18+

Adults only

This season covers the business of the adult industry: companies, money, technology and regulation. There are no images and no adult content here, but the topic is 18+. Please confirm that you are 18 or older.

← Back to internet history
← all stories

Era 2 · Early web: subscriptions and the birth of high-risk billing

1995–2023the Meow Media, Inc. brand legally ceased to exist in…

Persian Kitty

A free directory of links to sex sites that former accountant Beth Mansfield launched in 1995 as a traffic experiment, and that banner advertising turned into a business earning over $1 million a year by 1998. The directory lasted under the same domain for almost 30 years, and its trademark was canceled in 2023 simply over a missed paperwork formality.

Founders Beth Mansfield (webmistress, according to Forbes, 1998; legal entity Meow Media, Inc.)
Domains persiankitty.com
s3-adultdirectorytraffic-arbitrageadvertising

In brief

Persian Kitty was a free directory of links to sex sites that Beth Mansfield, a former accountant from Washington State, launched in October 1995 out of pure curiosity: to see how much traffic could be gathered out of nothing. Within a few months the page reached hundreds of thousands of visits a day, and one of the sites listed on it offered to pay the hosting bill in exchange for a banner, which is how an advertising business was born that by mid-1998 was bringing in about $125,000 a month in gross revenue. Mansfield set the business up as a company, Meow Media, Inc., turned down an offer to buy the brand for $10 million at least once, and signed a publishing deal with a major magazine house. The site lasted under the same domain for almost three decades, but legally the business ceased to exist in 2023, not because of competitors or a court but because nobody filed a mandatory declaration with the US Patent and Trademark Office; the domain later passed to another operator and today promotes other people's cam sites.

How it started (the founders)

Mansfield, a 34-year-old homemaker from Tacoma, Washington, had worked as an accountant until 1995 and had no technical business of any kind. By her own account to Forbes, the idea was not entrepreneurial but almost exploratory: she "put up a site to see how much traffic she could draw." The decision to build a site specifically of links to sex content came from a simple observation: she visited a third-party hit-counter service and saw that the most visited type of site, "by far," was pages of links to adult content. From there, in her own phrasing from the interview, came "monkey-see, monkey-do logic": since it worked for others, she did the same thing.

The result outgrew the experiment: within a few months the page was getting more than 100,000 unique visits a day. The hosting bill grew to $400 a month, a sum that a homemaker with no outside income could not write off as a hobby. The solution did not come from Mansfield: one of the sites listed in her directory itself offered to pay that bill in exchange for a banner. It was this inbound deal, not a deliberate strategy, that set off the advertising model that later brought in six-figure monthly income.

Year-by-year timeline

Lesser-known but significant facts

  1. Persian Kitty's first advertiser paid not for an ad campaign but for a hosting bill. $400 a month was the entire starting sum that turned a page of links into an advertising business; the idea came not from Mansfield but from one of the sites in her directory.
  2. The company filed for a trademark on the brand only in 2001, six years after launch, when it was already fending off $10 million offers and signing a publishing contract with a major magazine house. Legal protection came to the business later than the money did.
  3. One and the same cluttered set of SEO keywords in the page's meta tag ("kittycat," "meow," "purple cat-o-nine tails," all jumbled together) survived at least five versions of the site's content and stayed unedited from 2005 to 2026: twenty years of spam that was not updated even after the change of owner.
  4. In spring 1998 the site appeared in Yahoo! exactly once, even though AltaVista returned 1,570 matches for the same query. The difference lay not in the site's actual uniqueness but in the indexing methods of the different search engines of the era.
  5. The brand died not because of a court or competitors but because of a missed form. The USPTO registration was canceled in 2023 solely because nobody filed the mandatory declaration of continued use (Section 8): technical negligence, not a market defeat.

Legend vs. the record

1. Directories died right after Google. The legend paints a sharp cliff: Google devalued manual directories almost instantly. The record, the site's own archive in the Wayback Machine, shows a different pace: persiankitty.com was updated practically without interruption from 1996 through at least 2021, that is, 23 years after Google was founded (1998). The first sharp break in the archive is not 1999–2001 but 2022-09-02 (the first redirect); the only snapshot of the new template slips through on 2023-06-05, 18 days before the mark was canceled on 2023-06-23. Verdict: the legend is wrong in this case; the directory died not from competition with search but from bureaucratic negligence during a change of owner, two decades later.

2. The directory was bigger than the portals in traffic. The legend relies on figures that Mansfield herself gave the press: 525,000 unique visitors a day by early 1998 sounds impressive and at first glance comparable to the major sites of the era. The record, independent (not self-reported) panel measurements by Media Metrix from the same period, paints a different picture: in the top 50 "Digital Media Properties" for September 1999, AOL Network drew 52.4 million unique visitors a month and Yahoo 38.4 million, while the only separately counted adult site, Porncity.net, took just 44th place with 3.35 million, more than 15 times less than AOL. Persian Kitty does not appear on this list at all, nor in the December 1998 top 10 according to the same Media Metrix. Verdict: the legend is not supported, and directories were not bigger than portals. The difference probably lies in incomparable methods: a self-reported count of hits on one's own server versus a national panel estimate is a likely source of the legend, although panel methods undercounting niche categories may also have contributed estimate.

3. A directory of links is zero work, just a list of other people's sites. The stereotype paints an aggregator as a passive list that updates itself. The record shows the opposite: as early as 1996 the site was divided into dozens of categories with separate sections for daily, weekly, and monthly updates ("Day/Week/Month"), worked its way to the top position on JumpStation, and received a third-party "Site of the Day" award; curation was work that others judged, not a one-time publication. Verdict: the legend misses the main point: the directory's value lay not in the list but in the continuous curation that kept it current.

The first growth lever

The first growth lever was not advertising but an organic position in a system not yet captured by search engines: Mansfield discovered through a third-party counter that link directories gathered more traffic than anything else on the web, and repeated the formula. Within a few months traffic passed 100,000 hits a day, and as early as 1996 the site was "#1 on JUMPSTATION."

Monetization was set off not by a decision of Mansfield's but by an inbound offer: one of the sites listed for free itself offered to pay the hosting bill ($400/month) in exchange for a banner. This exchange gave birth to flat-rate advertising that scaled along with traffic: $80,000/month by early 1998, $125,000/month gross by the middle of that year, ~$93,750/month net. The second lever was reinvestment in curation: categories and links grew by 15–20% in just three months between the snapshots of November 1996 and February 1997, which gave visitors a reason to come back again.

The paired story

The link with Danni's Hard Drive in this issue is not a thematic contrast invented by the editors but a documented business relationship. Persian Kitty showed a paid Danni's Hard Drive banner continuously from at least November 1996 to 2000, and then, in a weaker form as an ordinary link or a weekly feature, until at least early 2005. This is exactly the model that Danni Ashe herself described to the COPA Commission in 2000: directories collect and resell other people's traffic, while subscription sites buy that traffic and convert it into paying subscribers through content and retention. Persian Kitty here is a seller of attention, earning not from subscribers but from advertisers like Danni's Hard Drive. The pair shows two mirror-image models of the same early web (aggregating other people's content without producing it, versus producing content and retaining direct payers), and in 1996–2000 they were not competitors but each other's direct counterparties. The years, founder, and commercial fate of Danni's Hard Drive are recorded in its own dossier and are not repeated here.

Parallels today (projects from the catalog)

What a builder can take from this in 2026

  1. A first deal can come from being ready to accept someone else's offer, not from selling. The advertiser itself offered to pay the hosting bill, so be ready to turn operating costs into monetization when an offer arrives on its own.
  2. Putting off the legal registration of a brand is a risk that materializes. The mark was applied for in the sixth year, when the company was already fending off $10 million offers, and it died not from competition but from a missed form twenty years later.
  3. A directory or aggregator is continuous curation work, not a one-time publication of a list. Categorization and regular updates of the collections (daily, weekly, monthly) justify charging for placement.
  4. A partnership survives a change in the monetization model, but not a change in the platform's owner. The Danni's Hard Drive banner held on for almost ten years and disappeared at the same time the rest of the site's structure began to fall apart.
  5. A brand and a domain are different assets with different fates. A trademark is canceled over one missed piece of paperwork, while the domain goes on earning, now for another owner and another business.

Discrepancies and what we could not verify

Sources (primary first)

Secondary (context, cross-checking):

Internet History Keeper

Every historical dossier here is free and open, and it will stay that way. If you want the series to keep going (new dossiers, checks against primary sources, the English edition), support it with a subscription for $9.90 a month. It is support, not access, and you can cancel at any time.

Become a keeper — $9.90 a month →

The same thing, about today

The same breakdowns, but of projects launching right now: what the product is, where the first users came from, how they charge. The card is free, the full dossier is $5 (the dossier itself is written in Russian).

Browse the dossier catalog →