When three things are due Friday and there's only one of you, the failure mode isn't dramatic — it's quiet. You spread yourself across all three, make partial progress on each, and deliver three things that are 80% done, which is to say three things that are late. Competing deadlines punish the instinct to be fair to every task equally. They reward the discomfort of choosing.
You cannot do them all at once, so stop pretending
The first move is to reject the premise that everything can be handled in parallel. Attention doesn't divide cleanly; it fragments, with a switching cost each time. Three things worked simultaneously finish later, collectively, than three things worked in sequence, because sequential work gets the compounding benefit of momentum and the switching tax paid once instead of forty times. Sequencing feels slower and is faster.
There's a second, less obvious argument for sequencing: it produces something finished early. Three parallel tasks all complete at the deadline or not at all. Three sequential tasks mean one is genuinely done by Wednesday — which is one fewer thing to hold, and one thing that survives if Thursday goes wrong.
Sort by consequence, not by anxiety
The loudest deadline is rarely the most important one; it's just the one attached to the person who emails most. Re-sort by what actually happens if each slips. A deadline where a colleague waits an extra day is not the same as one where a client contract lapses or a regulatory window closes. Lay them side by side with their real consequences visible — as ranked cards on an action board — and the order that felt impossible to choose usually becomes obvious once "who's actually harmed" is staring back at you.
Find out which deadlines are real
Deadlines look identical on a calendar and are not remotely equivalent. Some are external and immovable: a filing date, a court deadline, a contractual milestone, a printer's slot, an event. Others are internal targets someone picked to create momentum, and a third category is deadlines that exist only because a date got mentioned once and calcified.
You cannot tell which is which by looking. You have to ask, and the question is more useful phrased as what happens on the other side — "what's this feeding into on Friday?" — than as "is this a real deadline?", which invites a defensive yes. Half the impossible weeks in working life are impossible only because nobody asked.
Renegotiate early, while you still can
The move people avoid until it's too late: telling someone their deadline is going to slip before it slips. A heads-up on Tuesday — "I've got three things due Friday, yours may land Monday, is that workable?" — is a manageable conversation. The same news delivered Friday at 5 is a broken promise. Most deadlines have more give than they appear to, but only the ones you renegotiate in advance; the ones you blow past silently cost you trust you don't get back.
Make the conversation easier by bringing a proposal rather than a problem. "I can get you the analysis Friday and the full deck Monday, or the whole thing Monday — which is more useful?" gives the other person a decision instead of a complaint, and it usually surfaces that they only needed one part of it by Friday anyway.
Escalate the collision, don't absorb it
When the competing deadlines come from different people, each of whom thinks they have your full attention, the collision is not actually yours to resolve — you are just the place it became visible. Absorbing it silently by working a weekend teaches everyone that the collision didn't exist.
Take it to whoever can arbitrate, with the facts and a recommendation: here is what's due, here is what I think should give, tell me if you disagree. Being able to do this without it reading as a complaint is one of the more valuable professional skills there is, and the trick is entirely in bringing the proposed answer along with the problem.
Protect the finish, sacrifice the polish
When time is genuinely short, cut quality deliberately rather than letting it get cut randomly by the clock. Decide in advance which task gets your best work and which gets a good-enough version, then defend those choices. An intentional 90%-quality deliverable that's on time beats a would-be-perfect one that's late and unfinished. The skill isn't doing everything well; it's choosing where to be excellent and where to be adequate, on purpose.
Say which is which out loud where you can. "This is a rough cut so you can start reviewing; the polished version lands Monday" sets an expectation that a silent 90% deliverable does not — and the same artefact that reads as sloppy unannounced reads as considerate when framed.
Do the postmortem so it doesn't recur
A Friday crunch is sometimes bad luck, but a recurring one is a system problem wearing a scheduling costume. If you're constantly triaging collisions, the real issue is upstream: you're accepting more than you can hold, or commitments are being made without a look at your existing load. After the fire's out, trace back how three things ended up due the same day — the pattern you find there is worth more than any single heroic week of catching up.
The usual finding is unglamorous: each commitment was made in a separate conversation where the other work wasn't visible. The fix is equally unglamorous — look at the existing list before agreeing to a date, every time. See saying no at work professionally and weekly planning that actually works.
