Ben, you said you "stumbled across" this term. What was the context in which you found it?
Personally I would put "loose definition" in the same class as "concept by intuition" - something we abstract from a number of examples, as contrasted with a precise mathematical or other intensional definition - concept by postulate. I do not see "loose definition" as something new (having used this formulation in the past), but possibly as a formulational representation you may not have consciously noticed before.
As far as applying it to the terms 'good' and 'bad' as you have, well, we have the word "utility" from pragmatism. I view the word "cooperate" as requiring an animate subject, so this would be applicable to a living being, most likely human. If you intend to include non-living "stuff", then I would reserve the term "useful" to the purposes of the individual in question - relative utility - or "hazardous" to the purposes of the individual - relative disutility. Another factor, it seems to me, is the fact that persons often do not "know" what is in their "best" interest or contrary to that.
What a person "desires" is not always what they "need", and vice versa.
I would be inclined to include dating, indexing, context, and "to me" as qualifiers when applying such terms as "good" or "bad" - in other words, only apply them with consciousness of relativity to some explicitly stated situation.
"Someone's interests" however, it seems to me, is sufficiently vague as to convey very little information.
Interst in what?
I don't know what it might mean to further one's interests as contrasted with oppose one's interest when "one's interests" remains undefined.
Suppose "aiding one's interests" get to be viewed by another person as "contrary to that one's interests". How then can "good" and "bad" be "loosely" applied? Is the same thing or situation good for one and bad for the other?
What do you think of "I pointed out how something [I evaluated as] good in the short term [because it seems to me at the time to be in my interest at the time] might prove [me to evaluate as] bad in the long term [after abstracting and collecting more date], and vice versa."?
I previously related "loose definition" to "concept by intuition" in my prior post, and I contrasted it with "concept by postulation". Read Northrop for a more detailed discussion.
PS. You forgot "close enough for 'government work'".
I've been using the phrase "horshoes, hand grenades, and nuclear weapons" for decades now, probably ever since I was in nuclear power school back in the '60's.
The definition of "conceptual definition" appears somewhat ambiguous with respect to Northrop's classification, because it merely specifies defining one "concept" in terms of "other concepts". These can be as ambiguous or as fuzzy as "concepts by intuition", or they can be as precise as concepts by postulation. That it fails to make Northrop's distinction leaves it ambigous in this regard.
A "procedural definition" specifies human acts in terms of agreement with respect to the identification and use of objects and acts performed with these objects. Such a definition "assumes" that participants already understand less ambigous things and acts that can be shown and trained. It requires authoratative training to insure that the procedures are followed "in the same manner" using "the same" tools or objects. This is not altogether unlike the learning of the precise use of mathematical defintions, axioms, and rules of inference, although, in laboratory terms, it does depend on such concepts by intuition as "beaker", "scale", and other common objects, as well as acts done with these. Time-binding is of the essence in insuring consistency of use of operational definitions (as it is in learning and using mathematics).
Although I relate the two distinctions, I do so "losely". 
We can "define" "good" and "bad" in terms of "philosophy in the flesh" (relative to the individual), as that which we orient towards and that which we orient away from, respectively, noteing of course, that one source may result in individuals who orient both towards and away, and even a single individual who may orient ambivalently at different times.
See The Philosophy of Mobile Life and Think-Feel and Know-Act.
David, I did not associate "concept by intuition" with "immediately apprehended" in the sense that I infer you mean.
It is not the concept that is "immediately apprehended". The concept may take a while to learn.
Look at the definition given:
A concept by intuition is one which denotes, and the complete meaning of which is given by, something which is immediately apprehended.According to English, as I learned it, an insert that is set off by commas is not required. As I learned it, if the inserts was a required part, the commas would not be used. Perhaps this subtlety is not "universally applied", but applying it here "reduces" the required content of the definition to "A concept by intuition is one which denotes something which is immediately apprehended.". This does not say that the concept itself is immediately apprehended.
Once we have learned a concept by intuition, a process which may take more or less time and examples, we "recognize" the applicability of the learned concept to paradigm case examples immediately, and to many other extensions as rapidly as we recognize the similarity between the "immediately apprehended" object or situation and our previously learned paradigm case examples as modified by subsequent examples and learning. The more experience we acquire with using a "concept by intuition" the faster we can apply it - we can achieve the level of near instantaneous "signal reaction" in the majority of cases, and especially in the paradigm case examples.
A supporting "argument" for my position on this matter can be seen by considering that Northrop's distinction is intended to be a "binary" classification of concepts that goes along the extensional-intensional dimension. If we consider only those "concepts by intution" that people are capable of learning "immediately", then we no longer have a binary classification of concepts.
Since I personally grew up with and evolved a very precise notion of "definition", I consider the very notion of a "loose definition" a near oxymoron, but I'm not an authority in the sense of being able to command usage. To quote my high school English teacher, "Usage is correct.". Consequently I relate to your response in this regard. I think I might have tended to use the word 'notion' rather than 'loose definition' in some circumstantes, and 'characterize' or 'describe' in other circumstances. This approach, it seems to me, preserves some (verbal) awareness of the abstracting and inductive (not mathematical induction) nature involved.
When we "define" a (finite) set in mathematics by listing all of its elements, it is precisely defined by including all the elements that any variable on the set may denote. The word name of a concept by intuition gets its "definition" principally by example denotation, and its instantiation in any person's brain ultimately depends on that persons abstractive and generalizing capabilities.
We can note lots of arguments that we may infer indicate differences in apprehension (instantiation) of such a "concept", although the arguers would be described as immediatily apprehending the particular disputed example. ["Such a concept" may sound like it refers to a single entity, but I mean it as in the case of a map - the verbal formulation - that may be used to refer to many distinct territories - the instantiations in many brains. We have "direct" access to the formulation name - a single entity - but not to any other brain instantiations. This condition, it seems to me, applies to every word we use.]
Consider bean-bag "chairs" for example - a big loosely filled bag containing styrofoam or other packing material. The invention and use of these extended the notion of "chair", but in the early days only the inventors, and possibly a few others, applied the concept by intuition "chair" to these immediatily apprehendable objects, although prior to the common extension of the "meaning" of chair, immediately apprehending such objects did not necessarily invoke the functional knowledge "a place to sit".
The first one I remember seeing was at my first general semantics seminar. I was looking for a "chair" to sit in, and seeing none, Walt noticed my hesitation and pointed the "bean-bag chair" out to me. After the novelty and comfort, - like a big overstuffed chair, but one that hugs back, - it didn't take me long to update my "concept" of chair. I offered it to the next female entering the room. [Back in the days of chivalry].
David,
I believe Northrop must be referring to the former puctuated occasionally with the later (as in "show and tell").