1 · Welcome and account
Promise, expected effort, and familiar account actions. No tour of the full product.
Remember what works.
Set up your starting target, then go to Today. About 3 minutes.
Low-fidelity product planning artifact · 25 July 2026
A neutral storyboard for establishing a trustworthy starting point, acknowledging setup, and landing in Today ready to track. Grayscale means structure is under review—not identity, polish, or final copy.
Account creation and body inputs are kept compact. Sensitive fields explain their purpose at the moment they are requested; units stay visible and editable.
Promise, expected effort, and familiar account actions. No tour of the full product.
Set up your starting target, then go to Today. About 3 minutes.
Collect only the inputs needed for eligibility and the energy estimate.
These details help check whether an automated target is appropriate and estimate energy needs.
Decimal separator follows locale. Validate range and precision after entry.
Optional. You choose what to share and confirm every available value.
A distinct, contextual step explains the requested categories before the iOS system permission.
This is optional. We request read access only after you continue, and only the data you allow can be used.
Available values are attributed, editable, and mixed safely with required manual inputs.
Apple Health filled two fields. Review them and enter anything that is missing.
The safety questions themselves require clinical copy review. These screens focus on the resulting product behaviour: eligible users may review an estimate; automation-ineligible users never see an automated target—not even a neutral maintenance estimate.
Continue toward an estimate without overstating certainty.
Your answers are within the current boundary for automated guidance. The result will still be an estimate that you review.
Goal and activity context still need to be collected before an estimate can be calculated.
Confirmed policy branch. No automated maintenance estimate appears anywhere in this path.
Based on your answers, we will not calculate or recommend a calorie target. You can still use the food log with a manual target.
No suggested number, maintenance estimate, deficit, surplus, or automatic weekly calorie change will be shown.
For personalised nutrition guidance, talk with a qualified clinician who understands your circumstances.
The chooser describes intent rather than promising an outcome. Safety status changes the target path later; it does not remove or disguise a goal.
Lose, Maintain, Gain, and Recomposition are all first-class options.
This sets the direction of your target and what progress means. You can change it later.
Every goal asks for a destination after the goal choice. Where bounds are approved, the coarse slider and exact numeric field update one value. Recomposition can explicitly defer when no credible sourced baseline exists. A destination is intent, not a promised outcome.
Landing weight sits below current confirmed weight and above the safety-limited minimum.
This helps frame progress. It does not make the calorie target more aggressive.
Illustrative value. Locale, units, target BMI, and approved safety rules determine the real range.
Maintain defaults near current weight. A materially different destination prompts a goal change.
Choose a landing point close to your current confirmed weight.
Landing weight sits above current confirmed weight without increasing the conservative starting surplus.
This sets a progress destination. Training experience still controls the starting surplus.
Illustrative value. The destination does not promise that all added weight will be muscle.
Body fat is a progress reference only and requires a comparable measurement source.
A value, source, and date stay together. Changing the method creates a new observation; it never relabels an older one.
Now choose the destination. Future comparisons use a compatible measurement method. This value does not calculate calories.
Clinical bounds must be approved before this value can be confirmed.
Each goal branches into the minimum inputs currently supported by the recommendation proposal. Maintain’s detailed adjustment rules remain explicitly unresolved.
Selected pace, with safety caps applied later.
Slower rates may better support performance and recovery. The final target still respects the safety boundary.
No extra input is required for the starting estimate.
For eligible users, the starting target is the estimated maintenance value with no deficit or surplus.
Continue to activity context. The stability band and weekly adjustment cadence are still under product and clinical review.
Training experience controls the conservative starting surplus.
This helps avoid recommending a larger surplus than your current training stage supports.
Training behaviour and explicit priority determine whether a small deficit can be offered.
Calories alone cannot credibly recommend recomposition.
Apple Health improves activity classification when enough representative history exists. It does not add exercise calories to an activity-adjusted target. A clear manual route remains equally available.
Available recent activity and workouts prefill a reviewable category, not a calorie answer.
Available Apple Health data from the last 14 days suggests an activity category. Confirm it or choose another.
Based on 10 days with usable coverage. This is a category for the estimate, not exercise calories added to your target.
Connect through the distinct permission step or use the short manual assessment.
We use activity to classify your starting estimate—not to add exercise calories twice.
Eligible users see a transparent estimate and can switch to manual. Automation-ineligible users see only the manual entry state, with no anchoring estimate or implied recommendation.
Illustrative eligible-state values. The range and evidence remain visible.
Manual entry remains available without hiding the estimate the user declined.
Your calculated estimate was 2,250 kcal/day. It will not replace the number you choose.
The safety path contains no calculated value, range, confidence, or recommendation anchor.
Use a number you have chosen yourself or agreed with a qualified clinician. We will not suggest a number in this path.
Weekly reviews may show your own validated record, but they will not calculate or propose a calorie-target change.
This beat follows target confirmation in both paths. It marks the person’s first meaningful step and provides one unconditional next action: enter the home screen.
The eligible path acknowledges the confirmed starting estimate without repeating the preceding review screen.
The same acknowledgement, with the user-entered source preserved and no implied recommendation.
This destination reference is outside onboarding. Both acknowledgement paths arrive here directly, where food logging is available as the primary home-screen action.
Setup is complete. The real home screen presents today’s state and food logging without an onboarding fork.
Your starting point is set. Track food here when you are ready.
Confirmed constraints are built into the storyboard. Open items are labelled where they affect screen count, routing, or copy; they should be resolved before medium-fidelity interaction design.
| Status | Decision | Current wireframe treatment | Why it matters |
|---|---|---|---|
| Confirmed | Four first-class goals | Lose, Maintain, Gain, Recomposition receive equal chooser weight. | Prevents Maintain from reading as a fallback. |
| Confirmed | Every goal asks for a destination | Lose, Maintain, and Gain use landing weight; Recomposition uses an optional body-fat value/source/date observation and can store an explicit deferred state. Approved sliders synchronize with direct entry; unresolved Maintain and Recomposition sliders stay disabled. | Frames intent without promising an outcome or changing calorie logic. |
| Confirmed | Manual weight entry uses direct numeric input | Screen 2 shows one decimal-keypad field with a clear kg/lb control, locale-aware formatting, and commit-time validation. Wheel and hybrid concepts were rejected and are absent from the storyboard. | Keeps a short, known value fast to enter without presenting false product choices. |
| Confirmed | Apple authentication and Health access are separate | Continue with Apple signs in only. Apple Health uses an optional contextual permission, attributed editable values, and manual fallback. | Prevents implied consent and hidden data sourcing. |
| Confirmed | Safety-sensitive path is manual-only | No automated target or neutral maintenance estimate; no weekly auto-change. | Avoids an unsafe recommendation disguised as “neutral.” |
| Confirmed | Setup acknowledgement leads directly to Today | Both target paths retain distinct acknowledgement content and expose exactly one CTA: Go to Today. Food logging is a primary home-screen action, not an onboarding decision. | Celebrates progress without repeating the target review or introducing a false log-now/log-later fork. |
| Open | Maintain adjustment model | Initial estimate is shown only for eligible users; weekly adjustment remains off. | Stability band, persistence, and cadence are not approved. |
| Open | No-training Recomposition branch | Flagged at the goal-specific question. | Changes whether the flow reroutes to Maintain or preserves the label with no prescription. |
| Open | Manual target without calorie visibility | Manual entry is shown; targetless continuation is not assumed. | Affects accessibility, safety, and completion requirements. |
| Open | Account timing and skip behaviour | Account remains first; acknowledgement proceeds directly to Today. | Determines time to value and what can be stored before identity exists. |
| Open | Imported/repeat suggestion on first use | Deferred to Today’s food-logging surface; it is not part of onboarding. | A new account cannot honestly have a likely repeated meal. |
| Open | Mid-plan goal switching | Starting a new plan repeats goal, destination, baseline, safety, and target confirmation. Editing the goal inside an active plan is not demonstrated. | Determine whether any mid-plan goal change must create a new plan. |