How Buyers Actually Value AI-Built Apps (Forget the Textbook)

Forget what you read in MBA textbooks. At the size we're talking about, $5K to $50K, your app isn't worth what some DCF spreadsheet says. It's worth what one specific indie buyer, scrolling at midnight, decides it's worth in about 90 seconds.

That number is almost always a multiple of your trailing twelve months of profit. Everything else is decoration.

The TL;DR

1. ·The math: Asking price = TTM profit × multiple (usually 1.5x to 3.5x).
2. ·The multiple: Set by your churn, growth, age, tech stack, and how much of the work runs without you.
3. ·The discount: Tech debt, single traffic source, and AI-spaghetti code each shave 0.3x to 0.8x off.
4. ·The premium: Clean Stripe history, 12+ months age, and one boring tech stack each add 0.2x to 0.5x.
5. ·The reality: Most listings under $20K close at 1.8x to 2.4x. Anything above 3x needs a story.

The Math That Actually Matters

Here's the formula every buyer uses. It's not a formula they say out loud, but it's the one they're running:

Asking = (TTM Profit × Base Multiple) ± Adjustments

Let's make it concrete. You built a Notion-style writing tool with Cursor. Last 12 months: $14,400 revenue, $9,600 profit after Stripe fees and hosting. That's $800 MRR-equivalent in profit.

• ·Base multiple for a 14-month-old SaaS with <5% churn: 2.5x
• ·Tech: clean Next.js + Supabase, +0.3x
• ·Single channel (100% from one Reddit post that went viral): -0.4x
• ·One paying customer is 38% of MRR: -0.3x

Net multiple: 2.1x. Fair asking: $20,160. List at $24K, expect offers around $18K to $21K.

That's it. That's the whole game at this scale.

Multiple Ranges By App Type

Different categories of AI-built apps clear at different multiples. Here's what's actually moving on marketplaces right now:

Notice content sites use monthly profit and a 25-35x figure, not annual. That's a quirk of the SEO-acquisition world. Everyone else uses annual.

What Adds Multiple (Real Premiums)

Stop hand-waving. Here's what genuinely pulls your number up:

• ·12+ months of Stripe history: +0.3x to +0.5x. Buyers need a base rate for churn.
• ·One boring stack (Next.js + Postgres + Stripe): +0.2x. They can maintain it. AI spaghetti = the opposite.
• ·Documented onboarding, even a Loom video: +0.1x to +0.2x.
• ·Traffic from 3+ sources, none over 50%: +0.3x.
• ·Real category buyers recognize (CRM, invoicing, scheduling): +0.2x. "Niche AI productivity coach for left-handed pilots" gets discounted.

What Kills Multiple (Real Discounts)

The honest list nobody tells you:

• ·Code is 80% AI-generated with no refactor: -0.4x to -0.8x. Buyers open the repo, see 14 nested useEffect hooks, close the tab.
• ·Single customer is >25% of MRR: -0.3x to -0.5x. One churn event destroys the business.
• ·App depends on a free Twitter API or scraped data: -0.5x or more. Sometimes unlistable.
• ·You're the only support channel and you reply to every ticket: -0.3x. Buyer sees a part-time job, not an asset.
• ·Last commit was 5 months ago: -0.2x. Looks abandoned.

Real Talk

If your app is doing $400 MRR after 6 months, your "valuation" is probably $4K to $8K. Not $40K. Not $400K. The "10x ARR" multiples you read about on Twitter are for venture-backed SaaS doing $10M+ ARR with 130% net retention. You are not that. Neither is anyone you'll meet on an indie marketplace.

The good news: a $7K sale you close in 6 weeks beats a $40K listing that sits for 14 months and never sells.

The Three-Number Pitch

When you list on Birexit or anywhere else, lead with three numbers, in this order:

1. ·TTM profit. Not revenue. Profit, after Stripe, hosting, any API costs.
2. ·MRR right now, today. With a Stripe screenshot.
3. ·Time per week you actually spend on it. Be honest. 8 hours? Say 8 hours.

Those three numbers, plus a one-screen tech-stack summary, decide your multiple before a buyer reads a single sentence of your founder note.

What I'd Do With $20K To Spend

If I had $20K liquid and was scrolling Birexit tonight, I'd skip every listing that:

• ·Shows revenue but not profit (means there isn't much)
• ·Has only one screenshot and it's the landing page (no Stripe = suspicious)
• ·Asks 4x+ on TTM with no growth story (math doesn't work)
• ·Was built in the last 4 months (no churn data yet)

I'd zero in on a 14-month-old micro-SaaS at $600 to $1,200 MRR, clean stack, asking 2.2x. That's the sweet spot. You're paying for proof, not promise.

If you're selling, build your listing to be the one that buyer doesn't scroll past.

Your asking price is a bet. Make it a bet the math supports.