How to build your own training split from scratch

Everything downstream depends on this one number, so be ruthlessly honest. Not the week where everything goes perfectly. The number of days you will genuinely show up, most weeks, for the next few months. If work, family, and sleep mean three days, build for three days. A great three-day plan you follow beats a great six-day plan you abandon by Thursday.

The number of days sets the shape of the whole split. At two or three days, you want each session to cover most of the body, because you do not have enough slots to give muscles their own days. At four days you can split into halves, upper and lower. At five or six days you can group by movement or even give muscles dedicated days. The days come first, the structure follows.

Now spread the body across your days so each muscle gets trained often enough without any single session becoming a marathon. The common groupings are just different answers to that same problem.

Notice the pattern in the right-hand column. Almost every workable grouping lands each muscle at roughly twice a week. That is not an accident. A meta-analysis by Schoenfeld and colleagues found that, when total weekly volume was matched, training a muscle twice a week tended to edge out once a week for growth. A later, larger volume-matched review found the difference mostly disappears once volume is equal. The practical read is simple: aim for about twice a week because it spreads the work into manageable sessions, but do not treat frequency itself as a separate magic lever.

This is the decision that actually drives your results, and the one most people skip. Before you write a single exercise, decide how many hard sets each muscle should get across the whole week. A hard set means a working set taken close to failure, not a warm-up or a half-effort set.

Research shows a dose-response relationship: more weekly sets per muscle tends to mean more growth, with diminishing returns as the numbers climb. A reasonable place to start for most people is around ten hard sets per muscle per week, then add sets slowly over the following months if you are recovering well. Beginners can grow on less. Very advanced lifters often need more. Ten is a starting anchor, not a law.

1. Pick a weekly set target per muscle. Start near ten hard sets for the muscles you care about most. You can run less for small muscles you already train indirectly.
2. Divide that total across your training days. If chest gets ten sets and you train it twice a week, that is five hard sets per session. This is where the split becomes concrete.
3. Check the session is not bloated. If a day ends up with thirty-plus hard sets, you have crammed too much in. Spread it, drop junk sets, or add a day.

With volume decided per muscle, fill each day with exercises that deliver those sets. Two rules cover most of it.

Lead with the big compound lifts

Compound movements like squats, hinges, presses, and rows train several muscles at once and let you move real load. Put them at the start of the session when you are freshest, because that is when technique and output are best. The American College of Sports Medicine progression guidance recommends performing large-muscle, multi-joint exercises before smaller single-joint work in a session.

Fill in with isolation work

After the big lifts, add isolation movements like curls, lateral raises, or leg curls to top up volume for muscles the compounds under-serve. These are your fine-tuning tools. They cost less recovery, so they belong later in the session when a little fatigue does not matter.

A split with no progression plan is just a list of exercises you will plateau on. Progressive overload, gradually doing more over time, is what keeps forcing your body to adapt. It is the difference between a programme and a routine you spin in place on.

The simplest method for most lifts is double progression. Pick a rep range, say 8 to 12. Add reps week to week until you hit the top of the range on all your sets, then add a small amount of weight and drop back to the bottom of the range. Repeat. It is boring, and boring is exactly why it works.