By the time the clamps come out, the glue-up has already succeeded or failed. The clamps just hold things still while you find out which.
That's not cynicism. It's physics. A glue-up is four decisions that happen to occur in sequence: which adhesive, whether the surfaces are ready, what order to assemble in, and how much clamping pressure to apply. Get three of them right and the fourth can still wreck you. Get the first one wrong and nothing else matters.
Most woodworkers think the clamps are the hard part. They're not. Clamping is the last 10 percent.
The four decisions
The adhesive isn't a commodity. Yellow and white PVA, hide glue, polyurethane, epoxy — they have different open times, different cure chemistry, different strength profiles in different orientations, and different failure modes under different loads. Picking by color of the bottle is how you end up with a chair joint that fails six months in, not at the glue line but just past it, under cyclic racking.
The surface decides whether the glue bonds or sits. Adhesion is a surface phenomenon. The glue works by pulling into the wood cells at the interface. A surface that looks clean may have oxidized over 24 hours of shop exposure, or picked up silicone from a spray can you used on the table saw, or taken on enough moisture from a damp morning to change the chemistry at the joint. Freshly machined and uncontaminated — that's the standard.
The sequence is structural. Face frames, box assemblies, through-mortise frames — each type has an order of operations that keeps the assembly square while you have working time. Assemble in the wrong sequence and you're chasing alignment with clamps instead of setting it with order. Clamps cannot correct a bad sequence. They can only hold whatever result you've made.
The clamping pressure needs to be adequate, not maximum. That's a real distinction. Even, adequate pressure across the full glue surface. Not maximum force at the point of contact. Overtightened clamps crush wood fibers, squeeze out all the glue, and leave a joint that looks closed and isn't — a starved joint with a glue-free gap down the middle.
Why this step gets underestimated
Two reasons.
First: PVA is forgiving. Yellow glue has a long open time, reasonable gap-filling, and adequate strength for most furniture loads. Most glue-ups in most shops with most woodworkers produce joints that hold, even with imperfect preparation. This is genuinely true.
Second: the failures that happen with imperfect preparation don't happen in the shop. They happen five years later, under load, at a client's house. The builder never sees it. Or they see the failure, replace the joint with the same preparation habits, and conclude it was bad luck. The invisible successes and distant failures make it easy to conclude that technique doesn't matter much.
It does. But only at the edge cases that furniture regularly reaches: chair joints under repeated racking load, breadboard ends cycling with seasonal humidity, a dining table that lives in a dry-hot house in January and a humid-open-window house in July.
At those edge cases, "close enough" preparation produces joints that last five years instead of fifty.
What this arc covers
Six posts. The first four build the technical foundation.
Post 239 covers adhesives — PVA, hide glue, polyurethane, epoxy. When each one is the right choice and when it isn't. Post 240 covers surface preparation: what the surface actually needs, and what contaminates it. Post 241 covers sequence and staging — the order-of-operations problem for different assembly types. Post 242 covers clamping: pressure physics, caul geometry, reading squeeze-out as a diagnostic.
Post 243 closes the arc with a protocol card — same structure as the joint selection protocol and the finish selection protocol. One unified checklist from adhesive choice through clamp removal.
Before the next post
If you've got a glue-up coming, run the dry-fit to completion. Full assembly, no glue, every clamp set, diagonals measured. Then take it apart and stage everything in the order you'll assemble it with glue.
The dry-fit doesn't just check the fit. It runs the sequence. It finds the problems you can't correct once the open time is running.
You can't speed the clock. But you can run the sequence first.