Strategy guide

Every hero is five numbers from 1 to 10, and every call is a minimum for one or two of them. That's the entire mechanical layer. Nothing else on a hero's sheet is checked.

Combat

Anything that ends in a fight. Tested by the bank robbery.

Intellect

Investigations and containment problems. The missing person and the spill.

Charisma

Talking a situation down. Only the PR crisis needs it, and it needs a lot.

Mobility

Chases. The rooftop wants 8, which two heroes have.

Vigor

Staying in it. Paired with Combat or Intellect, never asked for alone.

A call with two requirements is scored on the one you clear by the least. Send Golem to something wanting Combat 7 and Mobility 6 and his Combat 9 buys you nothing — Mobility 3 is three short, so the call is three short.

1. 01Read both requirements before you look at the roster.
2. 02Find who clears the tighter one, then check they also clear the other.
3. 03Prefer margin over total. Clearing by 2 is a clean outcome; clearing by 0 is a scrape.

Some calls also flag heroes as wrong regardless of stats. Sending Golem to the chemical spill widens the contamination zone — the sheet says he could hold it, the story says he shouldn't.

An overused hero opens the next episode at -2 to every stat, and this is what ruins otherwise good runs. Prism is the only hero who clears the Episode 2 PR crisis — Charisma 9 against a bar of 8. Fatigued she's at 7, and nobody on the roster can take it.

Rotate from Episode 1

Fatigue carries forward, so the damage is done an episode before you feel it.

Watch the warning

A yellow icon marks a tired hero. Dispatching one loses No Hero Left Behind.

Protect the specialists

Sonar and Prism have exactly one thing each. Spend the generalists instead.

Multi-hero calls aren't additive — the right two buff each other. There's one pairing the game clearly favours.

Dialogue moves a hidden relationship number with each hero and with your boss, and it pays out episodes later rather than immediately. There's no manual save, so none of it is reversible.

Supportive lines compound

Consistent backing is what gets all seven heroes to +8 for the hidden achievement.

Siding costs the other side

Backing management gives you autonomy and slower hero trust. Backing the heroes inverts it.

The confidant is permanent

Whoever Robert opens up to in Episode 2 stays his confidant through Episode 8.

Tier here means “how often are they the answer”, not “how strong”. Coupé is S because nothing on his sheet drops below 6, not because anything on it is exceptional.

Generalists

Coupé and Malevola have no dead stat, so they're the fallback when your specialist is tired.

Specialists

Prism, Sonar, Golem and Invisigal each own one number — and have real gaps everywhere else.

Flambae

Combat 8 and Mobility 7 covers two lanes, but something always burns.

Sending your best hero

There's no best hero. Golem is the top pick for the bank robbery and the worst for the rooftop chase.

Saving Prism for later

The Episode 2 PR crisis is the call that needs her. Holding her back wastes the one thing she's for.

Rushing

Nothing in Dispatch is on a timer. A wrong pick is always a choice, never the clock.

Chasing every achievement in one run

Speed Runner and the exploration achievements can't coexist, and the Episode 1 finale branches are exclusive.

The success percentage the game shows you is not the real probability — AdHoc rigs it, and the developers said so at GDC. Knowing the guardrails changes how you read every call, and why the finale suddenly feels brutal.

76% auto-wins

Anything shown above ~76% simply succeeds. Once you do fail a high-odds call, three auto-wins switch back on to stop bad-luck streaks.

Low rolls floored at 15%

Any displayed chance from 1 to 14 is silently rounded up to 15 — there's no such thing as a near-zero call.

Over-stacking can auto-fail

Piling too much Combat onto a call can auto-fail it. More isn't always safer; match the requirement instead.

The finale removes all of it

In Episode 8 every guardrail is switched off, so an 85% can and does fail. That's intended difficulty, not a bug.

This is also why our simulator can disagree with the game: it shows the raw stat margin, while the game layers these hidden helpers on top. Source: AdHoc's GDC talk (reported by TheGamer, Mar 2026) and dev interviews (PCGamesN, Dec 2025).

After launch AdHoc published aggregate “Dispatch Wrapped” stats — a read on how the whole player base actually chose. Handy for sanity-checking whether your run went with the crowd or against it.

Episode 3 cut

Coupé was removed by 2.8M players versus 1.6M for Sonar.

Closest call

The Episode 4 Waterboy-vs-Phenomaman pick split almost evenly — Waterboy edged it at 53%.

Who got romanced

Invisigal ~1.9M, Blonde Blazer ~1M; 164,000 romanced no one and 66,000 failed both.

Shroud's fate

Shroud dies in 63% of playthroughs.

Official AdHoc figures, relayed by GameRant (Jan 2026).