Coming 2026 · iPhone
Postural measures your posture from two photos and scores three biomechanical markers as one number you can track, your Postural Age. It then builds a 3-minute daily routine from what your own scan found. The measuring happens on your phone.
iOS 18+ · Photos stay on your device unless you export or share one yourself
You're 29. Your posture reads 34, and forward head is carrying 4 of those 5 years, so that is what today's routine goes after.
The actual problem
Chin tucks and wall angels are free, correct, and everywhere. You do them for a week, can't tell whether anything changed, and stop. That is the failure, and it is not a shortage of exercises.
Posture also moves too slowly to feel. Without a measurement, week two looks the same as week one, so you have nothing to keep showing up for. Postural fixes the measurement first and hangs the habit off it.
How a scan works
Loose clothing hides the landmarks the scan needs, and a busy background costs you accuracy. Postural walks you through the setup before the camera ever opens, because a scan you have to redo is worse than one you spent thirty seconds preparing for.
Stand with your arms at your sides. Postural watches your pose, tells you when you're turned the wrong way, waits until you hold still, then counts down and captures. You don't need anyone to hold the phone.
Each view is read across 12 frames, and every joint is taken as the median of those readings, so one blurred moment can't move your score. If your shoulders or hips drifted mid-capture, it throws the set out and asks again rather than scoring noise.
Apple's on-device Vision framework locates 13 body landmarks. Postural computes five measurements from them: four angles and one height difference. Three map to the years they add, and that total becomes your Postural Age. You see the other two as well, but they stay out of the arithmetic.
Your daily routine then comes from a library of 28 movements, stretches and strengthening work, each with a video demonstration and its creator credited. Postural picks the ones matching your weakest markers and rebuilds the routine after every scan, so it follows your posture rather than a fixed plan.
The measurement
Postural Age is your age plus the years each deviation adds. These are the thresholds the app scores against, not a summary of them. The fifth angle has no thresholds because nobody has published any, so we left those two cells empty rather than fill them with a number we made up.
| Marker | What it measures | View | Normal | Severe | Max penalty |
|---|---|---|---|---|---|
|
CVA
Craniovertebral angle
Measured here from vertical, so a larger angle means more forward head. Clinical papers measure the same landmarks from horizontal, where the number runs the other way. See the note below the table.
|
Forward head: how far your ear sits ahead of the base of your neck, seen from the side. | Side | ≤ 40° | ≥ 55° | +6 yrs |
|
FSA
Forward shoulder angle
Measured from vertical too, so the published cutoff reaches us as its complement. See the note below the table.
|
Rounded shoulders: how far the shoulder rolls forward of the base of the neck. | Side | ≤ 43.5° | ≥ 55° | +5 yrs |
|
SAG
Sagittal alignment
We measure this on every scan and score it on none of them. Nobody has published a normal range to score it against. See the note below the table.
|
Trunk lean: the angle at your hip, from shoulder through hip to ankle. | Side | — | — | Not scored |
|
SYM
Shoulder symmetry
Measured on every scan and scored on none of them since 17 August 2026, on the advice of the physiotherapist reviewing our method. See the note below the table.
|
Left-to-right shoulder height difference, normalised against your height in frame. | Front | — | — | Not scored |
|
TILT
Lateral head tilt
|
Your ear-to-ear line, measured off true horizontal from the front. | Front | ≤ 3° | ≥ 10° | +2 yrs |
| Maximum total penalty, every scored marker at full severity | +13 yrs | ||||
On the two we measure and do not score. We measure sagittal alignment on every scan and score it on none. The measurement itself holds up: it is the hip angle used by SAPO, a postural assessment tool used in research, where five blinded raters agreed on it to an ICC of 0.93 on the right side and 0.86 on the left (Ferreira 2010, 22 adults, 88 photographs). Nobody has published a normal range for it in healthy adults. Measuring something consistently is not the same as knowing what a good reading looks like, and we would rather leave the cell empty than invent a threshold to fill it. So we compute the angle, store it, and show it to you next to the three that count. It adds nothing to your Postural Age. One thing to hold against that 0.93: it came out of a lab, from markers a clinician placed by hand on bony landmarks. A phone guessing at the same joints from a photograph has the harder job.
Shoulder symmetry joined it on 17 August 2026, and a clinician made that call too. Asked whether left-to-right shoulder height should carry up to three years, the physiotherapist reviewing our method answered that he would measure and display it but keep it out of the score, "unless there's a clear clinical or empirical basis for penalizing a specific degree of asymmetry." Two things we already knew argued the same way. The paper we had been citing measures rounded shoulders as present or absent by plumb line, not height difference, so it never matched what we compute. And a lower right shoulder is reported in roughly 68% of people, which means we had been charging most of our users for the population norm. The measurement stays on your breakdown. It costs nothing.
On the craniovertebral angle. Clinical literature measures it from the horizontal, where a higher number is better and normal is roughly 50°. Postural measures from the vertical, so our numbers are the complement: app angle = 90° − clinical CVA. Our normal threshold of 40° is the same posture as a clinical CVA of 50°. That 50°/35° convention is widely used but has no single consensus value, and published cutoffs range roughly 44° to 55°. Yip 2008, which we cite, does not state a cutoff at all. The clinical angle runs from C7 to the ear, which is why we measure to the base of your neck rather than to the point of your shoulder. Measure from the shoulder instead and the conversion stops holding: the shoulder joint sits in front of C7, which shrinks the angle and hides a forward head rather than reporting it.
On the forward shoulder angle. The same convention flip applies here. The published Shoulder Angle runs off the horizontal from the same neck-to-shoulder vector we use, so app angle = 90° − clinical. Ruivo 2014 measured 275 people and put the protraction cutoff at a clinical 52°, which arrives here as 38°. We published 20° until 29 July 2026, and that was wrong in an expensive direction: 20° on our scale is a clinical 70°, about 2.2 standard deviations above Ruivo's mean of 51.4° ± 8.5. Someone with an average shoulder angle was collecting roughly 2.7 of the 5 available years for standing normally.
We then went past the published cutoff, and a clinician sent us there. Reviewing our thresholds on 17 August 2026, the physiotherapist pointed out that 52° penalises around 40% of people who have no symptoms at all, and asked us to loosen it. He named 43.5° on our scale. Against Raine and Twomey, who found a mean of 53.7° across 160 asymptomatic people aged 17 to 83, that takes the share of symptom-free adults picking up a penalty from 42.1% to 19.8%. In Ruivo's 15-to-17-year-olds it goes from 52.8% to 28.2%. We report both because he asked for the age groups separately.
So 43.5° is no longer a transform of anyone's published cutoff. It is a clinician's number, chosen against a target rate of false alarms, and we would rather say that plainly than dress it up in a citation. One caveat we owe you: Raine and Twomey publish a mean but not a standard deviation, so the adult figures above borrow Ruivo's. The adolescent figures do not. The severe cutoff has not moved and is still ours rather than anyone's published figure.
Angles come from normalised image coordinates, so they are calibrated for what a phone camera can see rather than read off a clinical goniometer. At least two of the three scored markers have to be measurable before Postural will give you a score; below that it asks for a retake instead of filling in the gaps. Neither unscored marker counts toward that two, and when a scan rests on two rather than three the app says so under your number.
The method
Postural carries its own reading list, reachable from the app without leaving it, split by the job each paper does: showing that posture matters, showing that measuring it from a photograph is reliable, or supplying the numbers we set a threshold from. Every one links out to PubMed.
The reliability half of that list is the half we quote most carefully. Karbalaeimahdi pooled 29 studies and 1,910 participants and reported inter-rater agreement of 0.96 for head tilt, 0.90 for acromion alignment and 0.89 for the craniovertebral angle, which is the evidence that a phone photograph can measure this at all. The same review says the method needs more methodological rigour before it can reliably track clinically significant change over time. That limitation lands squarely on what Postural does, so we publish it next to the numbers rather than under them. Mylonas found the strongest evidence supports classic photogrammetry for reliability, and that for validity the evidence is not conclusive. Ferreira is on the list for one angle, sagittal alignment, and for one purpose: it shows we can measure that angle consistently. It says nothing about what a normal value looks like, which is why we measure that marker and never score it.
Being able to check the method means being told where it's thin. Two independent clinical reviewers went through this scoring, which is how the forward-head and forward-shoulder thresholds got corrected, the citation list got fixed, and we stopped scoring a fifth marker once we traced its thresholds back to nothing. The normal cutoffs for those two now follow published figures. A licensed physiotherapist reviewed the scoring thresholds on 17 August 2026, including both severe cutoffs. The years weighting — how many years of Postural Age an angle is worth — is ours, and it was not part of that review. The thresholds are reviewed but not clinically validated, and Postural says so inside the app. Read your score as a trend you can follow rather than a diagnosis.
Progress
Most habit apps sell you an improvement in week one. Twenty-one minutes of mobility work cannot move your structure, so that improvement is measurement noise with a ribbon on it. Week two, when the number wobbles back, is when you quit.
Your starting number, and the four markers behind it. Everything after this is measured against today.
You see how consistent you've been and nothing more. Postural won't show you a change it doesn't believe in yet.
Four weeks is our line, not the literature's: long enough for a real change to clear the measurement noise, short enough that you're still here. Your number moved or it didn't, and you get told which.
Between scans, every change is weighed against normal scan-to-scan variation before you're told anything, and the headline number is a rolling average of your last three scans' posture penalty. A single wobbly reading never gets presented to you as a step backwards.
We should be straight about the four weeks, because plenty of apps borrow a number from a paper that doesn't say it. The best available evidence on habit formation, Singh 2024 in Healthcare, pooled the literature and landed on a median of 59 to 66 days, means of 106 to 154, and individuals anywhere from 4 to 335. Nothing there supports a 28-day rule. What it does support is what Postural already does: put the routine in the morning, and let you set your own goal.
On device
Photos of your body are the most personal thing this app touches, so the app never sends them anywhere. The only way one moves is if you export or share it yourself.
Not out yet
That's the whole list. No newsletter, no drip sequence, no "we've been thinking about you". Unsubscribe in one click if you change your mind.
One email · No spam · Unsubscribe anytime
We'll email you the day Postural is live. Nothing before that.
Questions
Your chronological age plus the years contributed by three scored postural deviations: forward head, rounded shoulders and lateral head tilt. Each adds between 0 and its own ceiling of 6, 5 and 2 years, for a maximum of +13.
We measure a fifth angle, sagittal alignment, on every scan and show it to you, but it scores nothing. We can measure it reliably. Nobody has published a normal range for it in healthy adults, and we won't invent one to fill the slot.
It is a single number designed to be trackable and worth caring about, not a clinical grade. The direction it moves is the part to watch.
Postural uses Apple's Vision framework for 2D body pose detection, reading 13 landmarks and taking the median across 12 frames per view to damp out jitter. It cannot see depth, rotation, or muscle tone, because it works from a photograph rather than a 3D scan.
Lighting, how you're framed, and what you're wearing all affect the reading, which is why the app walks you through setup and refuses to score a capture where you drifted. Treat it as a consistent trend across time, not an absolute measurement.
No. They're written to the app's own directory on your device, flagged to be excluded from iCloud backup, and never sent to a server. Saving a timelapse or sharing a comparison card puts that image in your Photos library, which is the one path off the device and one you choose. An automated test verifies that exclusion on every build, so it can't be dropped by accident in a later release. Deleting the app deletes them.
Your scores sync to your account so your history survives a reinstall. Your photos stay on the phone.
Unlimited scans, your Postural Age, the reveal screen and the coach note on what changed are free, and stay free. That includes the Day-28 update itself: both photos, your before and after Postural Age, and the caption on what moved. When a scan puts years on your score, your home screen names the marker adding the most of them, at no charge. You can browse the exercise library and play any single movement from it without paying.
Postural Pro adds the personalised daily routine built from your own weakest markers, your scan history with the side-by-side comparison of any two scans in it, the full five-marker breakdown, and sharing the Day-28 card or the timelapse. It is $69.99 a year with a 14-day free trial, or $7.99 a week. The trial applies to the annual plan and converts to a paid subscription unless you cancel at least 24 hours before it ends.
Adults who spend the day at a desk and want to carry themselves better, whether the reason is how you look, how you feel, or catching yourself hunched over a laptop each afternoon.
It isn't for minors, and it isn't for clinical use. If you have pain, injury or dysfunction, see a physiotherapist; Postural doesn't replace one.
No. It's a consumer wellness app. Postural Age comes from geometric measurements taken by a phone camera and doesn't account for pain, strength, hypermobility or how your particular body is built. It does not diagnose, treat or prevent anything. For a clinical assessment, see a qualified provider.
Postural is in active development for iPhone, targeting late 2026 on the App Store. iOS 18 or later. Join the waitlist above and you'll hear on launch day.
Get the one email we'll send, the day Postural is live.
Join the waitlistiOS 18+ · Late 2026 · Photos stay on your device unless you export or share one yourself