A Tuesday gives you fifteen minutes. You would like them to count. The temptation is to take a session you already know and chop the ends off it — fewer hard pieces, less rest between them, and a warm-up squeezed into whatever is left over.
That is the order that breaks the format. Short intervals work because the recovery is part of the work, and the first thing a squeezed schedule does is take the recovery away.
Plan it the other way round. Fix the work-to-recovery ratio first, let the clock decide how many pieces fit, and then use the recovery itself as the check on whether you got the ratio right.
The core idea
The recovery is the test.
The talk test is the obvious way to judge how hard a piece is. Comfortable speech is likely possible when intensity is below the ventilatory or lactate threshold, and not likely possible when it exceeds it; the test is valid, reliable, practical and inexpensive, and it holds across walking, jogging, cycling, elliptical and stair-stepper work [3] Reed & Pipe 2014.
But the same review says plainly that the talk test may not be practical for high-intensity interval training [3] Reed & Pipe 2014. Which is a problem, because the work piece of an interval session is the one part of the session where you cannot sensibly ask yourself whether you could hold a conversation.
So do not use the talk test on the work. Use it on the recovery, where it works exactly as advertised. If you can speak a sentence between two pieces, the ratio is right. If you cannot, the recovery is too short — and you are not doing short intervals, you are doing a set of medium efforts with spikes in them, which is a different session with a different cost.
That check is free, it needs no equipment, and it is the only intensity control you will need. The reason the recovery is the right place for it is physiological rather than administrative: rowing is an expensive sport. It costs about 36 kcal a minute, among the highest energy costs reported for any predominantly aerobic sport [5] Hagerman 1984, and about seventy to seventy-five per cent of the energy in a standard 2,000 m comes from aerobic metabolism, with the remainder anaerobic [5] Hagerman 1984. A session built from short pieces and long recoveries is, by construction, mostly easy aerobic rowing with a handful of hard moments. Most of its minutes are the kind of low-intensity work that dominates elite endurance training — around eighty per cent of total training time, in athletes competing at national or international level who train ten to thirteen times a week [1] Seiler 2010. That is a population with ten times the volume of a busy Tuesday, so read it as a description of where the bulk of an endurance week goes, not as a prescription.
There is a cost to cutting the recovery, and it is a measurable one. The load of a session is its perceived effort multiplied by its duration [2] Foster 2001, and the method is valid across exercise types and, usefully, independent of intensity [2] Foster 2001. So eight pieces crammed into the same fifteen minutes, with the rest squeezed to compensate, is the same fifteen minutes at a higher effort — a bigger session, not a shorter one. The time you saved came out of the load.
The how-to
Pick the ratio first and write it down. A starting point for a compressed day is 30 seconds on and 90 seconds off, and the reason it is a good starting point is arithmetic rather than tradition: one cycle takes four minutes, so the clock decides how many cycles you can afford and you stop when the time is gone. Keep the ratio identical across every piece, because a ratio that drifts is a ratio you cannot compare anything against.
Warm up inside the compressed session. You do not have the luxury of a long warm-up, so do a short one — a couple of minutes at an easy rate with a few build-up strokes — and treat the shape as a starting point rather than a finding. What matters is that the first piece is not the first time the legs have moved at that rate all day.
Run the pieces at a rate you can repeat, not a rate you can hit once. The temptation on a short session is to make each piece harder because there are fewer of them. Resist it. Rowing's cost rises steeply with speed, because drag on a shell rises with the square of boat speed while metabolic cost rises with a lower power, and oarsmen are strongest in the low-velocity part of the stroke with seventy to seventy-five per cent slow-twitch fibres [6] Secher 1993. A pace that is only sustainable for the first piece is a pace that ends the session early.
Use the recovery as the check. Speak a sentence during each recovery. If you cannot, the rest is too short for the piece you just did.
End the session while the last piece still looked like the first. This is a conservative rule rather than a measured threshold, and it is worth being honest about why. The best available evidence on how a stroke changes under fatigue comes from nine elite female rowers doing a 2,000 m race simulation: by 1,500 m the timing of leg extension and arm flexion had dissociated, trunk extension had delayed, and abdominal muscle activity had increased — though trunk muscle fatigue was not evident [10] Pollock 2012. That is a study of a six-minute race effort in elite women, and it cannot tell you when to stop a short-interval session on a Tuesday. What it does suggest is that the trunk's role in the stroke changes as an effort goes on, which is a reason to stop before the change reaches you rather than after.
Log three numbers. The rate, the split, and how hard the whole thing felt out of ten. The perceived-effort scale that makes the third number mean something is Borg's CR-10, a category-ratio scale with verbally anchored categories and a true zero, presented in the 1982 paper alongside the 6-to-20 category scale [8] Borg 1982. A true zero is what lets you multiply the effort by the minutes to get a session load [2] Foster 2001, and the method has since been checked in rowing itself, where session RPE tracked a training-impulse measure closely on the ergometer across 194 sessions and thirty professional rowers over twelve months [7] Dai 2025.
The trade-offs
You cannot police the work piece. The talk test is explicitly less practical at hard intensities [3] Reed & Pipe 2014, and it was validated on walking, jogging, cycling, elliptical and stair-stepper work rather than on rowing [3] Reed & Pipe 2014. In a recovery piece it works; in a thirty-second effort it is not a tool you should be reaching for.
You cannot set the work pace from a threshold. The usual instruction is that the work sits below your lactate threshold and the recovery well below it. A threshold is a number that comes out of a graded incremental test, and a review of threshold concepts found twenty-five different definitions in use across thirty-two studies, with the variability of the individual difference between a threshold and the maximal lactate steady state rarely reported [4] Faude 2009. You cannot steer by a number whose own uncertainty is almost never measured, and least of all on a session you are assembling in the ten minutes before you start.
Your manufacturer's intervals are not this format. Concept2 publishes interval sessions, but they live inside a ten-week half-marathon plan: three ten-minute pieces at moderate intensity in the base weeks, four eight-minute pieces at threshold pace in week two, five eight-minute pieces at threshold in week five, three twelve-minute pieces at threshold in week eight, and three five-minute pieces at half-marathon pace in the taper. The plan's own notes tell you to keep long rows slow, to fuel and hydrate on them, that consistency matters more than any single session, and — the one that matters here — to limit rest periods during interval sessions to no more than three minutes [12] Concept2 2026.
Read that rest cap next to a busy Tuesday and the difference is obvious. In a race plan the rest is a ceiling, because the goal is race-specific work at a quality you cannot sustain in a race, and you want as little rest as will preserve it. On a compressed day the trade runs the other way: you want the volume you can repeat and a recovery long enough to keep the stroke honest. Their pieces are eight to twenty minutes long and their rest is three minutes at most. Yours should be short and your rest long. Borrowing their format and calling it a short-interval session is how people end up doing eight minutes of threshold work in a lunch break.
One session will not tell you whether this is working. Time trials vary by less than five per cent between attempts and time-to-exhaustion tests by more than ten, and a performance test has to be able to detect a change of about one per cent to mean anything in a sport decided by less than that [11] Currell & Jeukendrup 2008. A compressed session on a Tuesday is not a performance test, and its split will move for reasons that have nothing to do with fitness.
The multi-week trend is not a diagnostic. It is tempting to call the trend across weeks the thing that settles the question. The review of training-load monitoring is direct: very few of these markers have strong scientific evidence supporting their use, and there is yet to be a single definitive marker [9] Halson 2014. The same review insists that the individualisation of load monitoring cannot be over-emphasised [9] Halson 2014 — which is an argument for knowing your own numbers, not for a formula.
What to do tomorrow
Three things.
Write the ratio on the ergometer's display before you start — 30 on, 90 off as a starting point — and let the number of pieces fall out of the time you have. Do not take the rest away to fit more pieces in.
Speak a sentence during every recovery. If you cannot, the recovery is too short and the next piece will be worse than the last one.
Before you step off, write the rate, the split, and the effort out of ten. Then compare this week's three numbers with last week's, and treat a difference smaller than a couple of per cent in the split as noise rather than as progress [11] Currell & Jeukendrup 2008.
And when the format starts working — when the pieces look alike and the recoveries let you talk — extend the session, not the effort. A busy day is a constraint you are managing, not a ceiling you are training against.
Fix the ratio before you fix the clock. The number of pieces is what you cut on a busy day; the recovery is what you keep — and it is also the check, because the talk test works in easy rowing and not in a hard piece. Load is perceived effort times minutes, so a squeezed recovery makes the session harder rather than shorter. Log the rate, the split and the effort out of ten, and let a week of those three numbers tell you what a single session cannot.
Key points
- Fix the ratio before the clock. The number of pieces is what you cut when the day is short; the recovery is what you keep.
- The talk test is less practical at hard intensities, so use it in the recovery: if you cannot speak between pieces, the rest is too short.
- Load is perceived effort times minutes. The same minutes with less recovery is a different session, not a shorter one.
- Concept2's interval sessions are eight- to twenty-minute pieces in a ten-week half-marathon plan, with rest capped at three minutes — race preparation, not a compressed-day format.
- A single compressed session tells you almost nothing. Detecting a real change means detecting about one per cent.
- End the session while the last piece still looks like the first, and write down the rate, the split and the effort out of ten.
Editorial & Coaching Standards
Guides in the MyNextRow Learn library are created by Concept2 athletes and coaches, synthesizing peer-reviewed sports physiology, biomechanics literature, and authoritative rowing guidelines. Every workout protocol and technical cue is tested for safety, repeatability, and PM5 monitor compatibility.
Sources and further reading
- Seiler S. Best practice for training intensity and duration distribution. Int J Sports Physiol Perform 2010;5(3):276-291— About eighty per cent of training time at low intensity, taken from competitive endurance athletes training ten to thirteen times a week.
- Foster C et al. A new approach to monitoring exercise training. J Strength Cond Res 2001;15(1):109-115— Session RPE is valid across exercise types and independent of intensity. It was measured on cycling and basketball, not on rowing.
- Reed JL, Pipe AL. The talk test for prescribing and monitoring exercise intensity. Curr Opin Cardiol 2014;29(5):475-480— Comfortable speech is likely possible below the ventilatory or lactate threshold. The review says it may not be practical for hard intervals.
- Faude O, Kindermann W, Meyer T. Lactate threshold concepts: how valid are they? Sports Med 2009;39(6):469-490— Twenty-five threshold concepts and thirty-two studies. The variability around the threshold is rarely reported.
- Hagerman FC. Applied physiology of rowing. Sports Med 1984;1(4):303-326— Rowing costs about 36 kcal a minute, and seventy to seventy-five per cent of the energy for a 2,000 m comes from aerobic metabolism.
- Secher NH. Physiological and biomechanical aspects of rowing. Sports Med 1993;15(1):24-42— Drag on a shell rises with the square of boat speed while metabolic cost rises with a lower power, and oarsmen are strong in low-velocity work.
- Dai X et al. Session-RPE validity and reliability in trained rowers. BMC Sports Sci Med Rehabil 2025;17:196— 194 sessions across 30 professional rowers over twelve months. Session RPE tracked a training-impulse measure closely on the ergometer.
- Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 1982;14(5):377-381— The CR-10 category-ratio scale: verbally anchored and with a true zero. The 1982 paper introduces it alongside the 6-to-20 category scale.
- Halson SL. Monitoring training load to understand fatigue in athletes. Sports Med 2014;44 Suppl 2:S139-147— Very few load markers have strong evidence behind them, and there is not yet a single definitive marker.
- Pollock CL et al. Trunk kinematics and EMG during a 2000 m race simulation. Scand J Med Sci Sports 2012;22(4):478-487— Nine elite female rowers in a 2,000 m race simulation. The trunk changes its timing by 1,500 m and no trunk muscle fatigue was evident.
- Currell K, Jeukendrup AE. Validity, reliability and sensitivity of performance measures. Sports Med 2008;38(4):297-316— Time trials vary by under five per cent and time-to-exhaustion tests by over ten, and a test has to detect about one per cent to mean anything.
- Concept2. Half Marathon Row Training Plan— The manufacturer own interval work: eight- to twenty-minute pieces, and a note to limit rest in interval sessions to three minutes.