I've touched on reducing risk and how the pendulum keeps swinging away from Time and Materials contracts and to the Firm Fixed Price side. Then I wrote about how A-76 seems to favor the industry because it doesn't consider the total cost of ownership. Today, I'm going to focus on the third memo that was identified in the Washington Post article last week. This one is an easy one for me to write because it is something I have blogged about before.
Back in February I posted an entry about a web application that has gone relatively unnoticed in the federal sector. The Past Performance Information Retrieval System (PPIRS) should be a no brainer. Almost every solicitation indicates that the technical evaluation will be based partly on a review of the offeror's past performance. But in reality, PPIRS does not help in assessing that factor. Since my post 6 months ago, I feel stronger about this issue. I've used PPIRS, I have tried to verify contractor performance, and, I have found it to not be a useful tool.
The problem with PPIRS is data. The data in this application is terrible. I wanted to objectively evaluate the performance of several offerors. Each of them indicated that they were the prime on a government contract, one was even a DoD contract, and only a couple were actually in PPIRS. The DoD contract was not, which is surprising since PPIRS is run by the Navy. One of the offerors that was actually in the system was in there 4 separate times because their name was spelled 4 different ways.
Don't misunderstand, I really like the idea of PPIRS and I am all for requiring everyone to input more information. But before we start doing that, let's take a moment to identify the goals of what we are trying to do. This memo neglected to do that. The goal, I believe, is to objectively score past performance when awarding a contract. Based on that goal federal agencies should input information into a system (PPIRS) that will help acquisition offices in other federal agencies to form a clear opinion of that performance.
The problem, as I mentioned is data. What this system needs is a primary key that will allow contractor companies to be uniquely identified and trap nuances for spelling. Almost all of these recipients of contracts get paid for the services they perform and most pay taxes on those fees. As such, I would recommend the Tax Identification Number (TIN) be used to uniquely identify companies. Once the companies are known it is a lot easier to consolidate all of the information we have about their past performance.
Next, standardize the evaluation of the performance. This standardization should be on two fronts; qualitative and quantitative. The qualities that are evaluated should be as objective as possible. If you asked me, I would say, Cost, Schedule, Quality, Scope, Satisfaction and Risk, but I'm a Project Manager. In the qualitative review the contractor's performance in each of these categories should be evaluated. In the quantitative side, each of these should be scored from 0 to 10 (0 being the worst performance). These numbers could then be rolled up to identify an overall contractor score. Of course, this would greatly favor the small contractor that only has a few contracts and could hinder the larger vendors that have hundreds. As such, there should be a way to weight the amount of information we have to identify the confidence we have in the score that was assigned. As such small company ABC, that has only performed 1 contract, but did it well, could have a score of 59, but a confidence of 1. Whereas, big integrator DEF may have a lower score of 49, but a confidence of 200. This would give acquisition personnel what aren't familiar with those companies an opportunity to understand the rated performance and put it into perspective.
While getting on the right track in scoring past performance is a good idea and the right thing to do. What would make an even bigger impact would be to figure out a way of getting the government's Independent Government Cost Estimates (IGCE) under control. We are terrible at generating these. I am terrible at these, and I'm better than most. Estimates, regardless of where you are, are notoriously slippery. But, if we are going to be doing a better job with collecting information about the past performance of contractors, couldn't we use that information to help us generate better estimates too? Think about how many contracts the government awards per year. When I go to generate an IGCE for some segment of work, PPIRS should be the first stop to generate a parametric estimate.
No comments:
Post a Comment