Here’s the math problem facing early-stage venture capitalists today. The vast majority of “exits” for venture-backed software companies, those happy events where everyone gets paid, are acquisitions valued at less than $50 million. All those large companies that go shopping for startups — Google (GOOG), Microsoft (MSFT), Cisco (CSCO), eBay (EBAY), Yahoo (YHOO) and AT&T (T) — may be buying up dozens of companies every year, but mostly they arent paying out-sized prices and venture firms consequently aren’t getting out-sized returns.

As an entrepreneur, you might be very happy with someone buying your company for $10 million or $20 million. But if you are the VC who invested $2 million in the first round of financing, a return of 2x or even 3x the venture firm’s money doesnt move the needle much on a fund that might be anywhere from $250 million to $600 million. To compound the problem for venture capitalists, many startups these days simply don’t need much money to get off the ground, thanks to cheap hardware and software tools.

So if there is a tension between the needs of a typical software startup and the needs of a venture capitalist, what do you do? If you are Sequoia Capital, you don’t do anything. Of any early-stage firm on the planet, Sequoia gets first crack at the best ideas and the best entrepreneurs –- everyone wants to work with Sequoia — so the odds of them finding the next YouTube or PayPal along with the big returns they bring are as good as they get. If you are Kleiner Perkins Caufield & Byers you start focusing a good deal of your energy on areas like alternative energy which, unlike software, do need boatloads of cash to get products developed and to market.

The other option is to tweak the traditional VC model. Firms like Union Square Ventures, and First Round Capital size their funds and investments so that a relatively small exit is still a meaningful return. In the past few years, “super angels” like Ron Conway have used a shotgun approach putting small amounts of money into dozens of companies a year. The startup boot camp Y Combinator, and the knockoffs it has spawned also fall into this category — put a little money into a lot of companies. These might be seen more as institutional seed funds.

The latest twist on the VC model is Tandem Entrepreneurs. Started by three former big company guys who cut their teeth at places like Oracle and Xerox Parc and then went on to found their own companies, Tandem is putting small amounts of money, $850,000 per company over two years, into just six companies.

At first it sounds like a recipe for disaster: small bets and not many of them. The difference, says Tandem’s Doug Renert, is that he and his two other partners, Sunil Bhargava and Joyo Wijaya, act as the operating team for the startups in which they invest for the first year. “Finding money is not the problem for most of these startups,” says Renert. “It’s the human capital problem, hiring the right team, where these guys get stuck against the wall and often fail.”

He and his partners solve that problem, Renert says, by bringing their expertise and their networks into startups that are typically not much more than two or three engineers and a nascent product. The way they divide the duties is straightforward: Renert is the deal maker, Bhargava is the marketing and strategy guy and Wijaya is the technologist. “We are the extended operating team for each of the businesses we invest in,” Renert says. “They don’t have to hire anyone in the first year. We do it all, and they just focus on building their product.”

Since closing its $15 million fund earlier this year from the likes of early Google investor Ram Shriram and a stable of tech company CEOs, Tandem has invested in three companies: Attassa, a Seattle startupthat plans to launch a product early next year that helps you get control of your e-mail inbox; Localyte, a Silicon Valley company in stealth mode currently that is building an online marketplace where travelers can book local services at their destination; and Zecter, another Silicon Valley startup that is launching a simple and secure way to share documents over the Web.

Tandem’s focused approach limits them to six companies at any one time. The two-year active investment period, however, will allow them to bring new investments on board, either when a startup finds its legs after a year or if it is clear things aren’t working. “If you can’t make it work in a year, you probably can’t make it work at all,” Renert says. “But unlike traditional VCs that put so much money into these things [that] it’s very hard to let it go, we will.”

Gordon Ritter, a partner with Emergence Capital, a Silicon Valley VC firm that focuses almost exclusively on investments in software-as-a-service startups, agrees with the notion that consumer web companies can succeed on a shoestring and viral marketing luck. But selling to businesses is different, he says, and still requires the financial muscle of a traditional VC. “Though the enterprise on demand businesses require less money upfront to get started, they require sales, marketing and operational expertise and capital to get to scale,” Ritter says. “In the business market, if you use the model of small investments spread over a broad set of companies, your losers will still be losers and your winners will mostly likely be owned by the more deep-pocketed investors who have the expertise and capital to stay with the company for many rounds.” Emergence happens to have those deep pockets.

Venture capital could use some change, says Paul Kedrosky, a venture capitalist, blogger. “There’s lots of room for innovation and mucking about with the venture model,” says Kedrosky, a senior fellow at the Kauffman Foundation who focuses on the future of venture capital. “You are seeing new models popping up all over the place like crazy, and a lot of them are these quasi-institutional seed funds. The trouble they all have is that doing seed investments is no easier, and arguably harder than doing Series A — you just stand to make less money.”

Tandem’s partners say their particular structure accounts for that. “We are like the Southwest Airlines of the VC world,” says Tandem partner Bhargava. “We don’t want to fly everywhere, we just want to pick a few routes that are profitable. We want to do deals that are too small or too early for traditional large venture capital firms and make our money there.”