Correlation analysis and source selection

@FrankJon Two questions:

1. What do you mean by "approach"? What is an "approach"? Is it a process, procedure, or method?

2. What do you mean by "correlation analysis"? I know what it means in statistics, but I'm not sure what procedure you are talking about. What "correlation"?

3. What is your definition of "best value range"? How is it different from competitive range?

52 minutes ago, Vern Edwards said:

@FrankJon Two questions:

1. What do you mean by "approach"? What is an "approach"? Is it a process, procedure, or method?

2. What do you mean by "correlation analysis"? I know what it means in statistics, but I'm not sure what procedure you are talking about. What "correlation"?

3. What is your definition of "best value range"? How is it different from competitive range?

Vern -

1. What do you think of this source selection procedure? Specifically, utilizing a correlation analysis as a final step.

2. I'm referring to a statistical correlation analysis. The work is financial in nature and lends itself to this. If you'd like, you can think of this as any source selection consideration that is outside of the ability of offerors to control.

3. Recall that the agency intends to make multiple awards. I invented the term Best-Value Range to describe the group of prospective awardees once negotiations, final evaluation, and tradeoffs are completed. These are the offerors who will receive award unless the results of the correlation analysis dictate otherwise.

FrankJon,

Similar procedures to your suggestion have been utilized before. It’s somewhat related to cascading evaluations used to ensure small business receive representative share of awards.

Several agencies, including VA, also employed that procedure to provide adequate competition coverage under multiple award IDIQ contracts. The solicitation hypothetically covers multiple functional areas and offerors are free to propose on one or more functional areas. The solicitation states an objective is ensuring there are at least three awards for each functional area. The agency goes through something similar to your Phase 2. If an insufficient number of awards result for any functional area, they reserve the right to bring one of more offerors outside the “Best Value Range.” They simply picked who is next in line in ranking.

2 hours ago, FrankJon said:

Vern -

1. What do you think of this source selection procedure? Specifically, utilizing a correlation analysis as a final step.

Well, I don't really understand the procedure. It seems needlessly complicated. Why not just say that you're going to make multiple awards based on the number of "apporaches" proposed and let it go at that. (I assume that the proposed "approaches" are fully specified offers and that they will be incorporated into the respective statements of work upon acceptance.) Just treat each award as a separate source selection decision.

I don't understand the application of correlation analysis in this procurement. Correlation of what to what? Is it a price realism thing? If your team errs in its application or execution of the technique it might end badly.

But, again, I don't really understand what you're planning to do, so don't pay attention to me.

1 hour ago, formerfed said:

The agency goes through something similar to your Phase 2. If an insufficient number of awards result for any functional area, they reserve the right to bring one of more offerors outside the “Best Value Range.” They simply picked who is next in line in ranking.

It seems like the distinction would be that in my case, we wouldn't merely be bringing in a weak proposal for award, but swapping out an otherwise capable and deserving offeror for a weaker one because the weaker one happens to fill the remaining service gaps better than the stronger one.

1 hour ago, FrankJon said:

It seems like the distinction would be that in my case, we wouldn't merely be bringing in a weak proposal for award, but swapping out an otherwise capable and deserving offeror for a weaker one because the weaker one happens to fill the remaining service gaps better than the stronger one.

Wouldn’t it make more sense to just bring on an additional award rather than replacing one?

3 hours ago, formerfed said:

Wouldn’t it make more sense to just bring on an additional award rather than replacing one?

It makes more sense to me.

The correlation analysis thing makes no sense to me. It comes across as somebody trying to be "innovative".

15 hours ago, formerfed said:

Wouldn’t it make more sense to just bring on an additional award rather than replacing one?

11 hours ago, Vern Edwards said:

It makes more sense to me.

The correlation analysis thing makes no sense to me. It comes across as somebody trying to be "innovative".

Gents - To be clear, I’m not proposing to do anything. An agency has utilized this approach and intends to continue to do so. The reasons for doing so are rational and, in any event, the requirement office isn’t keen on changing its approach.

My question is whether this runs afoul of procurement rules or principles. Personally, it rubs me the wrong way to potentially disqualify an offeror over reasons outside of its control. But I struggle to articulate why it’s impermissible.

The best I can muster is that it’s akin to unstated evaluation criteria. Even though the agency alerts offerors as to its intention to conduct a correlation analysis, offerors have no way to prepare for it or guard against removal from the Best Value Range. It’s a crap shoot. They have to propose good value and then hope that their proposal is sufficiently complementary to other proposals.

1 hour ago, FrankJon said:

My question is whether this runs afoul of procurement rules or principles.

Possibly. But the answer depends on the procedural details. The process you described in your opening post strikes me as peculiar, but if fully explained in greater and accurate detail, and if clearly described in the RFP, it might make perfect sense and be okay.

I say let them try it and see what happens. Maybe nothing. Just make sure they can't blame you for an unhappy outcome

2 hours ago, Vern Edwards said:

Possibly. But the answer depends on the procedural details. The process you described in your opening post strikes me as peculiar, but if fully explained in greater and accurate detail, and if clearly described in the RFP, it might make perfect sense and be okay.

I say let them try it and see what happens. Maybe nothing. Just make sure they can't blame you for an unhappy outcome

Thanks for your thoughts, Vern.