Skip to main content

3 Agile Risk Management Tips from Shamu

Experienced agilists make it sound so easy, don't they?  They tell you over beers about the big, honking project they did last time.  They showed up, did some "workshopping," set up a little CI environment, and then, two weeks later, they began churning out working software.  And not just any working software:  high quality, low complexity, and perfect fit to business needs.  And why not?  The Product Owner was in the room to design and approve it!  Agile is fun!  Woo hoo!

This may not sound like the projects you do at all, where you experience confusion, resistance, fear, sweat, and tears, and you teeter daily on the edge of failure, public humiliation, alcoholism, and unwilling participation in corporate "resource efficiency" efforts.   How can you be an agilist when you don't share this ability to triumph over the grim corporate realities and do the impossible every day (twice on Sunday?)
http://simple.wikipedia.org/wiki/File:Shamu_in_SeaWorld,_San_Diego.jpeg
Here's a quick answer from Shamu, the quirkily named Orca(s) at the Sea World amusement park(s world-wide):  if it looks like a killer whale, and it swims like a killer whale, and it has killer whale teeth, then it probably IS a killer whale, and you need to be careful.  Agilists cannot be any more cavalier about project risks than anyone else.  Ask yourself:  are you "making a hard job look easy," or are you thinking that you have "changed the world so that we don't have risk any more." If it's the latter, please think again.

If you as an agilist are not finding, mitigating, and escalating project blockers, issues, and risks faster and with more rigor than your waterfall counterparts, then you are going to"fail fast," but not in a good way.   This flop is going to be without proper messaging, and on a schedule that is quite a bit faster than the one for which you, yourself, had hoped.

Every software project is a corporate financial investment backed by an explicit or implicit risk strategy.  Executives put money on the table to finance team activities with the expectation that the business will benefit from the expenditure.  That is why they are financing the project rather than putting the money into CDs and living on the interest (or buying a patent to exclusive-or bit-map representation and living off of the lawsuit proceeds).

When you are working on a big project, where a lot of money has been paid up front, then you own a piece of a sizable executive gamble.  What you do matters not only to you, and to your team, but to your whole organization, to your board of directors, and, depending on your industry, potentially to the whole world.

Which brings us back to Shamu.  What does Shamu have to recommend to you as an agile team participant, leader, or funder? 

1.  Risk is real.  Even in agile, and especially in agile, "risk management" is not a CYA activity in which boxes are ticked off, although that may be all you are doing today.
  • Effective project participants need to start with the expectation that the world tends towards entropy, and that if something can break, it will.  
  • Entropy needs to be constantly evaluated and triaged.  Do we have a crisis today, you should ask yourself, not just at the retro, but every day.  
  • If you do think you see something terrible about to happen, "pull the andon cord," as they used to say at Toyota.  Make a stink.  Get people's attention.  Make sure that the team acknowledges and fixes your perceived crisis-level problem or can explain to you why your perception is off, within a reasonable time frame.  And don't be too quick to think "it's just you." 
Problems come in many shapes and sizes but an actual crisis can come up on any day, and you want to be the one to point it out and prevent it, not the one crushed by it. 

2.  Put on a show, but be honest with yourself and your team, and fix your important issues.  As of 2010, Sea World has banned having trainers in the pools with the whales, after a whale trapped an experienced trainer at the bottom of a pool and killed her in front of a crowd.  For the 20 preceding years, the trainers had made it look easy, but it's not as though there hadn't been warning signs.  Wikipedia relates:  "On March 4, 1987, 20-year-old SeaWorld San Diego trainer, Jonathan Smith, was grabbed by one of the park’s 6-ton killer whales. The orca dragged the trainer to the bottom of the tank, then carried him bleeding all the way back to the surface and then spat him out. Smith gallantly waved to the crowd when a second orca slammed into him. He continued to pretend he was unhurt as the whales repeatedly dragged him to the bottom of the stadium pool. Smith was cut all around his torso, had a ruptured kidney and a six-inch laceration of his liver, yet he managed to escape the pool with his life."
  • Jonathan Smith is a bona fide hero for "going on with the show" under terrible circumstances to safeguard Sea World's flow of money in 1987.  
  • It's not as clear that Sea World looks good in allowing a stream of trainers to be injured before finally banning them from the pool.  And I don't even particularly want to get into the ethics of keeping Orcas at Sea World in the first place.
The point here is that it's a balance.  Your reputation as a successful software delivery person needs to be maintained, and face must always be saved.  But while you save face for yourself and your sponsors by staying positive, you need to be vigilent and powerful in getting the small group of people who have the power to fix your biggest risks to do so as quietly as they like, but quickly and efficiently.

3.  Please, never be optimistic.  Take the time you need to identify and mitigate identifiable risks before you start.  How much time you take, and how much you wait is going to vary by your context, by the stakes, and by the risk appetite of your funders.  Don't go for "fast" on principle.  Put this in context:  at its best, agile software development lets you find and fix your most important problems in a time frame that is months ahead of when the waterfall project would have even started standing up its first environment. 

Please, my friends, don't be macho.  Be smart.  As they said in the Sex Education Road Show at my college,  "Hope is Not a Contraceptive." 

Comments

Popular posts from this blog

How Do You Vote Someone Off of Your Agile Team?

One of the conundrums of agile conversion is that although you are ordered by management to "self-organize," you don't get to pick your own team.  You may not have pictured it this way, but your agile team members are going to be the same people you worked with before, when you were all doing waterfall!   I know I wasn't picturing it that way for my first agile team, so I thought I should warn you.  (I thought I was going to get between six and eight original Agile Manifesto signers.  That didn't happen.). Why "warn" you (as opposed to "reassure" you, say)?  Because the agile process is going to reveal every wart, mole, quirk, goiter, and flatulence issue on the team within a few hours.  In the old days, you could all be eccentric or even unpleasant in your own cube, communicating only by document, wiki, email, and, in extreme situations, by phone.  Now you are suddenly forced to interact in real time, perhaps in person, with written messag

A Corporate Agile 10-point Checklist

I'm pretty sure my few remaining friends in the "small, collocated team agile" community are going to desert me after this, but I actually have a checklist of 10 things to think about if you're a product owner at a big company thinking of trying out some agile today.  Some of these might even apply to you if you're in a smaller place.  So at the risk of inciting an anti-checklist riot (I'm sorry, Pez!), I am putting this out there in case it is helpful to someone else. From http://www.yogawithjohn.com/tag/yoga-class/ Here's what you should think about: 1.        Your staffing pattern.  A full agile project requires that you have the full team engaged for the whole duration of the project at the right ratios.  So as you provision the project, check to see whether you can arrange this staffing pattern.  If not, you will encounter risks because of missing people.  Concretely it means that: a.        You need your user experience people (if a

Your Agile Transformation Needs to Start with a Quiet Phase

From a really great blog post on agile adoption:  http://smoovejazz.wordpress.com/2011/02/16/an-agile-approach-for-adopting-agile-practices/ I've observed some different agile transformation patterns, and maybe you have too: Just Do Standups   (Shoot, then Aim):   some people feel that since you're "agile," you should just start doing stuff, like daily standups, and then build more of the the plan as you go.  Find a team and start doing some agile with them!  Start a revolution one practice at a time, one team at a time. Pros:   you're very busy from the start. Cons:   what exactly are you doing and why? KPI-Driven Change (Aim, then Shoot) : some people who have worked in large corporations for a while will tell you that to get the respect of the people, you need to start with a plan, support the plan with awesome printed and online collateral.  Then you "kick off," tell teams what to do, and measure them using "Key Productivity Indica