Loading page…
An online beauty store built on the brands department stores had no shelf space for, at a moment when investors were still saying women would not buy online — $10M of sales in its first year
120 employees · sold to Idealab for over $100 million in cash, two weeks before the NASDAQ crash — the employee count and the exit figure are the interviewer's framing, which the founder does not dispute
Their own market analysis said the category was closed: one group owned more than half of the beauty business and would not supply them. The thing that analysis missed is that the incumbent's share came from being able to afford shelf space — "the minute we opened up online distribution, smaller brands started taking share."
Fewer bars = easier, cheaper, or faster for an AI-assisted solo builder. Editorial judgments based on the case details.
Mariam Naficy co-founded eve.com in 1998, out of Stanford business school, to sell cosmetics online at a time when that sentence sounded implausible. She and her partner pitched SoftBank and were told, in her retelling, that women do not buy things online. Software was being sold online; clothes and shoes were not yet. They did the analysis a business school teaches you to do — a Porter five forces read of the market — and it came back negative. Supply was heavily consolidated, with the Estée Lauder companies holding more than half the beauty business, and those companies would not sign on as suppliers. On paper that is a closed category. What the framework did not surface is why the incumbent's share was so large: they could afford the key money required to get into department stores. Distribution, not product, was the bottleneck. Open a channel that costs nothing to stock and the small brands take share immediately. The second surprise came from customers. The plan was replenishment — creams and lotions you had already tried and were reordering. Instead, colour cosmetics flew off the shelves: blushes, eyeshadows, eyeliners, the things conventional wisdom said nobody would buy unseen. Women across the country were reading about small independent brands in magazines like Allure and had no way to buy them in their own cities. First-year sales were $10 million, which she says made them one of the largest e-commerce companies on the web at the time. The interviewer's framing is that the company reached 120 employees and sold to Idealab for over $100 million in cash — two weeks before the NASDAQ crashed.
The first customers were already looking for the product; they simply had nowhere to buy it.
The quick-reference card — acquisition channels, replication playbook, risks & traps, and our verdict — is members-only.
Join the Unicorn Club — $5/moCancel anytime, in one clickPaste into Claude Code / Codex and get a working version of this product end-to-end.
Two production-grade prompts per idea — a full build spec and an SEO growth plan — plus 10 tailored build specs a week. Members only.
Join the Unicorn Club — $5/moCancel anytime, in one clickEvery claim traces back to a listed source — open any of them and check the number yourself.
Data credibility: Founder interview on a podcast, recounting events from 1998–2000, roughly twenty-five years after the fact; everything is self-reported and nothing is sourced to a filing or third party. The $10 million first-year sales figure is the founder's own and she states it twice. ⚠️ The 120 employees, the six-month timeline and the sale to Idealab for over $100 million in cash all come from the interviewer's questions — she answers around them without restating or disputing them, so we have marked them as his framing. ⚠️ The $50,000 domain price is likewise the host's number; her own answer is that there was a cash payment and she does not remember the amount. The SoftBank anecdote, the Porter five forces analysis, the Estée Lauder companies holding more than 50% of the beauty business, the key-money explanation for department store distribution, the shift from replenishment skincare to colour cosmetics, and the domain story with Glenda and Eve Rogers are all her own recollections. ⚠️ The auto-captions garble several details: her co-founder's name is rendered only as "Vsha" so we have not asserted a spelling, the exit figure appears as "$und00 million" in one place and "a h 100red million" in another, and one date is spoken as 1989 where the context is unambiguously 1999. Location: the company was incubated inside Idealab's offices in Pasadena, which she states directly.
The first dollars arrived in the wrong category, and that is the useful part. The founding thesis was replenishment: skincare, creams and lotions — things a customer had already tried in person and was simply reordering, which is the obvious way to sell beauty to someone who cannot touch the product.
The loop ran through the supply side. Every small brand that had been locked out of department stores because it could not pay for shelf space now had a national storefront that cost it nothing to join, so brands came.
The hard stretch at eve.com came before launch, and it was made of expert opinion rather than silence. SoftBank told them women do not buy things online.
eve.com was an online store for cosmetics, launched in 1998 and trading from 1999, and its defining characteristic was its catalogue rather than its technology. The inventory was weighted towards small independent beauty brands that had editorial coverage and no national retail distribution, because the large houses would not supply an unproven online retailer. That constraint turned out to be the product. A customer in a city with no specialist boutique could read about a brand in a magazine and buy it the same day, which no other channel offered. The company was, in effect, a distribution network assembled out of brands the existing distribution network had rejected. The brand itself came with a problem she had to work around. Eve was, at the time, close enough to the name of a widely advertised American hygiene product that the association was the first thing she thought of, and part of the early work was steering the brand away from it. The domain came from Glenda Rogers, a programmer in Virginia who had bought eve.com as a present for her daughter Eve and refused to sell — she put the five-year-old on the phone and invited Mariam to convince her, which did not go well. Bill Gross at Idealab closed it instead, with an honorary board seat for the child, 1% equity in the company, a trip to Disneyland, free educational software for life, and a cash payment the founder no longer remembers. The two Eves are connected on Facebook today.
Retail margin on beauty products, sold direct to consumers, at a moment when almost nobody else was selling that category online. The business bought inventory from independent brands and sold it nationally at a markup, which is an ordinary model made extraordinary only by who was allowed to supply it and who was allowed to buy it. The economics that mattered were the ones underneath the margin. Beauty is a high-margin, low-weight, high-repeat category, which is what makes it attractive to ship — the founding thesis of replenishment was an attempt to lean on the repeat part. What they got instead was colour cosmetics, a discovery purchase with a different repeat pattern but an even better margin and a stronger attachment to the brand story that the magazines had already told. The cost side was customer acquisition. She says plainly that she spent money on acquisition at eve.com, and that the experience is why she chose a product with inherent virality for her next company. In 1999, with no organic search economy to speak of and a category that had to be explained before it could be sold, growth was bought. First-year sales of $10 million were, by her account, enough to place the company among the largest e-commerce operations on the web — a reminder of how small the whole channel still was.
Our read: the interview contains no technical or financial detail beyond the sales figure and the headcount, so this section is analysis of the shape of the business rather than anything the founder described. A 1999 online retailer carried costs that have since largely disappeared. There was no commodity e-commerce platform, so the storefront itself was custom engineering; there was no third-party logistics network of the modern kind, so warehousing and pick-and-pack were the company's own problem; and there was no organic acquisition channel worth the name, so growth came from paid placement and press. One hundred and twenty employees against $10 million of first-year sales tells you where the money went — this was an operations business with a website attached, not a software business. The part that has not changed is inventory risk. Buying stock from small independent brands means committing cash to products with no sales history, in a category where shade ranges multiply the number of units you must hold to look credible. The modern equivalent of eve.com would run consignment or dropship terms with its brands precisely to avoid this, at the cost of slower fulfilment and thinner margin. The founder's own retrospective judgement is the most useful cost lesson available: the acquisition spend at eve.com is what pushed her, a decade later, to pick a product that markets itself.
Copy the analytical move, which is the whole entry. The market looked closed because one group held more than half of it. She looked past the share to its cause — the incumbent could afford department store key money — and recognised that the moat was a fixed distribution cost, not a customer preference. When a new channel removes that fixed cost, share moves. Run this test on any category that looks consolidated: name the specific mechanism producing the incumbent's share, then ask what happens to that mechanism under the change you are betting on. Copy the catalogue strategy. Stock what has demand and no distribution. Editorial coverage with no purchase path is a measurable, findable thing, and it means someone else has already paid to create the want. Copy the willingness to be wrong about your own thesis in month one. They planned replenishment and sold colour, and they followed the receipts rather than the plan. Change the capital structure. This was a 120-person, venture-funded, inventory-carrying retailer. The version a small team can build today is a curation and distribution layer: consignment or dropship terms with the brands, no owned warehouse, and a content engine aimed at exactly the queries the magazine articles leave unanswered. Change the acquisition plan, following her own later conclusion. Paid acquisition was the only option in 1999. Today the same catalogue advantage can be converted into search and creator distribution — which is cheaper, slower, and compounds instead of resetting every month.
First, the evidence. The $10 million first-year sales figure is the founder's own, stated twice. The 120 employees and the sale to Idealab for over $100 million in cash come from the interviewer's questions; she does not dispute either, but she does not restate them. A $50,000 price for the domain is also the host's number — her own answer is that there was a cash payment and she does not remember how much. Nothing here is audited or sourced to a filing. The timing risk is the one the story itself announces. The company sold two weeks before the NASDAQ crashed, and her own account of why an early exit made sense is that nothing was going wrong — revenue was scaling nicely. Being right about a market and being right about when to take money off the table are separate skills, and this outcome depended on both. For anyone copying the pattern today, the risks are different from 1999. The distribution gap closes faster now: the same independent brands can open a direct storefront in an afternoon, which means a curator's advantage lasts only as long as its discovery and selection are genuinely better than a search engine's. Platform concentration replaces the department store — Amazon and the large marketplaces are the new fixed-cost gatekeepers, and they can list your catalogue too. And inventory remains the quiet killer in any owned-stock version: shade-heavy categories multiply units, and slow-moving stock from a brand nobody ended up wanting is cash that does not come back.
The four founder playbooks — first customers, first dollar, the flywheel, the silent stretch — are in the Unicorn Club, on this idea and on all 1,435 others. The story, the numbers, the sources and every later section stay open to everyone.
Join the Unicorn Club — $5/moCancel anytime, in one click