Automated Scanning: The State of Play

We at Houston Foresight get asked fairly frequently about what is the best automated scanning tool. Our response is that there seem to be lots of them out there, we watch, we ask, we try, and as of yet haven’t found one that’s “it.” That day will come. The tools are getting better all the time … as they have been for the last 25 years or so that I’ve been exposed to them. I think I have a healthy fear and paranoia about missing the next big thing. We need to be on top of this. At the same time, relax and breathe.

My sense of the human vs machine scanning is that much of the push to automate is based on what I’d say is a faulty assumption about what scanning provides. There is a sense that scanning will uncover the “hidden gems.” That it will find an insight that no one else know about. I say “good luck.” There is just too much info and too many people looking to make that any kind of reliable basis for scanning. There may be a few instances where such a focus is useful, but for the vast preponderance of foresight work, scanning is gathering and sorting signals in organized way to get a sense of potential change ahead in a domain. We like the simple three horizons framework to organize the signals. Our experience is that any “hidden gems” of change will be picked up fairly early in horizon three – long before they have an impact on the domain (that is before they reach horizon one). It takes years … really decades. If you do your regular scanning, you will not miss a hidden gem!

Put another way, in our typical three-month project, a team of 3-6 scanners can get a handle on the scanning/signals of change in a domain in 4-6 weeks doing it part-time. If any of my prior scanners want to chime in here, we’d love to here you confirm (or deny) this. Much more scanning than that for a project is overkill [note: ongoing scanning is a different thing].

Nonetheless, you may be getting pressure to update, automate, and take advantage of all the great big data, analytics, machine learning, AI and stuff that should be revolutionizing scanning. Be very wary about those who say they have the answer. I have heard these kinds of claims, and unless it happened yesterday, I personally haven’t seen it yet. Use your  good consumer commonsense. My advice, if you have the time and budget, is to go ahead and try as many of the tools as you can. Compare, contrast, and pick one that you believe in the people and in the tool. Sure, there are snake oil salespeople, but there are also smart, wonderful, and well-intentioned people trying to good things – when you find those people, go on the journey with them. There is no doubt in my mind that scanning will be more and more automated and in a useful way – eventually. – Andy Hines