuinely needs to make are limited. How long do we have at our current burn rate? Can we afford to hire before the next raise? Are we pricing at a level that reflects the value we deliver? Are the unit economics moving in the right direction over time? Those questions require a small number of metrics understood deeply — not a large number tracked superficially.
If the financial model can't answer "should we make this hire?" or "do we double down on this channel or cut it?" in thirty seconds, it's answering the wrong questions.
The test worth running is simple: look at your current tracking and ask, for each metric, what decision would be different if this number were 20% worse? If the answer is "none immediately," the metric probably isn't load-bearing. Cut it, or move it to a monthly review rather than weekly attention.
Runway stated precisely — not "twelve months"
Runway is the most important number in any early-stage business. It's also the number most frequently stated imprecisely — and imprecision here is specifically dangerous.
"Twelve months of runway" typically means: current cash balance divided by average monthly burn over the last three months. That's a starting point, not an answer.
Real runway accounts for the committed outflows that haven't yet hit the P&L — the annual software contracts that renew in February, the lease event in Q3, the payroll change when the new hire starts in April. It accounts for the step-changes in cost that are coming as the team grows, not just the cost base that exists today. And critically, it accounts for the lead time required to close a funding round or generate significant new revenue — because in the UK market, a fundraising process typically takes six to nine months from first conversation to cash in the bank.
A founder who believes they have twelve months of runway may have seven when those factors are applied properly. That's not a mathematical error — it's a planning error, and it's a common one. The correction is to state runway with the assumptions explicit: "We have twelve months at current burn, which adjusts to eight months once the April headcount change and the Q3 lease event are factored in. That means we need to be in active fundraising conversations by the end of this quarter."
That's a sentence you can act on. "We have twelve months of runway" isn't.
Gross margin is the number that tells you whether the business model works
Gross margin is the single metric that most early-stage founders underweight, and the one that experienced investors probe most carefully. The reason is straightforward: gross margin is the best early signal of whether the business model is actually viable at scale.
A business growing revenue at 40% year-on-year is impressive. A business growing revenue at 40% year-on-year on a gross margin of 20% is building a bigger version of an economics problem. The costs required to deliver the revenue are scaling almost as fast as the revenue itself, which means the business needs to grow very large before it generates meaningful cash — and that trajectory requires capital at a scale that changes the fundraising conversation fundamentally.
Gross margin benchmarks vary significantly by sector and model. SaaS businesses typically operate at 70–80% gross margin at maturity. Professional services businesses at 40–60%. Marketplace models and hardware-adjacent businesses can be structurally lower. What matters is knowing where you are relative to the model you're building, and tracking whether the trend is moving in the right direction as you scale.
Investors who know what they're doing assess gross margin trajectory before almost anything else. It tells them more about the business model's viability than the revenue figure on its own ever could.
Getting clarity on gross margin early — and understanding what's driving it, which cost lines are genuinely variable versus disguised fixed costs — is one of the most valuable exercises a founder can do before sitting down with investors. It is the lens through which every subsequent financial conversation will be filtered.
Burn decomposed — not as a total, but as a structure you can act on
Headline burn is almost useless as a management tool. Knowing that the business spent £180,000 last month tells you almost nothing about what you can do about it.
Burn decomposed into its three components gives you something to work with. Fixed costs — salaries, rent, committed contracts — that don't change regardless of revenue and that require structural decisions to alter. Variable costs — delivery costs, commissions, usage-based software — that scale with activity and can be managed through volume decisions. And discretionary costs — marketing spend, conferences, tooling that isn't business-critical — that can be cut quickly if circumstances change.
That third category is where optionality lives. A founder who knows their discretionary cost base is £22,000 per month knows exactly what headroom they have if the business needs to extend runway without a structural change. A founder who only knows their total monthly burn has no idea.
The discipline of decomposing burn monthly — not just tracking the total — changes how a founder thinks about the decisions in front of them. It turns the abstract question of "how do we extend runway?" into a concrete set of choices with known costs and timelines attached to each one.
What you can stop tracking without losing anything important
A 36-month revenue forecast with monthly granularity, built in the first year of trading, is not financial planning. It's financial theatre. The assumptions that underpin it will be wrong — often substantially wrong — and the time spent building and maintaining it is time not spent on the things that actually move the business forward.
The same applies to detailed cohort analyses before there are enough cohorts to be meaningful, to complex multi-touch attribution before the marketing model has stabilised, and to formal monthly variance analysis against a budget that was itself largely speculative when it was written.
These aren't bad tools. They're premature ones. They belong in a more mature business where the model is understood well enough that small variances are meaningful signals rather than noise. At early stage, the signal is almost always the direction of a small number of key metrics, not the precision of variance against a forward projection that didn't have robust inputs to begin with.
Three things tracked and understood deeply will tell you more about the health of an early-stage business than thirty things tracked superficially. The discipline is not in the complexity of the model — it's in the honesty with which the few things that matter are interrogated and acted on.
At Eranos, we work with early-stage founders on exactly this kind of financial clarity — helping them build the reporting and forecasting structure that's actually useful for running the business, not just presentable in a board pack. If your financial model isn't changing the decisions you make, it's worth a conversation about what a more useful one looks like.
Published by Eranos ·