Vaccine Management Systems: Why the easy choice was expensive, dysfunctional, and really no choice at all
Here’s something that drove me nuts recently. These are two news stories, one from July 2020, and the other from just this past February. They both involve no-bid contracts between governments and Deloitte (not to pick on Deloitte — they just happened to be the target of both articles).
The first is a $16 million deal with the State of California to update their unemployment system, which was groaning under the strain of applicants during the pandemic:
The second was a $48 million no-bid contract with the federal government to develop a vaccine tracking and management system — which, as it turns out, was rejected by most states:
Wait a minute…why did these government bodies choose Deloitte, without asking for a bid? They used the very unhelpful phrase “Deloitte was the only responsible choice.”
But was it?
Responsible Choice, Easy Choice, or No Choice?
Here’s my guess: These government bodies needed work to get done and get done quickly. Deloitte has already previously jumped through all the outdated bureaucratic hoops to be a vendor to the federal government. So the government went with what it knew, maybe with a sigh of relief that they didn’t have to go through the bid process.
And Deloitte got to name its price tag.
Notice what wasn’t part of the process: Demonstrating a track record of making systems that actually work, or proving the company’s ability to deliver when requirements are novel, nebulous, or even downright contradictory. (I’m not saying Deloitte couldn’t prove those things. But can we all just agree it’s weird that no one asked them to?)
This wasn’t a responsible choice. It was, in the eyes of everyone involved, the easy choice…which was, in effect, to make no choice at all.
Price Tags: How Complex is This Software?
So Deloitte got to name its price. And the scary thing is, this is not all that uncommon. The big players do it all the time.
What was behind that $16 million system? That’s the thing: No one knows. But based on what they said it could do, I would estimate that it would have been a $250,000-to-$500,000 bid.
To be clear, I am not opposed to companies making a profit. (I mean, I want that for my company, too.) And I understand companies pricing themselves according to value. But here we are talking about a 32X price increase over a conservative estimate.
Why do I say this? Because the cost of managing a system should not be based on the number of people that are going to be using it. It should really be based on the complexity involved in making a product like this.
And let’s face it: Software has come a long way. It can understand when a traffic light a block away has turned green, and what the traffic flow is doing around that. It can identify faces at the airport, and forecast the movement of planets centuries from now.
By comparison, being able to manage vaccine sign-ups and tracking should be child’s play. It’s Atari’s Space Invaders in an age of Call of Duty: Black Ops Cold War.
Just the Symptoms
The reason these caught my attention is that they represent problems that go way, way deeper than just devising vaccine tracking software. And so it is a good jumping-off point for other discussions I want to have.
Those discussions will take much more room than a single article. But maybe it’s worth bringing them up by asking some questions:
- Why does the federal government award contracts this way? Does this really help the taxpayer? Or the government itself?
- Why didn’t the government consider some other, smaller firms? (Deloitte is big and proven, but it also has a lot of mouths to feed. Is that behind the price tag?)
- If the government were open to smaller firms, how would they handle the project differently?
- How does a firm prove its ability to handle a project, regardless of size? And how can it do this when the requirements themselves are pretty nebulous?
- What is the right size for a project like this? How do you get a team that is nimble while still capable of the work?
I can’t count how many times I’ve seen it, whether it’s the federal government or not: Large organizations that have dreams of large software decide the solution is to get a large number of people to develop that software. And yet, most of the time, that is not the way to go.
I know that sounds counterintuitive. But if it turns out to be right in even one case…imagine how much time and money we would save. And perhaps have working software, too.
