Exercise snacks: do very short workouts actually work?
I look, on paper, like someone who does exercise snacks. Pull-ups partway through a walk with the stroller. Push-ups while the little one sleeps. A hundred kettlebell swings before anyone else is awake.
Then I went through my own training log to write this and found that almost none of it is an exercise snack. It's something else, with better evidence behind it, and the confusion between the two is worth clearing up.
The short answer
An exercise snack, in the research, is 20 to 60 seconds of all-out effort done on its own and separated from the next one by an hour or more. That specific thing has good evidence for improving cardiorespiratory fitness in inactive people. Almost everything else people call an exercise snack is a short workout, which is a different thing with its own, better evidence for a lifter. Both are worth doing. Expecting the first to do the job of the second is where people go wrong.
What the research actually tested
The headline study used stairs. Inactive adults climbed a flight as hard as they could for 30 seconds, three times per session, with more than an hour of rest between each bout, three times a week for six weeks. Total exercise time came to under five minutes per week. Peak oxygen uptake improved by around 3.5 ml/kg/min.
The result has held up. A meta-analysis of seven randomised controlled trials covering over 400 inactive participants found these short bursts significantly improved cardiorespiratory fitness in adults under 65. Wider work on brief vigorous activity reports benefits for glycaemic control, body composition and cognitive performance.
Under five minutes a week isn't a rounding error on a claim like that. It's a genuinely surprising result.
The two limits nobody repeats
The participants were inactive. That's why the relative improvement was so large. They started from a very low base, and you probably don't.
The outcome measured was cardiorespiratory fitness, not muscle or strength. Nothing in this literature says that 30-second bouts build muscle in someone who already trains, and it would be strange if it did.
So "five minutes a week transformed their fitness" is true, and it describes a specific population doing a specific protocol for a specific outcome. Almost none of that survives the trip to a social media caption.
Three different things are being called the same thing
| What it is | Duration | What the evidence says |
|---|---|---|
| Exercise snack | 20 to 60 seconds, all-out, hours apart | Improves cardiorespiratory fitness in inactive people |
| Single-exercise session | 2 to 15 minutes, one movement, several sets | No research under this name, but it's ordinary training volume |
| Short workout | 20 to 40 minutes, a few exercises | Rests on the normal resistance training literature |
The middle row is the one with no label and no studies attached to its name, and almost nobody does it. Fewer than 4% of sessions logged in RepCount are a single exercise, and most lifters have never logged one at all. Only about one lifter in twenty-five does them for even a quarter of their training.
I'm in that twenty-fifth. It isn't something I read about or picked up from anyone, it's just what a day with fifteen spare minutes turns into. The absence of a name for it is part of why: there's nothing to search for and nobody recommending it.
Being unnamed doesn't make it unproven, though. A hundred push-ups is a hundred push-ups whether you spread them over an hour or compress them into six minutes.
What my log actually shows
I expected to find snacks. I found the middle row.
More than a quarter of my sessions this year are a single exercise, with a median length of eleven minutes. In 2019 that figure was three per cent. Sixteen of my last twenty-four single-exercise sessions were under fifteen minutes. Some recent ones: a hundred push-ups across five sets in six minutes, fifty pull-ups in seven, a hundred kettlebell swings in seven.
The sessions with more than one exercise are longer than I'd have guessed too. My typical kettlebell session runs about thirty-seven minutes across three exercises, which is a workout by any reasonable definition, not a snack.
The one thing that has clearly improved is the swing. My best set is now fifty reps at 32 kg, against thirty-two reps at the same weight in 2020, and that gain was built entirely outside a gym.
I should be honest about what that number is, though. Fifty isn't a limit, it's a target, and 32 kg is simply the heaviest bell I own. Heavier would be better if I had it. What the swing is genuinely good for is putting a lot of leg volume in quickly, which is exactly the job a ten-minute session needs doing.
So which should you do?
If you're currently inactive, do the snacks. The return on five minutes a week is better than nearly anything else available, and the evidence is directly about people in your position.
If you already lift, do the short session instead. Ten minutes of one exercise, several sets, taken reasonably close to hard. It accumulates volume that a 30-second burst never will, and it fits the same gaps in a day.
Pick movements that need no setup. Pull-ups, push-ups, bodyweight squats, kettlebell swings, hangs, calf raises. Heavy compound lifts don't belong here.
Attach it to something that already happens. The kettle, the nap, the walk, the end of a meeting. Anything that relies on remembering won't survive a week.
Log it. This is the part people skip because ten minutes feels too small to count. It isn't. Short sessions are most of my training now, and if they weren't in the log I'd believe I had barely trained this year, which the log says is false.
The reason I do any of this is narrower than the research. On a day that falls apart, doing one exercise keeps me from writing the day off, and not writing days off turns out to be most of what consistency is.
RepCount is a free workout tracker for iOS and Android. Log a single exercise as its own session and it counts the same as everything else.
Frequently asked questions
What's an exercise snack?
In the research, an exercise snack is a bout of roughly 20 to 60 seconds of all-out effort, done on its own and separated from the next bout by an hour or more. The studied protocol is three 30-second maximal stair-climbing bouts, three times a week. A 10-minute set of push-ups isn't an exercise snack in this sense, even though it's often called one.
Do exercise snacks actually work?
Yes, for cardiorespiratory fitness in people who are currently inactive. Inactive adults doing three 30-second all-out stair-climbing bouts three times a week for six weeks improved peak oxygen uptake by close to 3.5 ml/kg/min, from under five minutes of total exercise per week. A meta-analysis of seven randomised trials with more than 400 inactive participants found improved cardiorespiratory fitness in adults under 65.
Can exercise snacks build muscle?
There's little evidence that 30-second bouts build meaningful muscle in someone who already lifts. The research base measures cardiorespiratory fitness, glucose control and body composition, and it recruited inactive people. If you already train, a short single-exercise session will do more for your strength than a snack will.
What's the difference between an exercise snack and a short workout?
Duration and evidence. A snack is under a minute of maximal effort with long gaps between bouts, and the research on it is about cardiorespiratory fitness in inactive people. A short workout is 10 to 40 minutes of actual training, and it rests on the ordinary resistance training literature. Both are useful, and confusing them leads people to expect muscle from something never shown to build it.
Is a 10-minute session worth logging?
Yes, because it accumulates real volume. A hundred push-ups in six minutes is still a hundred push-ups. RepCount lets you log a single exercise as its own session, so short entries build the same history as long ones and you can see whether they're progressing.