Sunday, March 11, 2007

Now is the winter of our discontent...


If this week's weather was any guide, we're looking at a glorious spring here at ASU. And in anticipation of that glorious spring, this past Friday, we had a little thank you picnic on the Hayden Lawn to recognize the hard work of the OASIS team, complete with yours truly in a Dunk Tank. Make no mistake. This was no "Mission Accomplished Celebration" to be sure. We're not hanging any banners yet. Everyone on the team knows we still have a long way to go, and a short time to get there. As I told the team on Friday, in the words of Winston Churchill, "This is not the end. It is not even the beginning of the end. But it is perhaps the end of the beginning!!!"

And taken all in all, it's been a good beginning. Given the way this team has hit its deadlines, deployed the systems, and shown an incredible ability in responding to problems to ensure that a system once deployed stays deployed, a beautiful late-winter afternoon was as good an excuse for a MidPoint Picnic as we needed.

For an ERP project of this magnitude, one that is just past its first birthday, OASIS is in awfully good shape. We've met every milestone so far. We're still under budget. The new system is already posting jobs and processing applicants. The schedule of classes is online. The student records are already converted, more than 50,000,000 of them. We're successfully processing student applications and we've been admitting students since November. We're computing their financial aid and we're registering them, by the thousands.

So while nobody's planning to don a flight suit and land on an aircraft carrier any time soon, its still useful to take a minute to take stock. And as I look at it, I have to say that ASU's strategy of Implement, Adapt, Grow is succeeding.

Recall that ASU opted for Implement, Adapt, Grow to avoid the Frankenstein Factor, a phenomenon I warned of in September of 2005. The Frankenstein Factor occurs when complex systems are developed, often over a period of years, without coming into regular contact with reality. Created in a castle laboratory, far from the pressures of the real world, the entire system is carefully designed and assembled, all the complex pieces carefully tried against test suites, and then assembled to create a whole. Only after the monster is complete do its creators jolt it with electricity in the appropriate fashion, bring it dramatically to life, and proudly lead it down to the public square to live among the people.

Who, of course, run screaming...

...because new systems always look different and are hence ugly. And taking a new system in all at once - even if the creating team has scrupulously given laboratory tours every Friday since the project's inception - when the monster finally comes alive and people are expected to work with IT every day, well...let's just say there's always an awkward period of adjustment. Even if the system eventually becomes a handsome prince.

Surprisingly to some, despite the best laid plans of the creators, every new system will have its flaws. As the military strategist von Moltke is credited with saying, "No battle plan survives first contact." No matter how carefully designed and implemented it may be, when a new software system first comes online, its flaws emerge, flaws that must be fixed in-vivo, with the electricity on. A project will be successful if it hits the right balance between upfront design and nimble response. Stay in the lab too long, and the costs mount without a guarantee that the system will be better for the time it spent there. Come out too soon and the flaws may be more than the institution can absorb.

ASU's strategy for avoiding the Frankenstein Factor has been to risk creating the system right in the middle of the public square, deploying the individual system components one-by-one as soon as they were able to stand on their own. In every case, their first contact with real data and real users has revealed problems, but so far they have been problems that the teams have been able to solve in-vivo. By staying on schedule, deploying 80% solutions and refining them in place, we are managing to advance the system on all fronts in an amazingly cost effective way that has so far avoided a full blown Frankenstein experience. This has been our strategy for managing complexity, and thus far, while not perfect, it is holding.

Which brings us to the winter of our discontent...

In my experience, in any major project there's a point in the middle somewhere where it begins to seem to some people like we'd all be better off if we'd never have started the whole thing in the first place. Surely, they say, someone should have planned this better. We should have bought a different product. We should have taken more or less time, or done a better job gathering requirements, or had a more complete understanding of how complex it would be. Think the deck of the Nina in early October, 1492. The third day at Thermopylae. A couple of days before the Battle of Agincourt. The midst of the London Blitz.  History is rife with examples to support any argument.
A period I would call "the winter of our discontent"

In the winter of our discontent, the focus can turn from looking forward to looking back. No matter how successful a system implementation may be, no one plays a perfect game, and when the winter comes, folks can start to dwell on what's gone wrong instead of what's gone right. While a complex system is going in, its construction makes lots of resource demands, but it has yet to provide most of its benefits. Deals get broken. Midnight oil gets burned. The mess of construction inconveniences everyone, making every job a little more difficult and some a lot more. In the middle of implementation, serious institutional challenges remain -- challenges of far greater importance than the system itself -- and these challenges are made daily more difficult because of the dust and general incompleteness of everything. During this period, in the middle of the tunnel, there is every temptation -- and every opportunity -- to attribute various institutional ills to the disruption caused by the construction process.

Consider the plight of the Dave's Dog House.

The Dog House is a popular spot on University Avenue here in Tempe and they make a mean Boston-style dog. But this year, because of the Light Rail construction on the street outside, some days it must be a lot harder to sell 'em than ever before. Dog House Dave may well be a Light Rail believer. He may be convinced that in the end he'll sell two dogs a day, every day to every Light Rail rider. But should he miss his 2007 sales goal, the disruptive effects of Light Rail construction will be the reason given. In the end, when the dust clears, and we can see with 20/20 hindsight, a careful analysis might find lots of factors went into a decrease in sales. Adding up all those factors, we might find Light Rail was nearly 100% of the reason for the shortfall, or less than 1%, because lots of factors go into determining how many hot dogs they sell besides construction traffic.

But if the Dogfather isn't selling what he needs to sell, I'll guarantee you he's not spending any time analyzing those causes. Because figuring the causes out isn't all that important when you get right down to it. In the end, what really matters to the Dog House is whether or not they sell enough hot dogs this year. So I guarantee you their spending all their time figuring out new ways to bring more new customers in the door for the first time and get all their current customers to come back for more. There'll be plenty of time to figure out the shoulda, coulda, woulda's after the tale is told. In the middle of the fight, it's about moving forward.

I used to work for this guy, Marty, who was great at surviving these hard winter periods. Deep in the middle of a project, often a skittish customer would grow concerned at a perceived lack of progress, and go over my head to Marty. Deeply concerned about impending risk, customers would sometimes call him and raise a host of issues, and complain of mistakes and mismanagement. In a couple of extreme cases, they even threatened to walk on a contract, or sue us, or defame us, or whatever.

Marty was great at this. He would listen carefully and patiently, without saying too much, 'til they'd had their say. Then he'd always tell them the same thing: "I understand that you're upset and that you feel that our team has taken certain actions and made certain mistakes that have put our joint project at risk. And I know that you understand that while we might acknowledge that we have made some mistakes, we feel most of what we've done has been right on target. We might even point to mistakes that you've made that have created challenges."

Right around this point, you could feel the customer getting ready for an attribution fight, a good old he-said/she-said, about who was right and who was wrong, and my lawyer will call your lawyer... But instead, Marty would say: "But none of this is going to help us succeed. We've both done things right, we've both done things wrong. And from where I sit, we can still be very successful in what we set out to do together, as long as we focus on moving forward. So let's put aside the conversation about how we got here. We're committed to helping you meet your goals, so let's focus on what's important to you, and figure out how we work together to advance..."
I can't remember a time he wasn't able to get the client to agree and turn a difficult situation into a success.

The OASIS story is full of good news, but it still faces its share of uphill climbs. And in some quarters, the challenges threaten to obscure the path to success. But if we follow Marty's advice, and continue to focus on the challenges and find solutions, we can bring this ship safe home.

There'll be plenty of time to look at the scoreboard at project's end. The tale will be fully clear by fall. And just so everyone is clear about it, the success of OASIS won't be measured by whether or not it came in on time and under budget. Those factors are important, sure, but the real test of OASIS' success will be:
Did ASU meet its 2007 enrollment goals?
Was ASU able to recruit the faculty and staff it needs to succeed?
Were we able to retain students in greater numbers for the fall?
Has OASIS positioned ASU to grow?

If the answers to those questions are yes, then OASIS will be judged a success. If the answer to any of the questions is no, then the judgment will be harsher, whether OASIS is 100% or 1% the cause. All we can do, every day, is all we've been doing every day: focus on the goals of the institution, and solve the short and long term problems that get us closer to meeting those goals.

It's a core part of the OASIS strategy. Implement, Adapt, Grow. All along, we're identifying errors and fixing them. We're streamlining processes. It is painful and difficult and disruptive -- and so far it's working.

Knowing what I know about these implementations wherever they have been done, if someone last year had shown me where we are in a crystal ball last year and asked "will you take this position or play the game again," I'd take this position in a minute. This game is ours to win.

So if now is the winter of our discontent... here's to making it a glorious spring...


Gene,  March 22, 2007 at 1:31 AM  

"Because of their courage, their lack of fear, they (creative people) are willing to make silly mistakes. The truly creative person is one who can think crazy; such a person knows full well that many of his great ideas will prove to be worthless. The creative person is flexible -- he is able to change as the situation changes, to break habits, to face indecision and changes in conditions without undue stress. He is not threatened by the unexpected as rigid, inflexible people are."

-- Frank Goble

Charlie,  April 30, 2007 at 9:37 AM  

I'm not the "sun of York" and haven't made any glorious summers lately, but I do have a question about the growing number of podcasts.

What steps are we taking to ensure that podcasts are accessible to students/public who are deaf or hard of hearing? I haven't seen any transcripts linked/associated with iTunes podcasts, those posted via the UTO pages, or those on the President's pages.

Are we talking to faculty about making their podcasts accessible?


Richard -- not the monster More/Shakespeare conjured!
aka Charlie