Bean On Bar September

Bean On Bar

Know the bean.
Brew the story.

A coffee app for iPhone and Apple Watch — for the bag in your hand

You don’t brew a dripper. You brew a bean. A cone has no opinion — what decides your cup was decided long before the water hit it: at altitude, at the picking table, at the fermentation tank, at first crack. Bean On Bar reads that first, then builds the recipe around it.

The Bean On Bar app icon

On the App Store in September.
iPhone and Apple Watch. Free to start · $29.99 once for Pro.

A pour-over on a scale reading 2:15 and 180.6 grams, water going in from a gooseneck kettle
Illustration2:15 · 180.6 g. The brew and its measurement are the same gesture.
Free with your next bag $29.99 once — no subscription, nothing renews Exportable — the notebook is yours because we never hold it

01 — On the wrist

Your hands are wet.
The guidance is not.

Bloom, pour, drawdown — each step arrives as a tap on your wrist while both hands are busy. The kettle is in one, the cone is in the other, and a timer that lives on the phone is a timer you have to dry your hands for.

An Apple Watch on the wrist showing a brew timer at 2:08 and the instruction: pour to 250 grams, swirl
Illustration2:08. Pour to 250 g, swirl. The watch face is a real capture; the wrist it sits on is not. The instruction goes where your hands already are.
Follow along
Start the brew on the phone and the watch takes it from there, step by step. Free
Start on the wrist
Begin the brew from the watch itself, phone in your pocket or across the room. Part of Pro
Every step
Dry end of bloom, each pour, drawdown — all timed, and a swirl only where the recipe asks for one
One record
Either way it is one brew, not two — what you logged on the wrist is the log the next cup is built from
Haptic
You can feel the next step without looking at anything

Following a brew on the watch is free. Starting one there, without reaching for the phone, is part of Pro — $29.99 once. What Pro adds

02 — The claim

Two lots, the same brewer.
Not the same cup.

Process decides more than technique does. A washed lot and a thermal-shocked one want different temperatures, different grind, different contact time — and the roaster prints two different sets of notes because they are, in every way that matters, two different coffees. No amount of pour skill makes one behave like the other.

Two pour-overs brewed side by side, each on its own scale, each with its lot card
Two lots, side by side, same cone and same scale. What separates them in the glass was decided by the bean, not by the pour.

03 — Context Score

It shows the reasoning,
not just the number.

Every score is an automated read of what the label actually discloses. Rarer disclosures appear as separate markers — they add depth, and never drag a plain bag down.

95Clearly documented
Provenance
Estate named · variety · process · roast date in window
Specificity
Notes and weight, capped so length cannot outscore
Depth
Estate named · price visible
Taste signals
Cited as badges, never folded into the score
Algorithm
Coach research-derived v1

Automated reads of visible listing detail, not verdicts on quality.

A shelf card and the bag it came from, side by side, both reading Corazon de Jesus, Costa Rica; the card adds region Brunca and altitude 1800 metres, which the bag does not print
One lot, described twice. The shelf card adds a region and an altitude — Brunca, 1800 m — that the bag itself never prints. A score can only read what is in front of it, which is why the same coffee can read differently in two places.

04 — Measurement

What you meant to pour,
and what you poured.

Connect a scale and the dose and water that actually happened are recorded separately from the ones you intended. The gap between them is where the next cup improves.

measuredintended
Water mass over time, from the scale rather than from the recipe card.

05 — The loop

Green, roast, pour, cup —
and back again.

Plenty of tools log a roast, and plenty log a brew. What is unusual here is that the file moves in both directions: your roaster’s .klog comes in, a green-adapted .kpro goes back out, and what the cupping table said becomes the next roast’s edit rather than a note you meant to reread.

first crack

Roast

Bean temperature against rate of rise, from your own .klog.

measuredintended

Brew

Seventeen brewers, custom cones, and the paper you actually use.

day 9–14

Rest

The window this lot is likely to be at its best, tracked per bag.

The Journey screen: twelve beans across five origins, with dated entries reading The cup you rated highest, your first was three stars, this one five
Twelve beans, five origins. Your first was 3★ · this one 5★ is the loop closing — the same bean, brewed better, and the app noticing rather than you having to.

06 — What it will never do

Some of the product
is the refusing.

Every one of these is a decision that cost something. They are listed because a tool that asks for trust should be checkable.

No analytics in the app. No exceptions.

No analytics SDK, no advertising SDK, no tracking identifier, no account. The app declares NSPrivacyTracking = false in its privacy manifest — a declaration Apple publishes but does not audit, which is why the privacy page states what the app actually sends, so it can be checked rather than believed. Serving this website leaves seven days of connection logs at the CDN for abuse prevention, and nothing beyond that.

No invented averages

Price per gram is compared against your own saved history in your own currency. No community average, no leaderboard, no number that needs other people to exist.

No fabricated reviews

Cited reviews and competition results appear as badges with their source. The app never invents review authority it does not have.

No baseless scoring

The scoring and the coach come from published research and stated method, not from a feeling about what tastes good. Each score names the algorithm version behind it, so it can be checked rather than believed.

No mailing list

No newsletter, no launch list, and no field anywhere on this site that asks for your address.

No subscription

One payment. Charging rent on a tool with zero marginal cost is the category habit this product exists to indict.

You put the cup down and the first thing you want is for someone else to taste it. Not a description of it — the cup. The bean, the rest it had, the bloom that made the difference, and the numbers that got you there, handed over whole.

A share is a copy, not a grant. It cannot be revoked, and anyone holding the link can read it — which is the right shape for a recipe, and we would rather say so than imply a control that does not exist.

How we handle data
David’s Bean On Bar share card: Nylon Coffee Roasters Nyeri AA, Kenya, SL28 and SL34, washed, light; the note cleanest cup yet, the 20s bloom did it; a V60 recipe at 16.5 g to 250 g, 94 C, 700 to 850 microns, with a QR code
David’s Nyeri AA from Nylon Coffee Roasters — Kenyan, washed, light. a V60 method of the kind James Hoffmann popularised, David’s grinder and bloom, his note, and a code anyone can read without installing anything. Whoever opens it starts where he left off.