If you’ve ever watched an experienced pilot get an airplane ready, you’ve seen something that looks almost casual: their hands move across the panel in a smooth, practiced sweep, touching switches and levers in a fixed order, and only afterward do they pick up a checklist. That’s not sloppiness. That’s the flow-then-check method, and it’s the backbone of how procedures are actually run in a well-organized cockpit.
What a flow actually is
A flow is a memorized, habitual sequence you perform by touch and sight in a fixed physical order — for example, sweeping left to right across the panel, or working around the cockpit in a consistent pattern. You do the actions from that practiced pattern, not by reading them one at a time. AOPA describes the method plainly in Flow, then check, and it’s reinforced throughout NASA’s human-factors work on cockpit procedure, Checklists and Monitoring in the Cockpit (NASA/TM-2010-216396).
The key idea is that the flow and the checklist are two different tools doing two different jobs. The flow is how you get the work done efficiently, in a natural physical order, without your head buried in paper. The checklist is how you confirm, afterward, that the work is actually done.
Organize both by phase of flight
Flows are typically organized by phase of flight — preflight, before-start, before-takeoff, climb, cruise, descent, before-landing, after-landing, shutdown — with each phase having its own flow and its own matching checklist (again, NASA/TM-2010-216396 lays this out). That symmetry is deliberate and useful: it means the checklist you build should mirror the flow you fly. Same phases, same order, so that when you finish the before-takeoff flow, the before-takeoff checklist is right there to confirm it.
This is why a checklist organized by phase of flight — rather than one long undifferentiated list — is so much easier to use. It’s shaped like the flight, and it’s shaped like your flows.
Memory items are the exception
There is one category that lives outside the flow-then-check pattern: memory items. These are a small number of immediate-action steps — think engine-fire-on-the-ground response — that must be executed from memory before any checklist is consulted, precisely because there’s no time to read. They’re pulled entirely out of the normal checklist and drilled separately.
It’s worth being careful about terminology, because two related ideas get muddled. The FAA’s DOT/FAA/AM-91/7 discusses “killer items” that may be performed as a flow when the need arises. Memory items and killer items overlap but aren’t identical:
- Killer items are the most dangerous-to-miss items on a normal checklist — fuel, mixture, flaps, gear, trim. They stay on the checklist; you just make sure they’re prominent and confirmed.
- Memory items are immediate-action emergency steps drilled out of the checklist entirely, executed from memory first, then backed up by the written procedure once you have a moment.
When you build an emergency section, mark the true memory items distinctly — a bordered box with a clear caption works well — so they read as “do this now, from memory” rather than blending into the read-and-verify steps that follow.
Why the checklist backs up the flow rather than replacing it
Here’s the safety logic. The benefit of running a flow and then a checklist is redundancy: if you miss a step during the flow — because you were interrupted, distracted, or flying an unfamiliar airplane — the subsequent checklist read is a second, independent chance to catch it. Two passes, two different mental modes, one shared goal.
NASA’s field observations also document the failure mode that undoes all of this: skipping the flow and treating the checklist as the primary action list — reading each item and doing it right then, “read-do” style, in a phase that should have been flown from a flow. That’s slower, and it collapses your two independent passes into one. Now a distraction mid-checklist can cause the exact miss the second pass was supposed to catch. The method only pays off if you keep the two passes genuinely separate.
Flying alone makes this matter more
This last part is an argument, not a cited fact — reasonable, I think, but framing rather than research. In an airline cockpit there are two pilots, and the second one is a living backup: one flies the flow, the other reads and challenges the checklist, and each catches the other’s misses. A single-pilot GA flight doesn’t have that. You are the flow and the check.
Which is why deliberately separating the two passes matters more for a solo pilot, not less. When there’s no one to catch your missed item, the discipline of “hands do the flow, then eyes confirm with the checklist” is the closest thing you have to a second crewmember. Build your checklist so that second pass is fast and unambiguous — short sections, phase-of-flight order, killer items obvious — and it will do that job.
None of this replaces training or the approved procedures for your aircraft. Flows and checklists are how you run the procedures in your POH; they aren’t a substitute for knowing them. Build the tool to match how you actually fly, then verify every item against your aircraft’s current POH/AFM.
Common questions
What's the difference between a flow and a checklist?
A flow is the memorized sequence of actions you perform by hand in a fixed physical order. The checklist is the separate, second pass you run afterward to confirm the critical items were actually done. The flow does the work; the checklist verifies it.
Are memory items the same as killer items?
No. Killer items are the most dangerous-to-miss items on a normal checklist and stay on it, just made prominent. Memory items are immediate-action emergency steps drilled out of the checklist entirely and executed from memory first, then backed up by the written procedure.