(FORTUNE Magazine) – A few weeks ago, Shaun Holliday was flipping burgers. Now, flush with venture capital, he's searching Oak Brook, Ill., for loftlike digs for his high-tech startup, eMac Digital. But this isn't another fairy tale from the days of Nasdaq 5,000. The majority shareholder in Holliday's new company is old-economy stalwart McDonald's. His venture capital comes from Accel-KKR, a firm formed last year with the goal of funding startups culled from old-economy corporations. And strangest of all: Holliday, a former CEO of Guinness and Living.com, isn't relying on the kind of blockbuster IPO that first lured him to the online world. There's a chance that eMac will become nothing more than a small company selling software to restaurants. Which explains his burger-making: Holliday was employing the very old-economy trick of trying to learn the needs of eMac's potential customers--the 28,000 McDonald's locations he hopes will buy his company's software.

Welcome to the new, slow world of the Internet spinoff.

Since 1998, FORTUNE 500 companies have been trying to figure out what to do with their Web businesses. They were driven first by fear, as nimble, tech-savvy online upstarts rose to challenge them. Then, when the market gave their fledgling competitors outrageous valuations, that fear turned to greed. Corporations approached venture capitalists who, in return for a minority stake in the new company, would provide the business-building expertise and connections to the online world that the corporations felt they lacked. The new entities that resulted--often referred to as carve-outs--were going to become bigger than the parent corporations and reap vast rewards for those involved.

That never happened, of course. Net stocks collapsed. The speedy, agile competitors suddenly looked young and clueless. And the carve-outs were left questioning their existence. But instead of trashing the idea of the carve-out as just another relic of the dot-com era, both companies and venture capitalists have begun to rethink the whole process. Carve-outs are here to stay. And this time they might work.

To see where this trend is going, we need first to revisit what appeared one of the most promising carve-outs: Toysrus.com. In the summer of 1999, Toys "R" Us watched helplessly as tiny eToys captured its customers--and an $8 billion market cap. The press was full of stories about how eToys would bury Toys "R" Us, which then had a functioning, but hardly impressive, e-tail site. Toys "R" Us management turned to Benchmark Capital for help. They offered what seemed a perfect deal: In exchange for a chunk of Toysrus.com, Benchmark would provide the startup skills and online relationships Toys "R" Us lacked. Benchmark bit at first but backed out when Toys "R" Us refused--among other things--to allow the online store to set its own prices. That left Toys "R" Us without a Net-savvy partner for the coming holidays, when online toy vendors do 80% of their business. By the time America's stockings had been stuffed, Toysrus.com was a self-declared disaster. Fulfillment problems had caused it to miss thousands of orders, leading the Federal Trade Commission to fine Toys "R" Us $350,000.

Chastened, a recommitted Toys "R" Us carved out its Website, offering a minority stake this time to Softbank Venture Capital and removing the restrictions that led to Benchmark's departure. "You have an experience like that and it calls into question some of your assumptions," says Rex Golding, one of the Softbank principals who handled the deal and now serves on Toysrus.com's board. "They realized their Internet business could benefit from outside partners."

Softbank and Benchmark weren't the only VCs to recognize the potential benefits of helping old-economy corporations achieve the revenue growth--and market caps--of their dot-com rivals. Whole new ventures sprang up with the sole focus of creating carve-outs. At the beginning of 2000, Accel Partners, which had already funded carve-outs from the Washington Post Co. and Wal-Mart, teamed up with LBO giant Kohlberg Kravis Roberts to found Accel-KKR. "With our perspective in the heart of Silicon Valley," says Accel managing partner Jim Breyer, "and KKR's strategic and judgment capability, I felt we had a huge opportunity to guide the leaders of the old economy into new Internet opportunities." Later that year, VC icon Kleiner Perkins joined management consultants Bain & Co. and private equity powerhouse Texas Pacific Group to form eVolution, a VC firm dedicated to corporate ventures. "Each party recognized the attractiveness of the corporate venturing space," says eVolution worldwide managing partner James Allen. "The notion was that if you could combine the business-building experience of Kleiner Perkins, the dealmaking skills of TPG, and Bain's network of corporate clients, you'd have something very powerful."

That's why these alliances seemed to make a lot of sense at the time. In an environment where speed to market was at a premium, where entrepreneurial talent was hard to find, and where strategic partners often insisted on equity in an independent entity, a VC steeped in the Internet universe could provide a great deal of value.

But just as the trend exploded, the fear that caused it disappeared. For one, the market tanked. eToys, the onetime Toys "R" Us killer, saw its stock plummet (the company will shut its doors in April). And the logistical problems everyone feared proved to be less difficult than imagined. Toysrus.com avoided another fulfillment disaster simply by partnering with Amazon.com, allowing the online retailer to host Toysrus.com's site and handle all its back-end processes. This past Christmas, Amazon claimed a 99% accuracy rate on its deliveries. "On one side, we have the No. 1 retailer and buyer of toys in the world," says Toysrus.com CEO John Barbour. "On the other side we're aligned with the No. 1 e-commerce player." Which raises the question, does a FORTUNE 500's dot-com project really need to exist as a separate entity?

For companies considering carve-outs today, the simple answer is yes--and no. No, when the carve-out is nothing more than an online means of distributing the company's main products. In these cases, as the carve-outs mature, companies often choose to reintegrate them--a step clearly at odds with the VCs' hopes for a venturelike payoff from a successful IPO. "When we go into any of these investments, our goal is to create a viable stand-alone business," says Theresia Ranzetta, a partner at Accel. "But there may be situations where it ends up becoming more valuable to be reintegrated with the parent company."

To avoid these complications, companies are increasingly carving out businesses that are separate from--though complementary to--their core business. The less they tie in, the better their chance of remaining independent--and keeping a smile on the faces of their VC partners. "The core should be something that a corporation owns," says eVolution's U.S. managing partner David Sanderson. "The Wal-mart.coms should remain a part of the Wal-Marts." Sanderson's firm has even coined a consultant-style phrase--which he and his partners use at every opportunity--for the carve-outs in which it is willing to invest: "non-core but mission critical." That is how he describes eVolution's $16 million investment in MarketMile, a joint venture between American Express (which contributed $17 million) and Ventro ($13 million).

On the surface, MarketMile appears to be another of the increasingly ubiquitous online marketplaces. Set to launch in March, the new company will give midsized companies a single portal--the MarketMile online marketplace--from which to buy office supplies and other products from suppliers like Office Depot and Boise Cascade.

Amex's reasons for carving out MarketMile rather than developing it internally in many ways resemble those of Toys "R" Us. "Given the other players in the market, it was far quicker for us to do it outside our walls, where the engineers would be free to create something on their own," says Jud Linville, Amex's executive vice president of corporate services. But Linville is quick to differentiate MarketMile from high-tech ventures the company chose to develop internally--programs, for instance, that keep track of card billing. "That's our core business," says Linville. "And it should be something we invest in within our own four walls."

Like many of the new carve-outs, MarketMile's goal is to succeed on its own while also fueling one of the company's businesses. "MarketMile is the new model," says CEO Gayle Sheppard. "Our relationship with Amex will enable both companies to prosper."

The business that needs boosting in this case is the American Express purchasing card, which has failed to capture the usage fees and overwhelming market share of its more glamorous sibling, the corporate travel card. MarketMile incorporated the languishing card into its exchange's purchasing process: Buy a stapler and the carve-out automatically bills your card. "We are certainly focused on the top- and bottom-line growth of MarketMile," says Pierric Beckert, Amex's senior vice president of Interactive Business Development. "But this investment is also strategic. We believe that the growth of MarketMile will represent an opportunity for growth for Amex's purchasing-card business."

This new focus on strategy returns the carve-out to its roots as a means for building a new business, not just for minting money. In 1970, General Electric's Technical Ventures Operation launched what was likely the first institutionalized carve-out program. Divisions that weren't performing well or didn't fit into GE's overall strategy were turned over to Technical Ventures, which carved them out, keeping a minority share for GE and distributing the rest to venture capitalists and carve-out employees.

"Internally, the business was not large enough to satisfy all the usual requirements and hurdle rates for the corporation as a whole," says TVO founder David BenDaniel, now a professor of entrepreneurship at Cornell. The December 1973 issue of FORTUNE called attention to the "rather extraordinary corporate strategy" designed by GE "in an effort to turn to its own advantage what has been a pervasive problem for many companies intent on advancing the frontiers of technology."

Shaun Holliday has finished his tour of duty at McDonald's. After a week spent grilling, slinging fries, and, of course, supersizing, Holliday thinks he may have gotten to know the concerns of the restaurant managers he hopes will become eMac Digital's first customers. Now he and his five employees have to develop the software that will integrate their customers' purchasing, labor, and cash-flow systems so managers are free to spend more time where they belong--monitoring employees and assisting customers. Though McDonald's restaurants will become the testing ground for these products, eMac eventually plans to market its products to other restaurants. Which is one of the reasons McDonald's decided to pursue a carve-out rather than build the business internally. After all, why would Burger King ever buy software created by its archenemy? But more important, McDonald's didn't want its carve-out to distract the company from doing what it does best--sell hamburgers. "For 45 years, we have been busy perfecting a business model we are in love with," says Mats Lederhausen, McDonald's vice president of corporate strategy. "Any loss of focus from our McDonald's business is just a stupid idea."

FEEDBACK: nstein@fortunemail.com