Yet from what I’ve read, as alleged, CPUSA extrapolated from one test that mutual aid is not effective and is pushing that as a line for the entire party.

The essay talks about a local chapter's stance on mutual aid. I didn't get the impression it reflects a broader party line. Frankly, the author is too light on details for us to really know what happened. Look at how oddly vague this is:

But our local leadership poo-pooed mutual aid as “ineffective” and “not worth it.” The metric being used was apparently the efficacy of converting MA recipients to dues-paying party members. Evidently they “tested” doing mutual aid work and opted against it, but from all appearances, the “test” was a one-off attempt

What does "poo-pooed" mean? Does the chapter want to reduce time spent on mutual aid but still do some, or do they want to do none at all? What do they mean by "apparently" and "from all appearances"? Do they know about the chapter's experiment(s) with mutual aid or not?

If nothing else, there's a lesson here about being specific when levying criticisms.

I broadly agree with all your points about how ideas should be tested -- you need at least a few tests to draw conclusions, and how you execute those tests can change a lot. Unfortunately, we're stuck talking about this in the abstract because the details here just aren't clear.