Qaeda-linked terror suspect Najibullah Zazi left a digital trail a mile wide for federal investigators to follow, but stopping seems to have required too much luck and footwork.Based on information that federal investigators have released to the public, it's clear that Zazi left a suspicious electronic footprint in a number of areas, including landlines, cell phones, his laptop computer and video footage of him buying bomb-making chemicals. It's all available to investigators and prosecutors because they know where to look for the data -- and they used enormous amounts of human intelligence to keep track of him and his whereabouts.

But what about the role of technology? Karan Hoss, CEO of the beauty supply chain where Zazi shopped for his weapons-grade hydrogen peroxide, told the Los Angeles Times he was glad to have invested in the many cameras that captured the suspect in the act of buying the chemicals. "Thank God for good technology and good camera systems," he said.

But just how much technology was actually used? Because preventing terrorism isn't just a question of software that can "listen" to wiretaps, but also a way for data mining applications to get into the disparate databases created by a wide variety of communications technologies.

In other words, the problem becomes finding a way to rein in what EMC president Mark Lewis referred to (in an entirely different context, by the way) as "information dispersal."

The reason I bring up Lewis and EMC (and data quality vendor Informatica for that matter) is that, like giant businesses, government has a lot of stuff stored in a lot of places; unfortunately, a plethora of technological and institutional silos keep that data apart, which can get in the way of timely analysis.

As Hoss said, thankfully we have the technology to gather all that data. Thankfully, we're also developing tools to sift through the residue of that technology, and finding correlations thanks to improved search capabilities incorporating associative semantics that allow us to really connect the dots -- even if we're not sure of what we're looking for.

Hopefully, the government is following the lead of some of the world's most progressive companies in developing information management strategies that apply those tools (as opposed to simple information gathering and storage).

As I noted in my opening, despite a relative (if understandable) paucity of information about the processes used to investigate Zazi, a pretty clear picture is emerging of lots of footwork and guesswork at the outset.

But sweat equity and getting lucky aren't substitutes for real information management strategies. I hope Vivek Kundra is listening, because preventing terror attacks isn't the kind of thing that happens by accident.