How to value a stalled startup
A practical framework for buyers and sellers, without the SaaS multiple LinkedIn cliches.
Most “what’s my SaaS worth?” calculators assume a healthy company with predictable MRR. That is not what we are trading on Failedups. We are trading half built things, often with zero revenue, that someone else will need to take over.
Here is the framework I have watched buyers and sellers actually agree on.
Start with replacement cost
Forget multiples. Ask: what would it cost to rebuild this from scratch?
Add up:
- Engineer time. A working SaaS MVP in 2026 takes roughly 200 to 600 hours of focused dev work. At a freelance rate of $80 to $150 an hour, that is $16k to $90k. Most listings here are at the lower end.
- Design and brand. Logo, basic landing page, system. Maybe $2k to $8k off the shelf, more for custom.
- Validation work. Customer interviews, surveys, the random $500 spent on Reddit ads to see if the idea has legs. Easy to forget but real.
That is your replacement floor. Buyers should never pay more than this. At that point, building from scratch is the better deal.
Subtract the “stale” discount
Code rots. Stacks change. Documentation thins out.
A reasonable discount:
- 6 months stale: 20 percent off
- 1 year stale: 35 percent off
- 2 or more years stale: 50 percent off, often more
Same goes for ecosystems that have moved on. A 2023 LangChain project written before the v1 SDK rewrite needs serious rework. That is a buyer only cost.
Add back the validation premium
Now the part most buyers undervalue: what did the seller already learn?
- Pre existing users. Even 5 active users in a beta is enormously valuable. Each is worth $200 to $2k of “would have spent on outreach to find them.”
- Distribution receipts. A blog post that got 10k views, a tweet that did 200k impressions, a launch on Product Hunt. Those compound for the new owner.
- Domain ownership. A clean .com is genuinely worth $500 to $5k depending on the name.
- A real founder note. A hard won “here is what I tried that did not work” saves the buyer 3 months of pointless A/B testing.
Negotiating
Most deals here close in the $500 to $10k range. The biggest mistake we see:
- Sellers anchor too high because of sunk cost.
- Buyers anchor too low because of opportunity cost.
The honest middle is usually replacement cost minus stale discount plus validation premium. Quote that calmly. If you and the other party can both write down the number with a straight face, the deal closes.
When to walk away
If the seller cannot show:
- Working code you can actually run
- A coherent answer to “why did this stall?”
- Honest run rate numbers (or zero, zero is fine)
… walk. The deals where someone is hiding something are the ones you regret.
Listing your project? Be the kind of seller buyers want to deal with. Honest beats polished, every time.