A valuation cap looks like a number. It's actually a relationship: between what the founder thinks the company is worth, what the market will pay, and what the next round will look like. Get the relationship wrong and the cap becomes a tax on the rest of the company's life.
When a cap is too low
A cap that's set well below the round's natural valuation is sometimes a founder being conservative. More often, it's an investor pushing for cheap ownership in a round the founder is too tired to renegotiate.
If your pre-seed cap is at $4M post and you have a working product, real revenue, and a team people would hire today, you're underpricing yourself. The next round will price up, and the SAFE investors will convert at the cap, capturing more of the company than they earned. Your dilution math gets brutal.
Test: model the cap table assuming the next round prices at three to four times the SAFE cap, which is the modal outcome for a successful pre-seed. If founder ownership ends up below 65% after a normal seed, the cap is too low.
When a cap is too high
A cap that's wildly above market is the other failure. Founders set them to avoid the conversation about valuation, or because someone in their network told them "raise on a high cap to preserve dilution." That advice is wrong when the cap is so high that the next round can't reasonably price above it.
Investors at the next round will not pay a premium just to validate your last cap. They'll price at what they think the company is worth, and if that's below your SAFE cap, your SAFE holders get converted at the cap (which is below the round price) and you've signaled to the new investors that the prior round was overdone.
A cap above market doesn't preserve dilution. It just delays a hard conversation and makes the next round harder to close.
What a reasonable cap looks like
For most pre-seed rounds in 2026, caps in the $5M–$10M post range are reasonable for first-time founders with early traction. Repeat founders, hard-tech, or strong early signals push that higher. Founders with no validation push lower.
These numbers move with the market. The right question isn't "what number do I want." It's "what number can I defend to the next round of investors, given what I'll know by then."
A defensible cap is one you can show a new investor a year later without flinching.
The negotiation, briefly
Don't let one investor set the cap unilaterally. Run multiple conversations. The cap that emerges from comparing three interested investors is almost always more accurate than the cap that emerges from negotiating against the first one who said yes.
If only one investor is interested, the cap they want is probably the cap your market will support. That's information about your raise, not a victory of negotiation.
What it means
Model the cap table. Compare offers. If you can't articulate why your cap is what it is to the next round's lead investor, the cap is wrong. Adjust now while it's a SAFE, not later when it's a priced round.