I mentioned in a previous column my penchant for certain types of television mysteries. One that has recently come to my attention is "New Tricks," and I thoroughly enjoy it. The premise is simple: three mature male police detectives who retired in the mid to late 1980s are called back to work on a special task force -- run by a woman -- closing unsolved crimes. Naturally, forensic science has changed in the intervening years (the use of DNA testing gets lots of use in "New Tricks") and a key element in the show is how these "old dogs" learn and make use of the "new tricks" of 21st century criminology. Much of the storylines demonstrate that police methods haven't changed much, although forensic technologies and disciplines have.
And that got me to thinking. Does the advent of a new technology, possibly a disruptive technology, ever require the advent of new methodologies in order for the technology to be successful? This question is more relevant than might be obvious. Some research for a client involved isolating disruptive technologies to determine if new technologies were causing problems in the data analysis.
What is "disruptive"?
The Cambridge OnLine Dictionary defines "disruptive" as "causing trouble and therefore stopping something from continuing as usual." I disagree. My definition would just be the second part of that phrase, "stopping something from continuing as usual." I prefer to see opportunities where others might see challenges and strongly believe in Lee Iacocca's statement, "We are continually faced by great opportunities brilliantly disguised as insoluble problems." I don't always or automatically believe what's going on right now is the way things should go; recognizing cause and effect where others might see just noise gets me invited to analyze best practices quite a bit.
Is it "best" practice or just "most often used" practice?
A "best practice" is "the most effective or efficient method of achieving an objective or completing a task" according to WordWebOnLine. A best practice may indeed be that, but you may want to consider some things before implementing a best practice in your business and work:
- Consider all the variables which affect what you want to achieve. When you've written them all down, go for a walk, go to the gym, get out of your chair and do something completely different. I'll bet other variables will well up out of your non-conscious. Some might be valid, some not. Add the valid ones to your list. Does the best practice you're thinking of implementing address all these variables?
- Has the best practice you're thinking of implementing successfully achieved the goal you wish to achieve historically, or has it merely come close? If the latter, how close, and is it close enough to achieve your goals in your present situation?
- Before implementing a best practice, ask to talk with companies where this best practice failed. Find out how and why it failed, and then determine how many of those failure elements are part of your business or business model. Can they be corrected before implementing the best practice or avoided during the best practice timeline?
This last element is a crucial one, I think, and I constantly discuss failure cases with clients and in presentations. I'm much more interested in knowing how, why and where things didn't work than knowing that things usually went well. An excellent for-pay paper on determining if a given "best practice" is right for your situation can be found in Business Process Management Journal (v11n6 2005).
A requirement of a best practice is that it can be continuously improved upon. This requirement insures that the best practice of Lee Iacocca's Chrysler is useful for today's Chrysler. Unfortunately, any "best practice" that proves successful here tends to become written in stone there. Given enough stones, a "best practice" that was successful in a few select situations becomes the "most often used" practice regardless of its efficacy. When failure occurs it's most often attributed to the individuals involved rather than the "best practice" not being best for the situation into which it's been applied.
Useful practices
This brings us back to "New Tricks" and answering the question, "does the advent of new technology require the advent of new methodologies in order for the technology to be successful?" There is a tendency in analytics communities and practices -- especially web analytics, business analytics, behavioral analytics-- to rely heavily on what's been in order to determine what will be. Frank Della Rosa, principle of FDR and Associates (USA East Coast), described this tendency in analytics as "looking in the rearview mirror to see what's coming at you on the road". Anyone doing analytics in any field knows the wonders and dangers of statistical analysis, and I'm sure everyone in the game is familiar with Andrew Lang's "He uses statistics as a drunken man uses lamp posts-- for support rather than for illumination."
Best practices do have a place. The idea of a best practice is an excellent one, and as an ideal it is something to be striven for. Experience has shown that many textbook best practices are like freshman engineering problems. They look good on paper but you couldn't build a working model if you were Scotty and the Klingons had given Kirk an hour to live (younger readers may translate that to "...if you were Geordi and the Borg were going to re-assimilate Locutus in an hour").
Here's a methodology for taking a best practice and making it a useful practice for your immediate situation:
Put yourself into the future: Assume the challenge (the reason you're implementing the best practice) has been met. Did another challenge grow out of meeting the original challenge? Can you live with what developed and the consequences?
- If not, what elements of the best practice need to be modified in order to avoid the future challenge?
- If so, what elements need to be modified in order to enhance the best practice's efficacy?
Put yourself into the past: Take the time to recognize clearly and understand as fully as possible everything that led to the present situation and the reasons for implementing the best practice. Does the best practice directly address all the factors involved?
- If not, how can the best practice be modified to address all the factors and how easy is it to do so?
Put yourself into the present: Take the time to imagine the implementation of the best practice. Go for what Einstein called a "Gedanken Experiment," and what cultural anthropologists describe as a journey. Allow yourself to envision the implementation of the best practice through to its successful completion. What happens along the way that is unexpected and unthought of? Can the best practice adapt to it without falling apart?
- If not, what needs to be changed in the best practice implementation today so that the things you've imagined won't blow it up down the road? How easy is that change to perform? Does changing the best practice initial parameters cause steps one and two above to need a little rethink time? Do so.
This outline provides a simple guide for learning how new technologies and new methodologies can be integrated into existing practices non-disruptively. My suggestion is to expand your practice knowledge to increase your expertise, thus increasing your worth to yourself and your clients. Doing so allows both yourself and those you work with and for to experience that one perfect moment in time when everyone learns some new tricks.
