How WizFit decides what workout to generate

Member

Location / equipment

Method

WizFit plan

Who trains

Where they train

How the brand wants to work

1. First: who is going to train

Two people training at the same gym do not necessarily get the same plan. The starting point is member context, not a template per level.

• Different goal: someone chasing hypertrophy and someone chasing recomposition do not get the same session structure.
• Different experience: a beginner needs fewer variations and more technical consolidation than an intermediate lifter.
• Different frequency: a 2-day week is not organized like a 5-day week.
• Different available time: 40 minutes and 75 minutes do not allow the same volume per session.
• Known restrictions: if the member reports a limitation, exercise selection adjusts to it.

This is what we call Hyperpersonalization at WizFit: the plan is built around the person, not around a standard routine that gets patched afterwards. WizFit does not provide medical diagnosis — it works with the information the member and the gym provide.

2. Then: where it will be executed

Good planning cannot ignore the environment. The chosen exercise has to exist and be executable at the location that member attends.

• The location's configured equipment defines what can actually be prescribed.
• Two locations of the same brand may require different plans for the same member.
• If the movement pattern stays but the machine does not exist, the decision is which valid alternative to use.
• The plan is generated taking the site's configured equipment into account, reducing the need for later manual adjustments.

WizFit does not detect equipment on its own: each location has its equipment configured in the platform, and planning works inside that set.

3. Finally: what framework it must respect

Automating planning does not mean AI can pick any exercise. The organization can define the methodological universe WizFit must work within.

• The approved exercise library defines which exercises belong to the brand's method.
• Criteria defined by the organization guide planning decisions when they are part of the WizFit configuration.
• Two brands with the same equipment can train differently, and that is respected.
• The method is not inferred: the organization configures it and WizFit works inside that framework.

Two things are worth separating: the approved library defines the universe of exercises, while operational rules and context — what Nexus represents in WizFit — guide how that universe is applied when they are configured. That is what keeps automated planning from becoming a black box: the gym sets the framework.

No single variable works alone

Each layer on its own produces an incomplete result. The value is in the intersection of all three.

• Highly personalized but hard to execute: exercises appear that the location cannot support.
• Executable but generic: every member at that location ends up training almost the same way.
• Every plan is different, but the brand loses consistency and the method stops being recognizable.
• High consistency, low relevance: rigid standardization that ignores the actual member.

WizFit combines all three layers in the same decision: individual context, location reality and organizational method.

A concrete example

The point is not to show a full routine, but how context leads to specific decisions.

Member

• Goal: hypertrophy
• Intermediate level
• 4 days per week
• 60 minutes per session

Location

• Configured equipment for that location
• Machines and racks included in that configuration
• Valid alternatives within the same equipment set

Method

• Approved exercise library
• Criteria defined by the organization
• Defined valid substitutions

Why this context leads to this plan

• Four sessions split by movement pattern, not daily full body — The declared frequency (4 days) allows volume to be distributed instead of stacked into every session.
• Per-session volume tuned to a real 60 minutes — Available time limits how many accessory exercises make sense before the session gets abandoned halfway.
• Main lift chosen among what exists at that location — The goal defines the movement pattern; the configured equipment defines the specific tool.
• Planning decisions stay within the criteria configured by the organization — When the brand has defined its own criteria, planning follows them instead of applying a generic pattern.
• Substitutions stay inside the approved universe — If an exercise needs replacing, the alternative still belongs to the gym's approved library.

The plan emerges from a combination of variables, not from a single instruction. Before it reaches the member, the gym team can review and adjust it.

Which part of WizFit takes part in each decision

This is a conceptual explanation of responsibilities, not an internal technical sequence.

• Hyperpersonalization — Individual member context: goal, experience, frequency, time and known restrictions.
• Nexus — Operational context and rules: how the gym or chain is organized and which configuration applies.
• Equipment-aware AI — Equipment reality: what can actually be executed at that location.
• Approved Exercise Library — Authorized exercise universe: which exercises belong to the brand's framework.
• Autopilot — Continuity afterwards: helps sustain update and renewal cycles for planning.
• Batch — Scale: running planning processes for groups of members more efficiently, keeping the context and rules that apply.

Staff keep oversight: WizFit prepares the decision, the organization validates and adjusts it.

The decision does not end with the first plan

WizFit should not be understood as “generating a routine once”. A plan loses value as soon as it is out of date relative to the member executing it.

That is why the decision model is applied again when context changes: frequency changes, the goal changes, the location changes, or what the member reports changes.

• Continuity: planning is sustained over time instead of being interrupted.
• Renewal: planning is updated in cycles instead of staying frozen.
• Feedback: the information and feedback available about the member can become part of the context used in future plan updates.
• Autopilot: helps sustain update and continuity cycles without relying solely on manual staff follow-up.

Why this matters at scale

A multi-location or franchise operation needs to combine three things that are usually presented as opposites: individual personalization, local reality and central consistency.

• Every member can receive something relevant to their goal and their moment.
• Every location can work with the equipment it actually has.
• The organization keeps its method as the network grows.

Scaling personalization does not mean removing the rules; it means applying the right rules to the right context.

The WizFit model and generic generation

A short comparison to place the difference in approach. Not every generation tool works the same way; this describes the most common pattern.

Generic generation

WizFit model

• Starts mainly from a request — Combines multiple layers of context
• May ignore the location's reality — Considers the configured equipment
• Broad universe of exercises — Can operate with an approved library
• Isolated personalization — Personalization inside the method
• A mainly generated output — A plan contextualized for the operation

The difference is not how much text gets generated, but how much operational context enters the decision.