This usually comes down to 2 questions:

• How many trees are you testing at a time?
  
  
• Are you testing the whole tree, or just part of it?
  
  
  

If you’re testing an existing tree for problems, before starting your IA redesign, the answer here is simple – you’re testing just the one tree.

If you’re revising the IA for a site, and you haven’t done a baseline test yet, it’s a good idea to test the “before” and “after” versions. At minimum, this means testing two trees – the existing one (to get a baseline score) and your revised tree (to see what improved and what didn’t).

As mentioned above, though, you really should be testing more than one alternative, so you can be sure your eventual new tree is as effective as possible. Typically, we’ll test 2-3 proposed trees against each other (and against the existing baseline tree), then we’ll test a “best of” hybrid of the two in a second round.

If you’re testing a small or medium-sized tree (say, < 500 items), you will normally test the whole tree – no major pruning required.

If your tree is larger (say, 500-1000) items, you have 2 options:

• Test the whole tree – Easy to prepare, but affects how many tasks you can ask each participant.
  
  
• Test a “pruned” version of the tree – Takes some effort on your part, but lets you concentrate on the parts you’re really interested in.
  
  

Finally, if your tree is very large (>1000 items), testing the whole tree may be feasible, but in most cases we recommend that you prune your tree to keep the participants’ effort from becoming onerous.

For more on pruning trees, see Chapter 6 - Preparing a tree for testing.

Next: Who will you test?