How Much Weight Should I Add Each Week? A Progression Rubric

There is no universal amount of weight to add each week. Add the smallest practical increment only when the planned repetitions are complete, the agreed rep standard is still present, effort has not exceeded the target, and the recent pattern supports another increase. Otherwise, repeat the load, reduce it for the day, or review the wider dose.

That answer is intentionally conditional. A novice may earn another increase after one successful exposure inside a linear progression. A later-stage powerlifter may need several comparable exposures or a full training week before changing load. The same athlete can also progress at different rates on squat, bench press, and deadlift. Practitioner texts describe those stage and lift differences, but they do not establish one optimum kilogram or percentage increase for everyone. [1, pp. 155-159; 2, chapters 6-8]

The five checks before you add weight

This rubric turns “add weight every week” into five questions. It is a coaching decision aid, not a validated prediction model.

RPE belongs in the rubric, but it is not a magic readiness score. In a three-week study, twelve nationally qualified powerlifters selected loads close to target RPE for single sets after familiarisation. The sample was small and the protocol did not test long-term progress or novice accuracy. [3] In an eight-week study of 21 resistance-trained men, both percentage-based and RPE-based groups improved squat and bench strength, with no statistically significant between-group differences. That study included no women and no deadlift training. [4]

Add, repeat, reduce, or change volume?

The “add” decision is not a promise that the new load will succeed. It is a test with a reason. Smith's practitioner manual recommends starting conservatively, tracking repetition maximums and prior training, and adjusting the rate to the athlete's strength, experience, lift, and observed response. Its progression tables are examples, not universal prescriptions. [1, pp. 155-159]

The “change volume” row is also a boundary, not a hidden volume template. This page identifies when load alone has stopped answering the problem. Choosing the number of sets, weekly frequency, phases, or a deload belongs to personalised programme design.

The site's separate training-volume, rest-day, recovery, and deload owners answer those wider questions when they are live. This page does not borrow their job by prescribing weekly set counts, a rest-day schedule, a readiness score, or a fixed deload. It only identifies the point at which another load increase is no longer the supported next decision.

Novice linear progression is a phase, not a lifetime rule

A novice is defined here by adaptation speed: the lifter can still recover from and adapt to a simple training stress before the next similar exposure. In that phase, successful completion can justify adding a small amount at the next session. Practical Programming for Strength Training describes that short feedback loop, then moves to smaller increases, repeated loads, back-offs, and longer weekly organisation when the same process stops working. [2, chapters 6-7]

Calendar age and years in a gym do not settle the stage. A new powerlifter with years of general training may not respond like an untrained novice. A later-stage lifter may still make a quick change on one lift after a technique adjustment. Use the fuller novice, intermediate, and advanced programming guide to classify the planning horizon; use this page only for the next load decision inside it.

Later training does not mean “never add weekly.” It means the week is no longer a guaranteed unit of progress. Load might hold while repetitions become cleaner, a planned volume target is completed, or fatigue falls before a later increase. The broader progressive-overload guide owns that principle. This page owns only the narrower question of whether the evidence supports changing load now.

The same plate jump is not the same progression

Suppose the smallest common change available is 2.5 kg across the bar:

• 40 kg to 42.5 kg is a 6.25% increase.
• 60 kg to 62.5 kg is about a 4.17% increase.
• 180 kg to 182.5 kg is about a 1.39% increase.

Those values are arithmetic illustrations, not recommended weekly rates. They show why “just add 2.5 kg” is not equally small for every lift or athlete. If the available jump is too coarse, repeating the load is valid. Smaller plates can make a smaller test possible when the gym has them, but microloading does not override missed reps, excessive effort, or changing technique.

Worked examples: decisions, not promises

Example 1: a novice squat earns a test

A hypothetical novice completes 60 kg for the programmed 3 sets of 5. Every rep meets the current squat standard, and the sets are reported at RPE 7, 7, and 8 against a target of no more than 8. Inside an established linear progression, testing 62.5 kg at the next exposure is reasonable.

That does not predict 2.5 kg every session or a future maximum. If 62.5 kg exceeds the target effort or changes the rep standard, the next decision can be repeat or reduce.

The standard itself belongs to squat-technique diagnosis. This example assumes the programme has already defined it.

Example 2: a bench press should repeat

A hypothetical lifter completes 40 kg for 3 sets of 5, but the final set reaches RPE 9 against a target of 8 and the last pauses shorten. Adding 2.5 kg would be a 6.25% jump on evidence that is already at the boundary. Repeat 40 kg under the same standard, or use a smaller available increment only after the response supports it.

The pause question itself belongs to bench-press technique diagnosis. This article does not redefine a valid bench press.

Example 3: later-stage deadlift needs a volume review

A hypothetical later-stage lifter plans 4 sets of 4 at RPE 8. The first two sets stay on target, but the fourth set repeatedly loses the agreed start position and exceeds RPE 9 across two comparable weeks. Adding load is not supported. The coach may hold the load and review whether the set count, distribution, exercise choice, or recovery context needs to change.

This is not a prescription to remove one set. It is a signal to review the whole dose. Deadlift positions remain the deadlift technique owner's job; sets and weekly structure remain the personalised-programming owner's job.

Limits and stop rules

• Do not turn one easy set into a guaranteed weekly increase.
• Do not use RPE as a diagnosis of sleep, stress, pain, or recovery. It describes effort in context.
• Do not keep adding load when prescribed reps or the agreed technique standard repeatedly fail.
• Do not use this rubric to train through new or worsening pain. It is a programming guide, not medical advice.
• Do not infer that a volume or deload change has one universal size. Those decisions require the full programme and recent response.

The practical rule is simple: earn the next test; do not promise the next number. Add when stage, lift, rep quality, effort, and repeated response agree. Repeat when the evidence is uncertain. Reduce when the session has already exceeded its intended dose. Review volume or programme structure when the pattern is bigger than one load.

Sources and how they were used

1. Chad Wesley Smith, The Powerlifting Program Design Manual (Juggernaut Training Systems, 2018), pp. 155-159, “Selecting and Progressing Weights.” Practitioner guidance for conservative starting loads, stage-aware progression, lift-specific response, and example progressions; not a controlled comparison or universal weekly increment.
2. Mark Rippetoe and Andy Baker, Practical Programming for Strength Training, 3rd ed. (The Aasgaard Company, 2013), chapters 6-8. Practitioner model for novice session-to-session progression, stalls, smaller increments, back-offs, and the transition to weekly and longer planning; school-specific examples rather than evidence of one superior programme.
3. Eric R. Helms et al., “Self-rated Accuracy of Rating of Perceived Exertion-based Load Prescription in Powerlifters”, Journal of Strength and Conditioning Research 31(10), 2017, 2938-2943, especially the abstract, Table 1, and Figure 2. Twelve qualified powerlifters over three weeks; supports familiarised single-set target-RPE selection, not novice accuracy or long-term progression.
4. Eric R. Helms et al., “RPE vs. Percentage 1RM Loading in Periodized Programs Matched for Sets and Repetitions”, Frontiers in Physiology 9, 2018, article 247, especially Table 2 and pp. 8-9. Eight-week controlled study of 21 resistance-trained men; both groups improved without statistically significant between-group strength differences; no women or deadlift training.