A plan is someone else's decisions written down.
That's it. That's what you're holding when you buy a set of drawings or pull one off the internet. A designer measured a space, picked a species, chose proportions, drew joints. Every dimension on that sheet reflects the lumber they had, the tools in their shop, the humidity in their region on the day they drew the thing. All of it was correct for their build. Whether it's correct for yours is a question the plan can't answer.
Most people don't think about it this way. They treat a plan like a recipe. Follow the steps, hit the numbers, get the result. And sometimes that works. But the first time it doesn't work, you need to understand why, and the answer is almost always the same: you're building from a set of decisions that were made for a different shop, different wood, and different tools than what you've got in front of you.
Here's what plans show you. Three views, typically: plan from above, front elevation, side elevation. Sections cut through the joinery. Maybe a detail view at a tricky corner. This gives you geometry. Proportions. How the parts relate to each other. That part is solid.
Here's what they don't show you.
Grain direction. An elevation shows a panel. It doesn't tell you whether that panel is flat-sawn or quartersawn, which way the growth rings run, or how much the thing is going to move after you glue it up. You make that call at the lumberyard, not at the drafting table.
Seasonal behavior. The dimensions were recorded on a specific day, in a specific region, at a specific humidity. A plan drawn in a North Carolina shop in July carries different assumptions than your shop in the Okanagan in February. The numbers aren't wrong. They just aren't yours.
Sequence. Plans show parts. They rarely show the order you build them in. Which assembly closes before which other one. Which surfaces need finishing before they're trapped inside a case. That knowledge lived in the original builder's head and it didn't make it onto the paper.
Then there's the dimensions problem. Three-quarters of an inch in a plan often means 23/32" in your shop, because that's the actual thickness of nominal 3/4" plywood. If the plan was drawn to true 3/4", every dado, rabbet, and groove needs checking against the material you actually bought. And lumber varies too. You machine 4/4 rough stock to 3/4" and get exactly that. Buy pre-surfaced 4/4 from a supplier and you might get 13/16" or 11/16", depending on when they last sharpened their planer knives.
The rule: measure your material before you cut a single mating piece. The plan tells you the relationship. This part fits into that opening. Your stock tells you the number. Reference the plan for the relationship. Measure from what's in front of you.
A plan can't know your tools. A router table operation in the drawings might need a dado stack at the table saw if you don't have one. Same result, different setup, different considerations.
It can't know your wood. A walnut plan built in cherry needs different movement allowances, different finishing strategy, different thinking at every wide panel.
It can't know your shop. Humidity, temperature, how many clamps you own. None of that is in the drawings.
None of this means plans are bad. A good plan gives you a verified geometry, a parts list, and a record of decisions a skilled designer already tested. That's valuable. But it's a starting point. The builder who cuts straight to the listed dimension without asking what it assumes is working on trust they haven't verified. The builder who reads a plan like an argument, who asks "what is this dimension based on?" and "what isn't this view showing me?" is doing the real work before the first board hits the saw.
The plan is an argument. Your job is to understand it well enough to know when you agree and when you don't.