A paltry metric
[Programming note: I am on vacation for the next week during the kids' spring break.]
Here's a fun trick. Want to get a contracting officer riled up? Tell them that the most important thing they should measure is their Procurement Administrative Lead Time. As a metric, PALT is an absolute lightning rod for acquisition-professional rage.
On paper, PALT — which is defined as the time between the solicitation date and the award date — is an incredibly powerful and useful metric.[1] Knowing how long something should take to complete, on average, is incredibly important for things like capacity planning and forecasting.
But, acquisition professionals have strong reactions to PALT because, in practice, times can vary widely depending on a given requirement. Some things take less time, some things take more. And yet, for a lot of acquisition managers, PALT is treated as a performance target and individual contracting officers may get strong or weak performance reviews depending on their awards' PALT time.
As a consequence, for many acquisition professionals, PALT is considered to be subject to Goodhart's Law: "When a measure becomes a target, it ceases to be a good measure."
Last week, the Government Accountability Office published a report — Defense Contracts: Better Monitoring Could Improve DOD's Management of Award Lead Times — that both proves the conceptual value of PALT and the continued misunderstanding of its purpose.
The report effectively did a data analysis of new DOD awards between fiscal years 2019 and 2022,[2] and evaluated PALT across different dimensions, including "total contract value, contracting approach, contract type, extent of competition, and the type of product or service procured."
Unsurprisingly, they found that PALT varies quite a bit across those dimensions:
We found that PALT can vary (1) depending on factors such as contracting approach, total contract value, and contract type, and (2) by the different types of products or services procured. For instance, with regard to contracting approach, we found that median PALT values for orders were much shorter than the medians for definitive contracts and indefinite delivery contracts.
Quite predictable and validating to any acquisition professional. Indeed, the report includes numerous line charts and bubble charts that emphasize how much PALT can vary. As a forecasting tool, then, PALT is actually pretty good for predicting the differences among acquisition strategy choices.
But the report also highlights the problem. For example, when describing how PALT is used within the Defense Department, the GAO explained that PALT is basically used as a managerial intervention tool:
The Army’s fiscal year 2023 PALT memorandum explains that contract award time estimates are to be used as targets, understanding that actual award times will vary based on complexity, dollar amounts, and other unpredictable situations. An Army official added that once PALT delays are identified, a team of representatives from all the contracting commands works together to review the causes of these delays. Officials with the Air Force Sustainment Center stated they have weekly or biweekly reviews of the status of all in-progress procurements to discuss constraints or other potential impediments which could delay a planned award. Similarly, officials with the Naval Facilities Engineering Systems Command stated that each month, the chiefs of each contracting office meet with contracting officials to discuss procurement obstacles and recommend potential streamlining measures to assist in reducing PALT.
Small wonder that contracting officials get riled up over PALT. On the one hand, the data demonstrate that PALT is highly variable. On the other, you have to sit through status meetings to talk about why your PALT is variable.
And then there's the "tool":
To address known data quality issues, DOD is pursuing measures to ensure the accuracy and completeness of data in its PALT Tracker—the tool DOD established to track contract award times for planned procurements with an expected value greater than $250 million. However, DOD components do not find the tool useful and reported that it is burdensome and duplicative of other systems they use. Additionally, DOD and its components have made limited use of the PALT Tracker.
Look, I don't know who needs to hear this. But if your solution to the problem of poor data quality is to add burden to the people who enter that data, you're going to be unhappy. And sure enough, based on the report and DOD's response to it, my guess is that the PALT Tracker will be decommissioned.
If you're in government, chances are that nothing in this article is particularly new to you. It is just common knowledge among acquisition professionals that PALT is both a useful metric and a terrible target.
But for those on the industry side, consider this the next time you're frustrated by slow deal cycles: the contracting officer is likely just as frustrated as you are. They might even be in a status meeting talking about it—possibly being judged by their bosses or auditors—based on an innocuous metric turned bureaucratic device. Or they might be entering data into a tool that no one uses.
But if you're inclined to ask them what's taking them so long, you've been warned.
[1] I've written before that the formal definition of PALT is probably wrong, and continue to hold that view. Whatevs, it's fine.
[2] As an aside, a thing that is often overlooked in federal procurement is that most contract actions are not new awards but are, in fact, actions taken after award, such as modifications and exercising options. But PALT ignores all that and only measures new awards.