How should my collection be?

[Libraraptor]

Mine are extremely carefully arranged in a manner that is aesthetically pleasing but which gives the air of being aesthetically pleasing by accident and not because I have obsessively arranged them. It's a subtle art.

That sounds interesting! Are there any pictures?

[GasmaskMax]

Mine always look ugly when their not with others of their respective line. Except cheaposaurs, their always ugly so I just lump them together at the bottom of my "display case"     

[Sigmasaurus]

I normally either organize them into companies, OR I organize them by their scientific groups.

[Moodyraptor]

Mine are extremely carefully arranged in a manner that is aesthetically pleasing but which gives the air of being aesthetically pleasing by accident and not because I have obsessively arranged them. It's a subtle art.

That sounds interesting! Are there any pictures?

Well, I was going to take some, but my new dog decided everything on that shelf looked super-interesting, and I only just rescued my Invicta triceratops from his jaws.  So I've just had to move them all out of his reach, and now they're all higgledy-piggledy   

So now I recommend the "emergency puppy eviction" look to OP. 

[BlueKrono]

And this is why, as a devoted collector, I don't have any free-roaming pets.

[The Atroxious]

And this is why, as a devoted collector, I don't have any free-roaming pets.

Same. I share space with a couple of cats, but there are some rooms I don't allow them in without supervision, and only then on very rare occasions. Having my bedroom and studio off limits to the cats has saved me a headache and a half, since they can neither disturb my sleep, nor mess with my collections and work. I've had bad experiences in the past where animals destroyed something important, so this way I get to keep my sanity.

[Moodyraptor]

Well, I like my dinosaurs, but ultimately, I like my dogs better, so it's worth risking a few accidents to me 

[Doug Watson]

False. The objectively best way is by the energy content of the toy, which can be measured by combusting it in a bomb calorimeter. Once you've measured the number of kilojoules released by its burning, you can replace it and arrange it accordingly.

Does this method include figures such as the Papo brown running rex and the Battat Diplodocus? Not that I have either, but given the rarity of those figures, wouldn't it be better to weigh them and figure out what materials they're made of in roughly what proportions, and come up with an estimate of energy content?

No, too much 'roughly' in that method. Combustion is the only way, only a profligate heathen-born would question this.

Correct. That way lies epistemological anarchy.

You guys cracked me up.

To get back on topic I go by company but I am a completest so it helps me keep track, but like most have said go with whatever method floats your boat, except for combustion that is.

[Georassic]

My quirks:
I display mine on shelves in a walk-in closet off my office. When I'm working, I often turn the light on and leave the door open, so they're always a glance away.

First, I pigeon-hole them into one of six scales...each scale gets one or more shelves: 1:12, 1:18, 1:25, 1:35, 1:40, and 1:55.
Then, I arrange each shelf chronologically, as well as geographically where applicable, so animals that coexisted (or closely so) are displayed together.
So for instance, all my 1:40 figures from Cretaceous Asia are in two groups--everything up to ~80MYA, and everything after. The figures between them on the shelf are Cretaceous animals that lived elsewhere in the same time span.
But I also group some figures by relation. Example: with 1:40 ceratopsians and hadrosaurs, there are many animals that seemed to overlap each other chronologically, so I display them with their relatives.

And of course, none of that addresses "tweeners," what i call figures that could be displayed at multiple scales... I have another set of quirks for them.

[Halichoeres]

My quirks:
I display mine on shelves in a walk-in closet off my office. When I'm working, I often turn the light on and leave the door open, so they're always a glance away.

First, I pigeon-hole them into one of six scales...each scale gets one or more shelves: 1:12, 1:18, 1:25, 1:35, 1:40, and 1:55.
Then, I arrange each shelf chronologically, as well as geographically where applicable, so animals that coexisted (or closely so) are displayed together.
So for instance, all my 1:40 figures from Cretaceous Asia are in two groups--everything up to ~80MYA, and everything after. The figures between them on the shelf are Cretaceous animals that lived elsewhere in the same time span.
But I also group some figures by relation. Example: with 1:40 ceratopsians and hadrosaurs, there are many animals that seemed to overlap each other chronologically, so I display them with their relatives.

And of course, none of that addresses "tweeners," what i call figures that could be displayed at multiple scales... I have another set of quirks for them.

Mine is really similar, except I've finally split them into 9 scale categories (the smallest is exclusively sauropods, and the largest is Pikaia plus some arthropods).

[Reptilia]

I'm waiting for a cabinet where I will finally display my Papos in a proper way, I should receive it in a couple of days. At the moment all my figures are displayed on a bookcase, it's been a "temporary solution" for about two years now.

[Cloud the Dinosaur King]

I forgot to update this but I have reorganized my shelf's to by company. It looks a lot nicer that way and it isn't as crowded as it was. That's how it's going to stay from now on.