ManageTime

Article

Managing competing priorities across several projects and managers

To manage competing priorities, decide the split between projects once a week and write it down, rather than re-deciding it every time someone asks. Left alone, your attention flows to whichever project sits furthest from finished, or to whichever manager will notice first. Neither rule tracks what the work is worth. Both of those defaults have been measured, and both have a known failure mode. This page covers what your allocation rule actually is, what rewrites it, and how to set a split you can defend when two managers both want their work first.

How do you decide which project to work on first?

Monday, four projects live, none of them finished last week. One is nearly done and needs a final review. One has barely started and is due in three weeks. Two sit somewhere in the middle, and each has a different manager attached. You open the laptop, start somewhere, and the somewhere feels like a decision you made.

It usually is not. Most people carry a default allocation rule and never state it, and in the lab that rule turns out to be fairly consistent. Schmidt and DeShon published a study in the Journal of Applied Psychology in 2007 on how people pursue two goals at once over time. What predicted where the next stretch of time went was the gap: the further a goal sat from where it needed to be, the more time it got next.

Distance pulls. That is the whole rule.

On a good week the rule is defensible. Feeding whichever project is furthest behind is how a set of deadlines all get met at once, and it is why the nearly-finished piece can wait, since it does not need much from you. The trouble starts when distance and worth point in different directions. A project that has barely started is furthest from done whether it matters or not. So a low-value piece of work with a long way to go can quietly outrank the one thing on your list that anybody will still be reading next year.

The first move is not to work harder on any of them. It is to say out loud what your split currently is, in hours per week per project, and ask whether you would defend those numbers if a manager wanted them justified. Most people cannot answer, which is the finding underneath the finding: an unstated rule is still a rule, and it runs whether or not it fits the quarter you are in. The methods on the prioritization pillar are all ways of replacing it with one you picked on purpose.

What do you do when two managers both want their work first?

Two requests, an hour apart, both described as blocking, sent by two people who do not talk to each other. Both cannot be first. Whatever you decide, one of them finds out on Thursday. Most advice for this moment is about the conversation: escalate it, ask the two of them to agree a sequence between themselves, get the ranking set above your head. That advice is right, and on the day it is often not available. So it is worth knowing what your own ranking does while you wait for an answer.

It bends toward consequence. In the same 2007 study, when both goals paid the same, time went to whichever sat furthest from done. When only one of the two carried an incentive, time allocation was largely decided by progress toward the rewarded goal. And when the two differed in framing, one phrased as something to reach and one as something to avoid failing, progress on the avoidance-framed goal predicted where the time went more strongly than progress on the other did.

Consequence beats payoff. Avoiding a bad outcome pulls harder on you than reaching a good one does.

Ballard and colleagues found that asymmetry moving over time. Their 2016 paper in the Journal of Applied Psychology tested a formal model of multiple-goal pursuit, and participants carrying one approach goal and one avoidance goal shifted priority from the first to the second as the clock ran down. Among people pursuing two approach goals the split depended on their sensitivity to time: some fed the goal with the larger gap, others the one closest to done. Raise the risk attached and every one of those effects weakened.

In practice that is why the manager who will be annoyed usually wins. You can see it coming on Monday rather than discovering it on Thursday. The project that pays off in a quarter is something you are reaching for; the report that earns you a pointed email is something you are avoiding failing, and that is the one that takes your afternoon. Writing both requests down before you answer is the cheapest counterweight available, and the Eisenhower Matrix is built to force the urgency question and the importance question apart at exactly that moment.

How do you split your time across several ongoing projects?

Three projects, and a week with maybe thirty usable hours in it once your meetings are out. The split that survives is the one you wrote before Monday, because a split decided mid-week gets decided by whoever asked most recently. Give each project a number of hours and a place on the calendar, and treat those numbers as a claim you can turn out to be wrong about, not as a target you have to hit.

Time Blocking covers the calendar half: each project gets named blocks, so your allocation is visible before anyone asks for it and any change costs you a visible trade. Personal Kanban covers the other half, and it fits better when your problem is count rather than hours. Its central rule caps how many items sit in the doing column at once, which attacks the state you are probably in: four projects half-open, none of them close enough to finish.

Fewer live at once, not more hours.

Deciding the split in advance matters because the pressured version of you does not compute one. Alister and colleagues ran a series of experiments in Cognitive Psychology in 2024 in which people made linked prioritization decisions while the researchers squeezed either their deliberation time or their attention. Squeezed, they did not just go faster. They switched to a simpler rule, and the most common one leaned disproportionately on goal deadlines rather than on how difficult or how valuable each goal was. A week that compresses does not degrade your split gracefully; it replaces it.

So put the review where the pressure is not. Once a week, at a fixed hour, list your live projects and write an hour figure against each. A quarter of an hour does it. Then let the split stand through Wednesday even when it feels wrong, because the feeling on Wednesday is the load talking and the number on Monday is not. If a genuine emergency takes one of your blocks, move that block to a named slot in the same week rather than watching it evaporate.

How does your working style change the answer?

Which half of this you lose depends on how you already work. The four names below are this site's own, from the working-style self-check on the home page.

If you lean Architect, the split usually exists already, written and reasonable, and you lose it to the Thursday exception rather than to any absence of a plan. The discipline that helps there is refusing to renegotiate the week's allocation in the moment, and parking the request for your weekly review instead. Sprinters allocate in bursts, which suits one project and fails several. A burst spent on whatever is furthest from done is the default rule running at full power, and the piece that mattered gets whatever is left of Friday.

Visionaries fail from the other direction. The long project stays vivid, the two dated obligations owed to other managers do not, and the month ends in an escalation that was avoidable a fortnight earlier. Improvisers are the group for whom a written split reads like fiction, since their weeks rarely arrive as forecast. Name one, not four. Pick the single project that gets your biggest share this week, defend that one, and let the rest float, which at least makes the trade explicit rather than accidental.

There are weeks this page does not describe, and the evidence has edges worth naming. All three studies above ran in a lab, with goals the researchers handed out and no manager attached to any of them. None reports how much a real workload shifts, and this page will not invent a figure for it. If every project on your list is mandatory and already late, what is failing is capacity rather than allocation, and no split repairs a week that was overcommitted before you touched it. That one is a conversation with the people who set it.

So take the two projects you have been alternating between without finishing either. Give one of them the larger share of this week, in hours, on the calendar, and tell the manager attached to the other what it is getting now and when it gets more. Your forecast will be wrong somewhere. It will still beat the week where the split gets settled at four in the afternoon by whoever emailed last.

Frequently asked questions

How do you manage competing priorities at work?

Set the split between projects once a week, in hours, and write it down, rather than re-deciding it every time someone asks. A split decided in the moment gets decided by whoever asked you most recently. Writing it down turns each new request into a visible trade against a named project. A quarter of an hour at a fixed weekly slot is enough to produce it.

What do you do when two managers both want their work first?

Get the sequence set above your head if you can, and in practice expect your own ranking to favour whichever outcome carries the sharper consequence if you cannot. In a 2007 Journal of Applied Psychology study, time allocation was largely decided by progress toward the rewarded goal once an incentive attached to only one of two goals. Progress on a goal framed as something to avoid failing also predicted allocation more strongly than progress on one framed as something to reach. So the report that earns you a pointed email tends to win your afternoon.

Should you finish one project before starting another?

Not necessarily, but keeping fewer projects open at once is the more reliable half of the fix. Capping how many items sit in progress is what Personal Kanban is built around. It targets the state most people with competing priorities are in: several projects half-open, none close enough to finish. Strict sequencing is rarely available when different managers own different deadlines, so cap the count instead.

How do you split your time across several ongoing projects?

Give each project a number of hours for the week and a named block on the calendar, then hold that split until the next weekly review. The figure matters less than deciding it before the week starts, because a compressed week does not produce a considered split. In a 2024 Cognitive Psychology study, people under time pressure or cognitive load switched to a simpler rule that weighed deadlines disproportionately against a goal's difficulty and value. Move a lost block to a named slot rather than dropping it.

Sources