I have to agree with a lot of what Johnk said. Too many people, including folks on this thread, get seduced by the "I modded it so it must be better" perspective. In all equipment, but especially with speaker crossovers, a "better" component may not actually result in better sound. There are too many other factors involved.
Consider another context. We all know that when we try different preamps or amps in our system, it is often easy to notice certain qualities that we think are improvements but it may take longer to notice failings. Well, the same thing occurs when swapping capacitors, chokes, resistors and wire in a speaker crossover. The new part will undoubtedly change the sound but is it really an improvement? Are there some negative qualities that go along with the positive ones? Often the negatives may seem minor at first but become major issues later on. It does indeed help to return to the unmodified version as a reality check. (I don't agree with making changes to one speaker and leaving the other stock but that was beaten to death earlier in this thread.)
Adding to the confusion is the breakin factor. Nearly all new component parts take time to reveal their true sonic character, so they can't be judged within minutes or even hours after being installed. My approach with crossover parts is to hook them up with a dummy load on another system and run them 24/7 with FM to let them break in prior to trying them in my speakers. Then, after I install them, I switch back and forth between the prior component and the new one every few days or so in order to get a real fix on how each of them sounds (in the context of my speaker). This takes a lot of time and I would not say it is always fun, but I don't know of any other way to evaluate parts objectively.
As an example, I have been trying Duelund CAST resistors in my speaker crossovers. These took a long time to reveal their true character. If I had only listened for a few hours, I would have said they were unacceptable---detailed and dynamic, but too wispy in the highs and lightweight in the bass. Yet after putting a lot more hours on them, I am now very pleased with how they sound. (Incidentally, the CAST resistor sounds nothing at like the regular Duelund resistor.)
So the bottom line is that you do have to go back and forth between parts to assess their true sonic qualities. Some people may be happy after installing a new capacitor and never look back. That's fine for them, and I am not critical of their experience if they are happy with the end result. But this approach won't work for all applications especially where the part being replaced is already of high quality. I would also add that swapping in a brand new component like a Duelund cap that takes a very long time to break in means that it will be difficult to state with any precision how it changed the sound since you are relying on memories from weeks or months earlier.
It should also be obvious that you can't change lots of parts all at one time and then try to say what changes were due to what part. This is one of the things that frustrates me so much about Jeff Day's long blog about his Tannoy Westminsters, but that is for another day.
