Progressive overload is the practice of gradually increasing the demand you place on a client's body so it keeps adapting. Training that stays the same stops producing change, because the body only rebuilds past its current capacity when something asks it to.
It is a principle, not a method. It says the stimulus has to rise over time. It does not say the weight on the bar has to rise every week, which is where most of the confusion sits.
The variables you can progress
Load is the obvious one and the one clients fixate on. It is one of seven, and the others matter more once a client is past their first year, when adding 2.5kg a week stops being available.
| Variable | What progressing it looks like | What it costs in recovery |
|---|---|---|
| Load | Same sets and reps, heavier weight | High. Heavy work taxes joints and the nervous system, not just the muscle |
| Reps | Same weight, more reps per set | Moderate. Cheap at low reps, expensive as sets approach failure |
| Sets | More working sets per session or per week | High, and it compounds. The fastest way to overreach a client |
| Density | Same work, shorter rest periods | Moderate, but it degrades load. Fatigue, not effort, ends the set |
| Range of motion | Deeper squat, fuller stretch at the bottom | Low to moderate. Adds soreness for a few sessions, then settles |
| Tempo | Slower eccentrics, pauses at the hard position | Moderate. Cheap on the joints, brutal on muscle soreness |
| Frequency | Training a movement or muscle more often per week | Depends entirely on whether total sets went up with it |
The costs are the point. Sets and load carry most of the recovery burden, so a program that only progresses those runs a client into the ground within a couple of months. The others raise the stimulus while the recovery bill stays roughly flat, which is what keeps a client moving through a year rather than a block.
Load and reps are usually enough for a beginner. The rest of the table is what you reach for once the simple levers have run out.
Why it matters more for online clients
In a gym you see the drift. A client on the same 20kg dumbbells for four months is obvious the moment you walk past them. Online, that client logs sessions every week, reports that training is going well, and nothing in the check-in tells you the numbers have not moved since March.
That is a record-keeping problem before it is a programming one. If you cannot see what a client lifted last week next to what they lifted this week, you are prescribing from memory. Logs that keep the numbers against the exercise — Fitsly's training tools hold per-exercise history the client sees while logging and you see when you review — make "are they progressing?" a question you can answer in ten seconds.
Overload has a ceiling, and it is not in the gym
A stimulus only becomes an adaptation if the client recovers from it. Push the demand past what their sleep, food and life outside training can absorb and the extra work accumulates as fatigue instead.
The usual limiter is food. A client eating well under their total daily energy expenditure has less capacity to add muscle and less tolerance for added volume, and protein intake sets how much of the training becomes tissue — worth checking a client's macros before you add a fourth session. In a long fat-loss phase there is a second effect: as the deficit runs on, non-exercise activity falls and the same loads feel harder. That is not a client who has stopped trying.
Progression also has to be interrupted on purpose. Planned drops in volume let fatigue clear so the previous weeks' work shows up as performance — the reason to program deload weeks rather than wait for a stall.
What it is not
It is not a personal best every session. Weekly progress is a beginner phenomenon and it ends; after that, progression is measured across a block and plenty of good sessions match last week exactly.
It is not adding weight at the cost of the movement. A squat that gained 10kg and lost 15cm of depth got easier, not harder — range of motion went backwards while load went up.
And it is not open-ended. The rate slows as a client gets more trained, and your job shifts from adding weight to finding which of the other six variables still has room.
Turning this into an actual week-to-week process — what to progress first, how to write it so a client can execute it without you in the room, and what to do when the numbers stall — is covered in progressive overload for online clients.
