Extraction Simulator
Dialing in espresso: by weight & ratio – decided by taste
Too sour, too bitter, or just right? For us, no clock decides that: we weigh coffee in and espresso out (for example 18 g in, 36 g out, so 1:2) and let the cup decide. Here you move the sliders and see which direction your espresso is heading.
The sliders: temperature in °C, coffee in g, ratio of coffee to espresso (1:2 means 18 g in, 36 g out), grind size as a relative scale, roast, puck screen.
The results: EstimateModel values, not measurements — they show the direction, not how your espresso will taste. The basis: a rule of thumb we disclose in full, built from experience in our courses and checked against studies. No guarantee — details right below the tool.
In more detail: sliders, model and sources
The sliders in detail: Temperature 88–96 °C in steps of 0.5 °C. Coffee dose 12–20 g in steps of 0.5 g. Ratio 1:1 to 1:3.5 in steps of 0.1 — that's coffee to finished espresso in the glass, 18 g at 1:2 gives 36 g. Grind size has four levels (very fine, fine, medium, coarse): this is a relative scale, not micrometres — “fine” here means finer than medium, not 250 µm. Roast light, medium or dark. Puck screen none, or 0.8, 1.0, 1.7 mm thick.
The results in detail: EstimateDose and yield are shown in grams — that's simply your setting multiplied by the ratio. The dissolved amount (extraction yield, EY) is shown in percent; the model moves between 14 and 26 %. This number is a calculated value from the model, not a measurement on your machine: a refractometer on your cup will show you something different. The six bars for body, sweetness, fruitiness, aroma, bitterness and acidity are indicators from 0 to 100, not measurements and not lab results. They show you the direction your espresso is moving in, not how it will taste. The scale “more sour – often balanced – more bitter” puts the dissolved amount in context: below around 18 %, espresso is often perceived as sour and sharp, above around 22 % as bitter and dry. That's a tendency, not a law — there is no espresso police.
The basis in detail: EstimateThe yield comes from a heuristic formula we disclose to you: 20 % as a starting value, plus 0.15 percentage points per °C above 93 °C, plus 2 percentage points per full step of ratio above 1:2, minus 0.5 percentage points per gram above 18 g, plus grind size (very fine +3, fine +1.5, medium 0, coarse −3). A puck screen lowers the effective temperature by 0.5 to 1.5 °C, depending on thickness. Every single one of these numbers is an experience value from our courses: estimated, not measured. This is a model, not a measured curve — it maps which direction each slider pulls in and how strongly. Across the full slider range, temperature (88 to 96 °C) moves the yield by 1.2 percentage points, grind size by 6, ratio by 5, dose by 4 percentage points — these four numbers are calculated from the formula; they come from the model, not from a machine. So temperature is the weakest of the four levers, but not zero.
What the studies say: Peer-reviewedThat ratio is the strongest lever and temperature a weak one matches measurements on espresso: Schmieder et al. 2023 (Foods 12, 2871, TU München) found only a small influence of temperature on the dissolved amount between 80 and 98 °C — clearly smaller than reported in older literature — and write that the brew ratio is the parameter you should set precisely first. Cameron et al. 2020 (Matter 2, 631–648) measured less dissolved material at the finest grind sizes than their model predicts (up to 13.1 % deviation) and explain it with clumping and uneven flow; less coffee in the basket plus coarser grounds gave more even, repeatable shots — that's what the channeling logic in the model is based on. That very fine with an even grinder and a stable machine works very well is our way in the cafés; the model doesn't capture those preconditions and calculates cautiously. That sweetness in coffee hardly comes from sugar was shown by Batali et al. 2020 (J. Sci. Food Agric. 100, 2953–2962) on filter coffee. Community measurementThere is no peer-reviewed study on the puck screen, but there is one careful individual measurement (Ribes 2021, published in the Decent documentation): with a screen, the dissolved amount rose by 2 to 3 percentage points, mainly at the edge of the puck. Our model calculates more cautiously than this measurement.
A contradiction we disclosed — and resolved in October 2026: The old version of the simulator carried the footnote that temperature “hardly changes total yield (Batali/Ristenpart 2020)” — yet the formula very much calculated with 0.15 percentage points per °C. And the source didn't support the claim: Batali, Ristenpart & Guinard 2020 (Scientific Reports 10, 16450) studied filter coffee, not espresso, and deliberately kept strength and yield constant. Their finding is: at the same yield, filter coffee hardly tastes different whether you brew at 87, 90 or 93 °C. It is not that temperature doesn't change yield. We've replaced the footnote in the tool; it now cites the espresso measurement by Schmieder et al. 2023 for yield, and Batali et al. 2020 only for what it supports: taste at a fixed yield. Also corrected: the channeling source was named “Hendon/Cameron 2020”; the correct citation is Cameron et al. 2020, Matter 2, 631–648.
A working model, not a fact: The sequence “acidity first, then sweetness and body, bitterness last” appears in almost every espresso text, including ours. What espresso measurements actually support is the sequence of individual compounds (trigonelline, chlorogenic acid, caffeine), not “acidity versus bitterness”. Useful as a working model for direction, not as a fact. More on this on our page about flow profiling.
The Extraction Simulator is loading … If nothing appears, please reload the page – or have a look at puck screens.
This is a simulator. It works with a simplified model and shows you relationships — not measurements and not an assurance of any kind. What ends up in your cup depends on your machine, your water, your beans and the way you work.
Our recommendations are based on experience from consultations, courses and the showroom. They are non-binding and provided without guarantee. We accept no liability for damage arising from their use. When in doubt, what your manufacturer writes in the instruction manual applies.
EstimateThe numbers this tool shows above — the dissolved amount in percent and the six taste bars — are calculated values from the model, not a measurement on your machine. Dose and yield in grams are simply your setting multiplied by the ratio.
How to go about it: one adjustment at a time
Too sour or sharp? Let more run into the cup (longer ratio, for example 1:2.5) or grind a little finer. Too bitter or dry? Less into the cup (shorter ratio, 1:1.5) or grind a little coarser. Only ever change one thing, weigh both sides on the scale — and taste. If you like it, you're right. You're welcome to note the time the shot runs; it's an observation, not a target: in our cafés, some shots run for over a minute, others taste fantastic after ten seconds.
Ready to bring it into the cup?
Browse our coffee grinders, coffee scales and beans from our own roastery – or discover all interactive coffee tools at a glance. How pressure and pre-infusion shape the shot is shown in the Pressure Profiling Simulator. Questions about your setup are answered by Jackson, our AI advisor, around the clock.
