The seed extension is now the default, not the failure

Somewhere around month fourteen after your seed closes, you will do the arithmetic and realise the numbers do not reach.

You raised eighteen months of runway. You spent four of them hiring. The metrics are real but they are not the metrics a Series A investor wants to see yet, and the six months it would take to get there is five months more than you have.

The old script said this was a failure state. You had missed. You would go to your existing investors with your hat in your hand and take whatever terms were on offer.

That script is out of date. In 2026 roughly 38% of seed-funded startups raise an extension before they ever see a priced Series A term sheet, the median extension runs $1.5M to $3M, and the gap between seed and Series A has stretched to 18 to 24 months. Only 15.4% of the 2022 seed cohort reached a Series A within two years, down from 30.6% for the 2018 cohort. That is the lowest graduation rate on record.

When more than a third of your peers do something, it is not a failure state. It is a stage. And stages can be planned for.

Two things moved at once, and they moved in opposite directions.

Seed rounds got bigger. Series A bars got higher. The Series A that used to want $1M of annual recurring revenue with decent growth now frequently wants $2M to $3M with efficiency metrics attached. Meanwhile the seed round that funds you to that point did not double in size.

So you have the same eighteen months of money and roughly twice the distance to cover. The extension is arithmetic, not judgment.

There is a second factor people talk about less. Series A investors have become much more comfortable waiting. They would rather see four more quarters of data than price a round on a trend line. Waiting is cheap for them. It is expensive for you. That asymmetry is the whole reason extensions exist.

The typical structure is boring, which is the point.

Most are raised on a SAFE rather than a priced round, at flat valuation or a modest 10% to 15% step-up. Dilution usually lands somewhere between 5% and 15%. No new board seat. Existing investors lead, sometimes exclusively, sometimes with one new name brought in to signal outside validation.

The size clusters at $1.5M to $3M because that is what buys nine to twelve months at a seed-stage burn, and because anything larger starts to look like a Series A that failed to get priced, which raises the question you are trying not to raise.

Terms get worse in predictable conditions: when you ask with under three months of cash, when your existing investors are not participating, and when you cannot name the specific thing the money buys. All three of those are avoidable and all three are about timing, not performance.

Every good version of this advice reduces to the same test, and I have not found a better framing than this one.

Are you bridging to a milestone that changes the price, or are you bridging to more time?

If you can finish this sentence in one breath, take the bridge: "In 120 days we will have X, and X makes the Series A happen at a better valuation or with better odds." X has to be specific and externally legible. Two named enterprise logos live in production. Net revenue retention above 110% on a cohort with twelve months of history. A second product line at $40K in monthly revenue.

If X is "more traction" or "a better story," you are bridging to time, and time by itself does not change your price. It just moves the same conversation to a moment when you have less leverage.

The founders who get clean extension terms are not the ones with the best metrics. They are the ones who walked in with a dated, falsifiable claim about what the money produces.

This is the practical part, and it is unglamorous.

At month nine of an eighteen-month runway, sit down and build two versions of the next twelve months. The first assumes you raise a Series A on your current trajectory. The second assumes you do not and need an extension at month fifteen.

Write down, for the second version: how much you would need, what milestone it buys, which existing investors have follow-on reserves, and what you would cut if you raised 60% of the target instead of 100%. That last one is the exercise most founders skip and the one that matters most, because a partial raise is the most common outcome and the worst one to improvise.

Keep both versions somewhere you can update monthly rather than rebuilding from scratch. A spreadsheet works. So does Notion, or a planning tool like Foundra that keeps your financial model and your strategy documents in the same workspace so the runway number updates when the hiring plan does. The tool matters far less than the habit of opening it every month.

By month twelve you will know which version you are living in, and you will have known it for three months rather than three weeks.

Do not open with the word bridge. It carries a decade of connotation and half the room hears distress before you finish the sentence. Extension is the better word, and it is also the more accurate one when you are extending a plan rather than rescuing it.

Ask early. Month twelve, not month sixteen. The ask at month twelve is a strategy conversation. The same ask at month sixteen is a rescue conversation, and rescue conversations price differently.

Be direct about the reserve question. Most seed funds hold follow-on capital for a portion of their portfolio. Asking your lead whether you are in that portion is not rude, and the answer, whichever way it goes, is the single most useful input for your next two quarters.

And if the answer is no, hear it as information rather than a verdict. A fund can pass for portfolio construction reasons that have nothing to do with your company. But a no from your lead does change what an outside investor will assume, so you need to know it months before you need the money.

Sometimes the extension is the expensive way to avoid a decision you already know you need to make.

If your burn assumes a team you built for a growth rate you are not hitting, the extension funds that mismatch for another nine months and then you face the same choice with more dilution behind you. Cutting to default alive is unpleasant and it preserves the thing an extension spends: your optionality.

If the business works but will not be venture scale, an extension delays a conversation about a different path, whether that is profitability, a small acquisition, or simply running it well. None of those are failures. All of them get harder to reach the more preference stack sits on top.

The extension is the right tool when the trajectory is real and the timing is short. It is the wrong tool when it is buying you time to not decide.

• About 38% of seed startups now raise an extension before a priced Series A. It is a stage, not a verdict.
• Median extension: $1.5M to $3M, on a SAFE, flat or a 10% to 15% step-up, 5% to 15% dilution, no new board seat.
• Only 15.4% of the 2022 seed cohort reached Series A within two years, versus 30.6% for the 2018 cohort.
• Take the bridge if it reaches a milestone that changes the price. Think much harder if it only buys time.
• Build the extension plan at month nine and update it monthly, including the version where you raise 60% of target.
• Ask your lead about follow-on reserves at month twelve. The answer shapes everything downstream.

Does raising an extension hurt my Series A? Far less than it used to, because so many companies now have one. What hurts is a pattern of repeated small raises with no milestone attached to any of them, or an extension raised at three weeks of runway on visibly distressed terms.

Priced round or SAFE for an extension? SAFE, in most cases. It is faster, cheaper, and it avoids setting a price at the exact moment your leverage is weakest. Price it at the Series A when the data is better.

What if my existing investors will not participate? Find out early, then treat it as a signalling problem to solve rather than a secret to keep. New investors will ask. Having a clear, unembarrassed answer, such as a fund that stopped writing follow-on checks entirely, matters more than the fact itself.

How much runway should an extension buy? Nine to twelve months. Under nine and you are fundraising again immediately. Over twelve and the round size starts inviting Series A scrutiny without Series A terms.

Is a flat extension bad for my team's options? Flat is fine. Employees are diluted either way. What damages morale is the surprise, not the number, so tell your team the extension is a plan before it becomes a scramble.

What if we hit the milestone early? Then you raise the Series A early and the extension capital becomes cushion. That is the good outcome, and it is a decent argument for sizing the extension at the lower end of the range.

1. Seed Extension Round 2026: Median Size, Dilution, and Why 38% of Seed Startups Now Raise One, Value Add VC
2. Bridge rounds in 2026: when to take one, when to die instead, VC Boom
3. Bridge Rounds Are the New Series A: Inside the Widening Gap Between Seed and Growth, SeedScope
4. Seed-to-Series-A graduation rate benchmarks in 2026, Causo Hub
5. Don’t Say “Bridge” Round, Ash Rust
6. Extension Rounds: How to Structure One, 2026 guide