Non-technical strategies

In the last few weeks I've written about ways to get peers to adopt a technology you like, and how to get your managers to adopt it too. Today I'd like to point out some "non-traditional" strategies you might employ when those fail. This list was created when talking with a reader about how to get approval for installing a trouble-ticket system.

Often the non-technical push-back is against the entire concept of ticket systems and nothing will be "good enough". In that case, don't bring a knife to a gun fight. In fact, find a way to avoid the fight entirely.

The Art of War and other strategy books would suggest alternate strategies like these:

  • Privately confront the primary dissenter directly: talk privately with the person to find the reasons behind their actions and settle those issues. Enlist them as a supporter.
  • Go around the dissenter entirely: set up the ticket system of your own choosing for a project they are not involved in, when it is successful it will be politically difficult not to expand its use to all projects.
  • Go over the dissenter's head: get the dissenter's boss on board.
  • Leverage influential people: If there is someone that the dissenter feels walks on water and can do no wrong, get an endorsement from that person.
  • Act faster: install something and put it into action before they can push back.
  • Act slower: are there benefits to putting off the decision? For example, will the dissenter retire or change jobs soon? (You may not be allowed to know that they are on the way out. If your boss smiles knowingly when you ask, maybe they know something you don't know.)
  • Produce more data: Gather data and produce charts that show undeniably you are right (don't show a single charts that disagrees; if the dissenter doesn't have the raw data, they can't make those charts).
  • Produce less data: Work in secret to build the system.
  • The power of crowds: Can you get a lot of other people on board such that the dissenter is outvoted?
  • The Power of the Demo: Are they rejecting a system they haven't actually used? Install your preferred solution on a VM and give demos to likely supporters. (The secret to a successful demo is doing at least 5 dress rehearsals)
  • Divide an conquer: Find out where the opposition isn't in agreement with each other and play one side against the other.
  • Isolate dissent: Identify the dissenters and exclude them from the process (find a politically viable justification for this).
  • Overload the dissenter: Give them so much other work to do that they don't have time to dissent; or put so much of the research on their shoulders that they ask to be taken out of the decision process.
  • Reduce the choices: Don't show 15 different models and hope they pick the one you want. Only show options that you will accept.
  • Give too many choices: Show so many potential products that they are overwhelmed; declare your expertise and recommend the one you want.
  • Selective comparison: Show 1 really awful system followed by a perfect demo of your system. (In a related note: At a singles bar always stand next to an ugly person.)
  • Force a "win": Get agreement to default to your solution if a decision isn't made by a certain date ("because we can't delay ProjectX"). Make sure you've given them more work than can be accomplished by that date so-as to trigger the default.
  • Make the dissenter think they are making the decision: If you ask a child "what do you want for dinner?" they'll ask for ice cream. If you ask, "Should we have hamburgers or hotdogs?" they'll think they're making the decision even though you've already made it for them. (Worst of all: don't list choices one at a time, they'll keep saying "no" until you run out of choices: "Do you want hamburgers?" "no" "Do you want hotdogs?" no "Umm... well, we have ice cream" "yes!").
  • Take advantage of emergencies: In an emergency the normal decision process goes away. Can you create a situation (or wait for a situation) where you can get permission to install RT or ORTS "just for this one emergency" and then take advantage of the fact that "nothing is more permanent than a temporary solution"?)
  • Bullies only respect other bullies: Declare that your solution is the ONLY solution and brow-beat anyone that disagrees.
  • Discredit the enemy: If the dissenter is always going to find reasons to reject something, don't try to deal with the points they bring up; discredit the dissenter's opinions. ("He isn't a real stake-holder, why should we listen to him?" "He rejects anything new, remember the time....", "He won't even be using the system, why is he causing trouble for us?")
  • Running code beats vaporware: a running system beats the theory that it won't work.
  • Avoid the issue: Find another project to work on that will make you a success; leave this "can't win" situation to co-workers that are suckers.

If done right, these strategies could work or could get you fired. Proceed with caution. Work with your boss, or if you boss the problem, confer with peers.

Please post comments with your suggestions and experiences. (This website now supports OpenID and other systems.)

Posted by Tom Limoncelli in Technical Management

No TrackBacks

TrackBack URL: http://everythingsysadmin.com/cgi-bin/mt-tb.cgi/1147

Leave a comment