Many have postulated that every doomed dot com of the late 1990s is actually a viable business started in a reckless environment and way too ahead of its time.

Apparently the same goes for trends in the venture capital world.

All sorts of harebrained ideas for remaking, scaling, or innovating on the company formation and funding models that had worked in the Valley’s sleepier decade were tried in the late 90s. And almost all failed. The industry reverted to its boutique roots after the crash with some firms, like Benchmark, staunchly refusing to ever again try tactics that seemed like shortcuts.

Never say never. Benchmark has stuck to venture capital classics, but one by one the industry around it has begun embracing tactics that failed in the late 1990s. And the stunning thing is that many of them seem to be working. A business that everyone convinced themselves only worked when slavishly applied to a boutique, single-office, apprenticeship model can actually be modernized. (Or so it seems in 2012, at least.)

The latest example comes from First Round Capital. Today, the firm may be the first venture firm to ever offer a Cyber Monday deal. That either signals innovation or the apocalypse. No, this doesn’t mean you get a 20 percent discount on your Series A. Instead, First Round has brought together more than 25 of its ecommerce companies to offer discounts on a collective site. Some biggies are in there, including Fab, One King’s Lane, and Birchbox, along with more obscure names.

It’s impressive that the giants would share the spotlight with some of First Round’s more fledgling ecommerce companies on the biggest ecommerce day of the year. It’s also a part of something bigger: an aggressive strategy to create a network effect among its portfolio companies.

Other efforts have included a Quora-like platform where CEOs, VPs of marketing, founders, designers, and others within its portfolio can ask questions of other portfolio members. They range from banal (What should I pay per square foot in San Francisco for office space?) to more in-depth gut checks (My co-founder isn’t pulling his weight; what should I do?).

In the past, you’d take these questions to your investor; the investor might poll the portfolio and get back to you. That service is one of those much talked about “value adds” of venture capital. First Round is simply trying to make itself stronger by taking out the middleman.

It also offers one-off mini-workshops around specific tasks, like how to acquire users via Facebook; it builds apps and calculators to show what others in the portfolio are paying for certain roles in various geographies; and it has built a platform called “HackPR” that helps reporters reach more startups for trend stories, and in the process generates more press for startups. The firm has tried to take every common question or request and turn it into something the portfolio can help each other out with, whether that’s an event, product, or service.

If you’re new to the Valley, the theory behind this approach may sound novel. To me, it harks back to the old Kleiner Perkins idea of the “keiretsu.” Take this passage from a 1998 Fortune story, one of many on the strategy:

Only recently has KP hit jackpots that have given it cachet beyond any other VC firm. Beginning in 1994 with Netscape, KP has funded some of the most important and richly valued companies to have emerged from the Internet – Amazon.com, @Home, AOL, and Excite, to name a few. It’s done so with a strategy almost perfectly suited to the Web boom: nurturing alliances and partnerships among the companies it creates, weaving a web of relationships so dense you need a map to find your way…

Other VCs talk of building “networks” or “families” of companies, but KP speaks of a keiretsu, a metaphor Doerr came up with in the 1980s, when everyone thought Japanese conglomerates would eat American tech companies for lunch.

Sound familiar? The article goes on to say that KP had tried to build this network-centric approach during three previous waves of technology investing. But this time around, it was the Internet itself that had made it so powerful.

Here’s more from Fortune on what a membership in the keiretsu could do for you:

Most important, portfolio companies often engage in crucial partnerships. Early in their development, @Home, Excite, and Verisign, a provider of security for online transactions, forged deals with Netscape that helped lend them credibility. This was particularly true for @Home, whose CEO, Tom Jermoluk, was persuaded to leave the presidency of Silicon Graphics because Netscape was investing in and supporting @Home. KP put money into AOL back in 1987, and the relationship still pays dividends. Both invested in Preview Travel in 1995, and in the past few years AOL has made deals with seven other KP companies, two of which involved multimillion-dollar investments.

Alliances in the Kleiner keiretsu tend to be mutually beneficial, boosting Website traffic, revenues, and even the stock prices of both partners. Take the deal Netscape and Excite announced last May. Excite paid Netscape $70 million to become the primary provider of Web searches on Netscape’s Netcenter home page, an arrangement that guarantees Excite millions of additional page views a day. The boost in traffic bumped Excite ahead of Infoseek into the No. 2 slot in the bitter wars to dominate Internet search-site traffic. It now trails only Yahoo.

Sometimes a KP company goes public on the strength of its keiretsu partnerships. Take Preview Travel, which completed a successful IPO last November. Founded in 1985, the San Francisco tour packager was repositioned by CEO Ken Orton in 1994 as an E-commerce travel site on the Web. Kleiner Perkins and AOL invested in early 1995. A year later, with no involvement from KP, Orton struck a simple test-the-waters deal with Excite. But when he wanted to expand the deal and make Preview Excite’s exclusive online travel agency, Orton asked for Kleiner’s help. “We took full advantage of the keiretsu. We did not want another travel company to be in a KP family site,” says Orton, who recalls that Excite was also talking to Microsoft’s Expedia and Sabre’s Travelocity. Excite board member Khosla championed Preview with Excite CEO George Bell and Will Hearst, who was on Preview’s board. Lo and behold, within a month Preview scored a five-year exclusive deal that included co-branded travel content, revenue sharing from sales of airline tickets and vacation packages, and the guarantee of a steady stream of online visitors. In return, Excite gets $24 million over five years.

Fast forward a few years, and the word was scrubbed from KP’s website, never again to be referenced. What happened? I’m not sure KP has ever publicly said. Certainly, the firm still does some of these things, like helping with recruiting and hooking up interested potential partners.

But the spirit of the thing – that dot coms helped make other dot coms money, pushing them into the public markets before they might otherwise be ready – was one of the daisy-chain excesses that made the unraveling of the early Internet economy so dire. Not a thing to advertise in its aftermath. One of the biggest, boldest moves of the KP keiretsu was the multi-billion flame out of pushing Excite and @Home together.

This time around, the Cyber Monday site is the closest First Round’s companies come to actually partnering and doing business with one another. The spirit of its network is more about helping one another out, entrepreneur to entrepreneur, without seeking a direct ROI.

The only example First Round’s Phineas Barnes gave me about using the group to facilitate  partnerships had more to do with hooking up giant companies with startups. It polls corporate giants like Procter & Gamble and asks what their Web-related initiatives are in the next few years, and then scours its portfolio for fits. It’ll arrange one day when P&G executives can come in and meet with those startups all in a row.

“Actual partnerships have emerged from that six months or more later,” Barnes says. But that’s a world away from portfolio companies financially propping up fellow portfolio companies.

First Round’s version of the keiretsu – I’m calling it that, by the way; they wisely do not – is only one example of this reinvention of the late 90s. The most glaring has been the incubator. Not only did it fail the most dramatically in the late 90s, but – of all of these returning trends – it has succeeded the most dramatically today. Y Combinator is one of the most powerful forces in the startup world. The incubator has contained its losses by investing a small amount of cash in each startup, while two staggering wins – Dropbox and Airbnb – have so far buoyed the entire portfolio.

More nascent is the rise of professional value-added services inside VC firms, pioneered in modern times by Andreessen Horowitz and its “agency” model of venture capital. In the late 90s this took the form of superstar partners placing CEOs of one failing company into another, or perhaps having a single HR or PR guru on staff. Andreessen Horowitz has taken this to an extreme, employing a small army of experts to help companies with specific problems, most notably recruiting talent.

Andreessen Horowitz can do this because it has some $3 billion under management. Many in the industry who don’t have so much money under management quietly gripe that a venture firm with that level of headcount is wrecking the industry’s economics. In public, they may applaud how entrepreneur-friendly it is, but secretly they hope the strategy flames out à la the keiretsu.

There has also been the return of early-stage firms doing growth-stage deals – a phenomenon that had disastrous consequences in the bubble, and which many firms vowed they would never do again. Fast forward to 2011, when Russian billionaire Yuri Milner had amassed a fortune doing what early-stage VCs said they wouldn’t, and the mood has changed dramatically.

The trend has had mixed results. There is a massive difference in the market conditions between the two eras. In the late 90s, these were heady pre-IPO bets anticipated to double or triple money rapidly; in recent years they have been ways of providing financing to a company that was putting an IPO off for several years. Many firms have made money, but the results have been less stellar than hoped as the mega Web companies, such as Facebook, Groupon, and Zynga, have floundered in the public markets.

Groupon is trading well below its mega VC round of “like, a billion dollars” raised in 2011 – ironically one of the only ones Milner didn’t do. At the time, some of the participants privately crowed that, now the usual Sand Hill Road gang was onto the trend, Milner wouldn’t be able to get in on future lucrative deals like these. Ooooof. That’s not quite the deal to brag about, in retrospect. At least they were smart enough not to say it on the record. Some lessons from the 90s have been learned.

AngelList is even the incarnation of early crowd-sourcing attempts to let everyone be a VC. The most egregious of the late 90s was Draper Fisher Jurvetson’s meVC, a publicly traded entity aimed at letting everyday folks invest, that turned out to be a total disaster. Clearly, AngelList’s approach is far better. It gives actual value to startups first and foremost, and it doesn’t rely on the public markets. Instead, its founder Naval Ravikant has paved the way to let everyone be an angel via its platform by, you know, going to Washington and changing the law.

I was waiting for someone to recreate another DFJ innovation, its so-called affiliate network of venture firms spread all around the US with local partners. The hope was that the Internet had made starting a company anywhere possible, and that the collective might surface unknown gems in other markets, leading to superior returns. It didn’t, and most of the offices were shuttered.

Lately, 500 Startups is starting to look like an international version of that. It’s hard to know how 500 Startups’ international strategy will fare, but one thing is certain: There’s far more of a track record of startups in other countries producing returns today than there was in the late 1990s. Almost all of Sand Hill Road is investing around the world in some way, a giant industry-wide departure from the way business was done just 15 years ago. 500 Startups is just systematizing it in a way few incubators or firms have… much like Draper did with the increasing trend of VCs investing in other pockets of the US back in the 90s.

In this iteration, several of these trends have been mashed together, or pulled apart and distilled. It’s as if Andreessen Horowitz and First Round have both taken Kleiner’s keiretsu and applied different parts of it – expanding it dramatically through headcount in the case of AH, and in product in the case of First Round. 500 Startups is a mashup of the incubator trend and the affiliate network, with far lower costs of operation than either trend had in the olden days.

Many of the homages to late-90s VC may be unintentional. Industries have a way of reinventing the same thing over and over again. After all, these seemed like great ideas in the 90s for a reason.

But it seems the reason that many of these trends are working this time around has been driven by two simple economic realities, the same two that, fittingly, have also made failed consumer Internet companies of the late 90s viable new ideas.

The first reality is that the Interent is no longer a cottage industry. It’s a massive force with 3 billion people shopping, buying, selling, and clicking all over the world.

The second is that the cost of building a Web company has plummeted to some 10 percent of what it was back in the late 90s. These twin themes – the ability to make money from billions more people much easier and spend far less doing it – have made incubators viable on far less money, while enabling AngelList and Kickstarter to put together rounds that get companies to a product, and make even an affiliate network look like a decent gamble.

There may be one trend, however, that hasn’t returned and likely won’t – at least anytime soon. That’s the creation of publicly traded companies whose entire business is creating other companies and spinning those out into the public markets. Or as we knew it back then: CMGI, possibly the biggest incubator flameout of the era. That one relied on the market insanity of the late 90s, the ability to predictably flip new things out into the markets quickly.

If Facebook can’t make it, odds are there is no CMGI coming that promises to throw together the next 10 Facebooks with the reliability that public markets demand.