Baseline Member app · interactive prototype · synthetic data Onboarding All screens
06:47
06:30
Thursday 6 August
Quiet hours are enforced server-side
Nothing is delivered between 22:00 and 06:00 in the member’s own timezone — enforced on the server, so a phone with the wrong clock can’t cause a 03:00 push. Crisis follow-up is the only thing that overrides it.

Tap a notification to follow it into the app

Good morning
Thu 6 Aug

This morning

129 / 80 mmHg

Logged at 06:44

What’s driving your numbers Top factor
130/80 30-day average
129/80 7-day average
Streak 23 days

Last 14 days. Green means you completed the check-in.

Day 4
What the first week looks like

Your latest

136/ 88mmHg
◑ Stage 1
Building your baseline

4 of 7 readings logged. Two more mornings and your first personal insights unlock.

Why we wait
We could show you population averages on day one. We’d rather wait until we can tell you something true about you — and say plainly when we’re not sure yet.
General guidance meanwhile

Based on your intake profile, not yet on your own data.

Keep sodium under 2,300 mg a day
75–150 minutes of movement a week
No more than 2 units of alcohol a day
Before you measure
5:00

Sit quietly for five minutes first

Get it right
1 Back supported, feet flat on the floor
2 Arm resting at heart level, cuff on bare skin
3 No caffeine or exercise in the last 30 minutes
4 Don’t talk during the measurement
5 Take two readings a minute apart — we’ll average them
Enter your reading
/

Try these to see the classification change

When systolic and diastolic fall in different categories, the higher one decides — so 128/82 is Stage 1, not Elevated.
Omron M7
Connected over Bluetooth

Reading 2 of 2

128/ 79mmHg
◐ Elevated

Pulse 71 bpm

Both readings Averaged, per protocol
First · 06:44 130/81
Second · 06:45 127/78
Average 128/79
Transfer log
Service 0x1810 discovered
Characteristic 0x2A35 indicated
2 records decoded
Add context
Optional — it sharpens the model
Which arm
Position
Timing
Anything unusual?

Seek care today

186/ 118
▲ Hypertensive crisis

A reading above 180/120 needs medical attention today. Sit quietly for five minutes and measure once more. If it stays this high, contact your doctor or go to an emergency department.

Call emergency services
No AI on this path
Crisis readings bypass the model entirely. No insight, no recommendation, no dismissal into a normal flow — just the clinical instruction.

Today’s check-in

Six questions about yesterday. Most of it is already filled in from your watch.

45sTypical
23Day streak

Sleep

How long did you sleep last night?

Apple Health says 7h 36m Pre-filled
7.6

How rested do you feel?

WreckedGreat

Stress & work

How stressed were you yesterday?

CalmOverwhelmed

How demanding was work?

Stress management became a first-line strategy in the 2025 guideline. It’s here because it earns its place, not to pad the questionnaire.

Movement

Yesterday you moved a fair amount

9,412 steps · 64 active minutes Pre-filled

Any deliberate exercise?

Food & drink

Much salty or processed food yesterday?

Water

2.4 litres

Caffeinated drinks

3 cups

Alcohol & smoking

How many units did you drink?

One unit is a 340 ml beer, 125 ml wine, or a single spirit.

1
This is your strongest weekend pattern. Each drink past the first shows up in tomorrow morning’s reading.

Did you smoke or vape?

Medication

Did you take everything as prescribed?

Atorvastatin 20 mg
Evening · 21:00
Vitamin D3 1000 IU
Morning · 07:00
You’re not on blood-pressure medication. Your doctor put you in the lifestyle-first window the guideline recommends for Stage 1 with low overall risk.

Done — 24 days running

That took 41 seconds.

Something we noticed New

Your last four nights all cleared seven hours — the longest run since May. Your morning average over those days is 127, down from 136 in June.

Patterns
Systolic against diastolic Last 30 days

Each dot is one reading; where it sits on the plane is its category. Recent readings are drawn stronger.

90 days at a glance

One square per morning, coloured by category. The dark band through the middle is quarter-end.

Morning versus evening

Your evenings run about 2 mmHg higher than your mornings — normal, and not something to act on. What matters is that both are trending down.

History
179 readings
Your drivers
Ranked by effect on your morning reading
Effect size, in mmHg 95% intervals
How this is worked out
A per-user model fitted on 89 days of your readings and check-ins, with each factor lagged to when it should act. It explains 76% of the variation in your morning systolic. Nothing is shown until its confidence interval excludes zero.
Short sleep
Sleep
+9.1 mmHg systolic

+4.2 mmHg diastolic

High Confidence
53 Days
6.2 to 12.0 95% CI
When the effect lands

Shifting the measurement window backwards kills the effect. That’s what a real timing relationship looks like — and what a spurious correlation doesn’t.

What to do

Suggested actions
Evidence
What if…
Move a slider, see the projection

Projected 30-day average

130 / 80

This isn’t a lookup table
The projection runs the same fitted coefficients the Insights screen reports. Move sleep and the number moves by exactly the effect size we measured on your data.
Risk score
Raised

−9 this month

What makes it up
Clinical context

Using PREVENT-style inputs, your estimated 10-year cardiovascular risk is around 8.4%. Your doctor uses this kind of figure to decide whether medication is warranted alongside lifestyle change.

Not a diagnosis
This score is a wellness indicator built from your own tracking. It is not a medical assessment and does not replace your doctor.
Experiments
Test one change, measure the result
Running now Day 14 of 21

Two alcohol-free evenings a week

136 Before
129 So far
−7 Change

Finished
Try next
More
Monthly report
July 2026
Sipho Nkosi
52 · Member BL-4471
30-day average
130/80
Baseline at intake
135/82
Change
−5/−2
Morning average
129/80
Evening average
131/80
Variability
±5.6
Readings taken
179
Check-in completion
100%

Identified drivers

    Generated from self-measured home readings. This is a wellness summary, not a diagnosis. Discuss any changes to treatment with your doctor.

    Share with your doctor
    Who can see it
    NK Dr N. Khumalo
    GP · Randburg · added 2 Jun     Active
    LN Lindiwe Nkosi
    Emergency contact · read only     Limited
    What gets shared
    Monthly report
    All readings
    Check-in answers
    AI insights
    Their system, not ours
    Readings export as FHIR R4 Observations (LOINC 85354-9), so a practice EHR can ingest them without anyone re-typing numbers.
    Medications
    Baseline never recommends starting, stopping or changing a medication. It records what you take and whether you took it — nothing more.
    Devices & data
    Recent sync
    Privacy centre
    Every permission, revocable
    Your data stays in South Africa
    Stored in AWS Cape Town (af-south-1). POPIA §72 has no adequacy list, so keeping personal information onshore removes the cross-border transfer question entirely.
    Settings
    SN Sipho Nkosi
    52 · Johannesburg · joined 9 May 2026
    Reminders
    Morning · 06:30
    Evening · 21:00
    Quiet hours 22:00–06:00
    Appearance
    Theme

    The app defaults to dark because you measure at dawn and after dark. Light mode is here for anyone who prefers it.

    Units & language
    PressuremmHg
    Weightkg
    LanguageEnglish