AI Directories: $10,000 a Month from One-Time Purchases, Built on a Chore He Automated

Every time he shipped a new product, Sergiu Chiriac repeated dozens of manual submissions to listing directories. So he automated the chore for himself. People with the same problem started asking, and he began selling it. That became “AI Directories”, now the largest income source in his portfolio.

What he sells is the submission work

The product is simple. He hand-curates a database of directories that list AI tools and startups, checking whether each is still active, whether it has SEO value, and what a submission requires. Buyers order bulk submission into that list.

The official pricing has three tiers.

Every tier includes a report, a listing in his own directory, a list of paid directories, and one newsletter mention. Separately, a listing in his own directory sells standalone for $49, including, per the interview, a DR70 dofollow link.

What buyers are really paying for is time. Take the Premium framing at face value, $199 for 70 hours saved, and it prices out below $3 an hour. Whether that beats doing it yourself depends entirely on what your hour costs.

The substance of this product is a list. He says he has spent “countless hours finding startup and AI directories, checking whether they still work, and gauging their SEO value.” What blocks entry isn’t code. It’s this accumulated research. But lists rot: directories fold, submission requirements change. The merchandise depreciates, so maintenance itself is a cost of goods.

The three numbers don’t agree

The first thing to check in this case is that the published numbers don’t line up.

All are self-reported. None are third-party verified. The natural reading: $9,500 was an unusually good month, and $6,500 is closer to normal cruising. The 900-vs-1,000 gap is probably just update timing.

Still, the headline “$10,000+/month” refers to yet another level, distinct from both the $15,700 peak and the product page’s $6,500. Three different figures for the same business, at the same time. That state of affairs is itself worth registering. We use the more ordinary-looking $6,500 as the base below.

Divide $6,500 by the $99–$199 price band and you get roughly 33–66 orders a month. One or two a day.

Revenue that doesn’t compound

There is no recurring billing. He concedes as much: “mostly one-time purchases rather than subscriptions, so months swing a lot.”

The product makes this hard to avoid. Directory submission is a do-once job. The same customer has no reason to buy again next month. Every month restarts from zero toward those 33–66 orders.

There is an exception. As his own directory grew, a second line appeared: guest posts and paid listings. Unlike submissions, this sells slots in a media property, customers rotate but the business continues. Amounts aren’t public, so the mix is unknown, but it bends the pure one-time picture.

Among our cases, one-time-heavy portfolios like Marc Lou’s, stacked to $1,032,000 a year across small products, share a pattern: low prices, spread across many products. Chiriac likewise runs TransClipper (short-video transcript analysis) and BrandReply (mention tracking inside AI assistants, in development) in parallel.

Being his own first customer

His refrain is that “distribution is often harder than building”, and this business literally sells that difficulty. Because he sells the chore that bothered him, improvements come from felt experience rather than imagination, he argues.

“Solve one problem properly first. Once you’re trusted, ask the same customer about their other problems. Expanding within the same customer is much easier than hunting new markets.”

On market entry he is equally crisp: “Research before you build. Study competitors before writing a single line. Competitors are a good sign, proof a market exists.” Treating competition as validation echoes the API developer who only picks markets with 2–3 competitors making $20,000–$80,000 a month.

SEO is the main channel

Channels cited: SEO, GEO aimed at AI search, listings in other product directories (eating his own dog food), founder communities on X and Reddit, and cold email to startups. He calls SEO “the biggest growth channel” and invests heavily in content and backlinks.

Positioning organic traffic as “the most durable source of customers” is consistent with the merchandise: a business that sells backlinks acquires customers through backlinks.

What a buyer cannot verify

The record lacks the numbers a buyer would care most about: approval rates after submission, traffic gained, ranking movement. None are published.

How much directory backlinks move search rankings has long been contested. We have cases like SEOJet, a backlink-planning SaaS, but few sellers publish measured effects. DR70 describes the linking site’s strength, not what happens to the buyer’s site.

This is not an indictment. Demanding controlled effect measurement from a $99–$199 solo product misses the point. But note that the “70 hours saved” value is explained, while the “rankings improve” value is not. The former is the sound reason to buy.

What this business needs to keep running

Demand flows from the fact that AI products keep launching in volume every month. If new launches slow, submission customers thin out, and with no recurring revenue, the decline shows up immediately.

His own listed failures point the same way. Several co-founded projects didn’t work. He concluded solo suits him. He burned out on 14-hour days and now keeps weekends off. And he names “staying focused” as his biggest challenge, writing that he should have spent more time improving the products that earn, instead of starting new ones.

A man with a $6,500-a-month one-time-purchase business still itches to build new things. Revenue that doesn’t compound and a self-confessed focus problem are, probably, two sides of the same coin.