How to revive a dead SaaS in 30 days, a buyer's playbook

The week you wire money for a stalled SaaS is the easy part. The 30 days after are where most buyers either build a real thing or quietly add another zip file to their hard drive.

I have watched both versions play out on Failedups. The buyers who restart abandoned startups successfully are not faster coders or better marketers than the ones who stall. They follow a pattern. Here is the one I keep seeing work.

A note before we start: most projects need 90 days, not 30, to feel alive again. The first month is not about growth. It is about proving to yourself that this thing is viable in your hands, so you actually keep going. That is the whole game.

Week 1: get it running, find the broken thing, take the keys

Day one through three is unglamorous. You are pulling the repo, reading the README, and trying to get a local dev environment to come up. Half the time the README is wrong. The seller meant well, but they have not run the project in eight months and node has had three releases since.

Budget the whole first week for this phase. Not because it is hard, but because if you rush you will skip the audit step, and the audit is the entire point.

What I do, in order:

1. Get it running locally. No shortcuts. If bun install fails, fix that before reading another file.
2. Get it running in staging on your own infra. Do not keep using the seller’s Supabase project, their Stripe test account, their Resend API key. You are going to forget you have not migrated, and one day they will rotate keys and your “live” SaaS will silently 500.
3. Read the database. Open the production data, even if it is 12 signups and a test row called “asdf.” This tells you what the seller actually shipped versus what the marketing site claims.
4. Find the broken thing. Every project I have taken over has one. A webhook that no longer fires. An OAuth flow that breaks on Safari. A pricing page that links to a Stripe product that does not exist. Find it before week two.

The handover call with the seller belongs in this week. Keep it to 45 minutes. Ask three questions: what do you wish you had done differently, who were your most engaged users, and what is the one thing you never got around to fixing. Write the answers down. They are worth more than the codebase.

Week 2: ship one visible thing, then tell the 12 signups

By the end of week one you have a working clone, a list of broken things, and a hand-drawn map of where the bodies are buried. Now you ship something a stranger can see.

One thing. Not three. The temptation is to rewrite the homepage, refactor the auth, and add a blog before launch day. Do not. Pick the smallest visible improvement that signals “someone is home now.”

Good candidates:

• New homepage copy that actually describes the product instead of the founder’s vision deck.
• A bug fix the original users complained about. Check the support inbox. There is always something.
• An updated changelog or a public roadmap with two real items on it.

Then, and this is the step buyers skip the most, you email the existing users. All 12 of them. Do not write a marketing email. Write a short note from a real human. Hi, I just took over this project from [previous founder]. I fixed [the bug they complained about]. Here is what I am working on next. Reply if you want to chat.

The reply rate on this email, in my experience, is close to 30 percent. That is wildly higher than any cold list you will ever buy. These are people who already raised their hand. They are the warmest possible audience to take over a SaaS project from, and most buyers leave them in the dark.

Week 3: one channel, all week

Now the hard part. You pick one acquisition channel and you do nothing else for seven days.

The mistake is dabbling. Two cold emails on Monday, a Reddit post on Tuesday, a half-written Twitter thread on Wednesday, despair on Thursday. None of those got a fair test, and now you think nothing works.

Pick one of these and commit:

• Cold email. Build a list of 50 ideal customers by hand. Write a short, specific message. Send it. Follow up once.
• Reddit. Find three subreddits where your customer actually hangs out. Spend the week being useful in them, then post one honest “I just took over this project” thread.
• Product Hunt re-launch. A new owner is a legitimate launch reason. Set the date, line up a few hunters, and ship.

You are not trying to win the channel in seven days. You are trying to learn, with both hands, whether this audience exists for this product. By Friday you will know.

Week 4: charge somebody

The last week has one job. Get money into Stripe.

Even ten dollars. Even one customer. The amount does not matter. What matters is that someone, somewhere, decided your half built thing was worth more to them than a coffee. That single transaction changes how you talk about the project, how you allocate your weekends, and whether month two even happens.

Most revival attempts die in the gap between “this works” and “someone paid me.” Closing that gap, while crude, is the most important move in the 30 days.

If by day 30 you have a working clone on your own infra, one shipped improvement, a tested channel, and a single paying customer, you have done it. You are not at growth. You are at the line where most projects on Failedups stop and yours keeps going. From here it is just iteration.

If you want to see what this looks like a year out, we collect the post-revival stories on the revived projects page. The pattern is almost always the same. Slow first 30 days, real first 90, profitable by month six.

The takeover is the easy part. The first month is where you prove to yourself that you meant it.