Search all the business literature you can and you’ll never find data analysis compared to romantic love. But, hey, why not? Love’s trajectories might hint at what the business world’s newly enabled generation of data analysts can expect.

These data analysts tend to be independent, are often creative and at least partly self-trained. They’re strapped to rockets from Tableau, Lyzasoft, Predixion, and others, tools that are at first deceptively toy-like. Aren’t they analogous to the garden variety teenager? Bothg groups revel in newly discovered tools, while both pursuits are fundamentally social — as Lyzasoft CEO Scott Davis observes about data analysis. His blog post got me thinking about this.

Everyone shows up ready to rumble. They’re fascinated with the possibilities, they experiment in private, later they have a blush of quick results followed by a long trail of self-training on finer points.

Each group’s toolset is potent and designed for early success but never early mastery. They make lots of mistakes. In love and analysis, people fall for the wrong data, mess up good data and dates, do all kinds of things they wish they hadn’t.

Without realizing, they face danger. I’ve noticed that behind most good trends comes a rotten sibling right behind it. Think of the history of other social events: Hippies begat the Summer of Love and then came Altamont. We celebrated “free love” and then came a surge of sexually transmitted diseases. Baseball begat the World Series and then came batters on steroids. PageMaker begat self-publishing but then came the ugliest lost-cat posters ever tacked on a telephone pole.

You may already wish that bad analysis would go away. Pete Warden, for one, warns of some fabulous ways people trip over new data. We could easily call this stuff “data porn” and ignore it.

But there are even more treacherous pitfalls. These potent tools can change everything in a flash (at the “speed of thought”). One minute you’re in orbit, and the next minute you wish you were dead. With sex comes the hazard of a painful breakup, and with data analysis comes the danger of unwanted speech that’s too hot for any public platform. Oops!

We have ways to deal with all that, but it’s never pleasant. The rejected lover picks up and leaves, and the analyst just finds his creative viz zapped off the cloud — by those who are themselves learning a new role.

The lover and the analyst both feel hurt, perhaps betrayed. Wasn’t each playing by the rules? Wasn’t each part of the group? Suddenly each one feels rejected for reasons that a hasty explanation doesn’t quite calm the hurt feelings.

In hindsight, we realize we shouldn’t have been surprised. Social pursuits can be like this.

By the way, who said good tools were the end of the story? Well, most vendors did. Some teenagers think so, too. But even slightly more advanced users know that technical proficiency is only the price of entry. We do the real work in many long conversations and collaborations with words, data, gestures, misunderstandings and reconciliations, and on and on.

Here the analogy breaks. The tools will keep getting better while the bodies fall apart. But the lesson’s the same: Tools enable, but conversation — better known in the business world as collaboration — is really at the heart of our work.

How do you approach unfamiliar data? An investment banker I talked to last week — one I know from a client’s whitepaper — rejects the “don’t think” method, advocated in my earlier post about Dan Murray. Instead, he thinks first, on paper.

“My approach is driven by having a bunch of gray hair,” says Michael Princi, managing director of ThoughtStorm Strategic Capital, a boutique investment bank and advisory firm in northern New Jersey. “I want to use my business acumen to tease out what might be the underlying issues.”

Experience counts The naive mind is prone to bad mistakes, he says. Take, for example, the 22-year-old analyst in India he employed who spotted this correlation: a firm’s revenue correlated with the number of Bobs in the workforce.

Hypothesize on paper “I first think through my hypothesis on paper,” he says. “‘It gives you a starting point.” If the model is wrong, as it often is, he just tries another one.

Test and repeat If the actual numbers are somewhat close to his expectations, he knows he’s on the right track. It’s the traditional consulting confirm-or-deny method. When the data does confirm his hypothesis, he’s able to run through the data again and again in iterations.

Test and repeat If the actual numbers are somewhat close to his expectations, he knows he’s on the right track. “It’s the traditional consulting confirm-or-deny method. It’s the quickest way I know. When the data does confirm his hypothesis, he’s able to run through the data again and again in iterations.

Does Michael Princi really analyze data differently from Dan Murray? They’ve never met, but Michael guesses not. He says of Dan, “I think he’s probably mapping it out intuitively.”

Do you have a routine for analyzing unfamiliar data? Please introduce yourself here.

I like the sound of Government 2.0: Collaborate with citizens online and you can change government from a sewer-dwelling raccoon into a purring housecat.

Social media lets us try for a kind of politics that was impossible until now. I hope for great results. For many, Government 2.0, or “collaborative government,” will mean just “friending” a local cop. But in full flower, Government 2.0 can mean far better service, and far more government-and-citizen collaboration than ever before.

Even before we had social media, the glare of public attention was a proven antidote for bad politics. Citizens getting up their elbows in policymaking has always been another strong medicine.

Trouble is, that “sewer-dwelling raccoon” is always smarter than people think. When he’s hungry, he purrs like a housecat and covers stinky laws with high-minded names. Advertising fools just enough voters — so often complacent and impatient — to throw a new law onto the books. On and on it goes.

Such a stinky new law is just what Californians got in 2000. Proposition 34 was sold to voters as campaign finance reform. It was a ruse. (A few days ago, a report confirmed suspicions, and a major drafter of the proposition insisted he was “outraged.” Yeah, and round up the usual suspects.)

One other fix, more honest, came 100 years ago: California amended its constitution to give citizens the ballot proposition. It was the only way for voters to bypass the paralyzed Legislature and loosen the Southern Pacific Railroad’s grip. It worked. But more recently, ballot propositions have helped tie the state’s budget in knots.

In the long run, who knows how social media, visual analysis, and other tools may be used in government? What will matter most of all is who uses them. If it’s “the people,” which people?

I hope this new, pervasive politics mobilizes a new wave of smart activists — the way desktop publishing and, later, weblogs enabled editors and writers. Or the way tools like Tableau and Lyza are enabling independent-minded, creative analysts today.

As these activists learn about politics, I also hope that more citizens than ever before step up to watch, push, and verify. Such a voter would be Citizen 2.0, the real hope.

Otherwise, it’s going to be that raccoon again — this time on Twitter.