With others, I have decided to base the design on the principle that each piece secures the previous piece during assembly. My cabinet is one example of this:
You could chose a bolt for this purpose, which is what I did for my experimental dinner table for 2:
I chose these special hinges from Hafele:
Despite my experience in understanding and drawing technical diagrams, I still found it hard going to ensure the cavities and pre-drill holes for these were in the right position. I wanted to ensure that the cabinet doors left a 2mm gap all the way around between them and the cabinet carcass, so that meant I had to apply an offset to the dimensions shown in the Hafele diagram.
In the end, I got the milling for the hinges correct but not for the door catches. I had to screw those in manually.
One of the first pieces I drew in CAD was a queen size bed frame (based on the idea from Spore) which used an additional assembly method which worked as planned but was more difficult to actually assemble than I thought.
The main difficulty with assembling this though relates to the way the slats needed to be inserted; I needed the bedroom to be longer than the mattress plus the length of the slats because they are slid into place small edge first. I had to turn the frame towards the bedroom door and begin with the slats poking out through the door.
One assembly difficulty I hadn't thought of was how difficult it was going to be to simultaneously bend some kerfed plywood pieces and slide them onto a frame with dovetail catches. There was some serious hammering going on with these:
Assembling these meant I needed to use scrap wood as a buffer between the hammer and the part I'm hammering, and that gave me a handy new idea.
By the way, I've not shown my coffee table in my posts yet, so here it is. Note the legs - they might look round and branch-like, but they actually have 8 faces (octagonal). I might explain more about the issues with this kind of framework in a future post.