Government waterfalls and data compasses
A book recommendation, why government (and other) projects fail, and numbers (yay!)
What if I told you that for every 1% of your casework backlog you reduce, your case submissions increase by 1.4%? How about the return on investment (ROI) of a forensic laboratory is far, far better than even the most profitable Fortune 500 company? Better than Apple? Or if there may be an excellent reason to routinely over-test your toxicology samples?
In this Substack, I haven’t talked much about one of my main research topics over the last decade plus, that is, the performance of forensic laboratories. Project FORESIGHT is a benchmarking process that hundreds of forensic service providers (FSPs) around the world use to evaluate their operations and improve their performance. Because most FSPs are governmental, that puts some constraints on how they can go about fixing things. There’s hope, however, in a wonderful new book that I’ll talk about and how the traditional (read: bad) way of project management is not helpful.
To that end, I’m always on the lookout for like-minded systems/process/design geeks who want to improve things, especially government operations. One of those people is Jennifer Pahlka, the founder and former Executive Director of Code for America. In her recent book, Recoding America: Why Government Is Failing in the Digital Age and How We Can Do Better, Pahlka talks about her time working with government to improve their performance and make their benefits and resources more accessible. She works to get our government out of the rigid, Industrial-era culture that distances, or even disconnects, policy from implementation.
Her work shows that there are three overlapping problems that are common to many government agencies. The first is that, over time, policy complexity builds up and is never cut back. Like geologic formations over millennia, layers upon layers of policy accrue, making it difficult to understand what can and cannot be done because of the density and complexity of rules. This leads to the second problem, namely, staffing. Learning all those nuances of policy implementation and interpretation takes time. If it takes years to become even minimally competent in a job, hiring more people is not a solution. The third and final problem is a top-down, risk adverse culture. If “management” is worried about how it might look if you addressed a problem (“Problem? Problem!??! Our agency does not have a problem! Do you know what would happen if we admitted we had a problem?”) and the front-line employees cannot opt for change to improve their governmental processes, then government becomes mired in mediocrity and inefficiency. As Pahlka told NPR, “In some ways, I think this is probably the core issue here because when you're in a culture in which the people at the bottom aren't able to tell the people up the chain what's wrong and how to fix it for fear of optics, you can't really solve problems.” Sound familiar? Yup, thought so.
In some ways, I think this is probably the core issue here because when you're in a culture in which the people at the bottom aren't able to tell the people up the chain what's wrong and how to fix it for fear of optics, you can't really solve problems.
I highly recommend Pahlka’s book and her website for it. One of the core concepts she espouses is that data is a compass, not a grade. To that end, let’s talk about a compass that was built specifically for FSPs.
Project FORESIGHT
FORESIGHT was created in 2008 and dozens of academic papers and articles have been published on its use, utility, and benefits. Each year, my colleague on FORESIGHT, Dr. Paul Speaker, produces a report to provide summary metrics of all the data for that year. Moreover, each participating FSP gets a tailored report, showing them exactly where they are in their performance ratings. Municipal, state, and national FSPs have used FORESIGHT to improve their performance, reduce backlogs, increase salaries, and trim operations to maximize their limited resources. The American Society of Crime Laboratory Directors (ASCLD) awards the top-performing FSPs with the Maximus Award, recognizing those who operate in the top 10% of efficiency of all FSPs. Coming soon from CRC Press is our book, The FORESIGHT Manual: Benchmarking Performance for Forensic Service Providers. It is a step-by-step approach to help the curious and wary adopt benchmarking and how to use it to do more good with what resources you have.
The Waterfall Model of Project Management
Another of her core concepts is “The waterfall is a pledge not to learn.” What’s a ‘waterfall’? The waterfall model is a project management approach that breaks a project into distinct, sequential phases. The phases are completed consecutively, similar to the flow of water in a waterfall. Each phase must be completed before the next phase can begin. The waterfall model is popular in software engineering and product development.
The waterfall model has a serious flaw, however: Once a phase is completed, it can be difficult (or impossible) and costly to revisit a previous stage. This means no take-backs, no revisiting, no recursion, and no feedback. This is a disastrous way to manage a project once you think about it. It flies in the face of any systems or quality process, which are inherently iterative.
You’ve probably seen Gantt charts in one or more projects you’ve been involved in: They only go forward. Has any project ever gone only forward? No, not realistically.
Basing projects on a waterfall model almost guarantees failure at some level. Information, decisions, and resources only go one way with no opportunity for downstream results to inform upstream actors. As Pahlka says,
Waterfall software development may result in poor government services, but when insights, agency, and power flow down a hierarchy, but rarely bubble back up, you’re caught in a waterfall culture that goes far beyond tech.
Agile project management is much more realistic in its outlook. Agile project management is an iterative approach to managing projects, involving teams working on different phases of a project simultaneously. Agile project management promotes flexibility and collaboration. It enables teams to work quickly, adapt to changing requirements, and ensures that feedback can be acted on quickly.
Forensic science providers seeking to use benchmarking to improve their government services would do well to read Pahlka’s book, read the forthcoming Project FORESIGHT Manual (shameless plug, I know), and adopt a systems and agile approach to managing your process improvement projects.