When the team is overloaded, what criteria do you use to decide what to work on first?

Question

Accepted Answer

**First I make the overload visible and stop pretending everything fits.** When a team is overloaded, the failure mode is doing a little of everything and finishing nothing. My job is to force explicit prioritization and protect the team from thrash.

## The criteria I apply

- **Impact vs. effort.** I rank work by value delivered against cost. High-impact, low-effort items go first; high-effort, low-impact items get cut or deferred openly.
- **Reversibility and urgency.** Anything that is irreversible or time-boxed (a security fix, a contractual deadline, a launch dependency) jumps the queue.
- **Cost of delay.** Some work gets more expensive the longer it waits (compounding **tech debt**, a deadline penalty); some is the same next month. I prioritize by how fast the cost grows.
- **Blocking others.** Work that unblocks other people or teams has a multiplier beyond its own value.
- **Risk.** What's the cost if this is wrong or late? Protect against the expensive failure first.

## A simple framework

| Bucket | Examples | Action |
|---|---|---|
| Critical + urgent | Production incident, security hole | Do now, drop other work |
| High value, not urgent | Key roadmap feature | Schedule and protect |
| Low value, urgent | Loud but minor request | Push back or delegate |
| Low value, not urgent | Nice-to-haves | Say no, cut explicitly |

## How I run it

- Make a **single ranked list**, not five parallel priority-ones. If everything is a priority, nothing is.
- Communicate trade-offs to **stakeholders** in plain terms: "We can do A or B this sprint, not both. Which matters more?"
- Protect focus: limit work in progress, reduce context switching, and shield the team from new interrupts mid-sprint.

## Pitfalls

- Letting the loudest stakeholder set priority instead of the data.
- Saying yes to everything, which is really saying "I'll let randomness decide."
- Cutting quality silently to fit scope; that just moves the cost to production.

## Why it matters

Under overload, ruthless, transparent prioritization is what turns a stressed, thrashing team into one that ships the things that actually matter, and it keeps trust with stakeholders by being honest about trade-offs instead of quietly missing everything.

Bucket	Examples	Action
Critical + urgent	Production incident, security hole	Do now, drop other work
High value, not urgent	Key roadmap feature	Schedule and protect
Low value, urgent	Loud but minor request	Push back or delegate
Low value, not urgent	Nice-to-haves	Say no, cut explicitly