Uh, Oh. Here Comes the Cavalry

By now, every one from the press to Obamacare's critics to the Obama Administration themselves have come to the same conclusion: that the issues with healthcare.gov are not just a few glitches that can be resolved in a few weeks but rather large-scale, fundamental system design problems that are going to take an awful lot of work to fix.

Today comes an announcement from Kathleen Sebelius that they are sending in the cavalry: a "technology surge" that includes a management expert to provide executive-level guidance for the massive project and "additional experts and specialists" that include "veterans of top Silicon Valley companies" to diagnose the issues and help fix them.

Rather than filling me with confidence and making me think "Great, now things are finally going to get done," this announcement left me saying, "uh oh." For, boy, have I ever been there.

It's a natural response to a technology crisis: throw everything you've got at it, bring in strong leadership, and get in as many experts as you can to iron things out. The problem is, they're not, metaphorically speaking, taking a defective car back to the shop to inspect, overhaul, and fix it; instead, they're trying to diagnose what's wrong with the car as it careens down a freeway and repair it without taking the foot off the gas.

When you bring in the proverbial army of experts, the first thing they need to do is get up to speed on the whole project and the whole complex system. They don't have the context, the history, the background into why whatever decisions were made ended up being made the way they were. At best, trying to bring the new experts up to speed just distracts the efforts of the team that's already in the weeds and trying their best to fix the issue. At worst, it throws a whole bunch more cooks--and pushy, confident ones at that--into a kitchen that's already way too crowded.

Any bright software engineer can take a look at the work of another engineer--especially one who is perhaps not as talented--and identify a zillion problems in the technical design and in the code and a zillion things that could be changed to "make it better." (The classic "not invented here" bias.) But, how many of those changes would actually improve the overall system performance versus making just slight improvements that are insignificant in the grand scheme of things?

It takes a very rare engineer indeed to be able to look past all the warts and knots and be able to find the small handful of changes that can bring stability and improved performance to a very troubled system. And, if you quickly determine--as I'm sure many of the experts will--that the whole system was architected wronged, well, how do you fix that, without going back to the starting line and rebuilding the whole thing?

My hope is that all these specialists and experts are truly cream of the crop and that they have the exceptionally rare capability to bring perspective and balance to the fire-fighting operation. My hope is that they can support, mentor, and guide the many teams of engineers as they frantically try to right the ship instead of just second-guessing and distracting them and sending everyone running off in all sorts of new, chaotic directions.

My experience, unfortunately, suggests its far more likely to be the latter and, if so, healthcare.gov is in for a very long, bumpy ride.