The final proof is the rise after shaping: the loaf sits in its basket until it is ready for the oven. How long that takes depends mostly on the kitchen temperature and on how far the dough got during bulk. This guide gives the times the bake schedule calculator plans for a proof at room temperature, says when to start checking, and explains why other pages give such different figures.
The short answer
After a bulk that reached a 50–75% rise, in a 22°C (72°F) kitchen with 20% levain, the model predicts a final proof of about 3 h. Start the poke test after about 2¼ h; the model's upper end is about 4½ h. At 24°C (75°F) it predicts about 2½ h, and at 18°C (64°F) about 4½ h. The time runs from the end of shaping, not counting the pre-shape and bench rest.
Final proof by kitchen temperature
The table gives the final proof the model predicts when the shaped loaf proofs in a room that holds the temperature in the first column, with 20% levain (levain weight as a share of the flour you add, as explained in how much starter in sourdough). The bulk column is there for reference: the proof times only hold after a bulk that ended at about a 50–75% rise, domed and jiggly, as in bulk fermentation by temperature.
| Kitchen | Bulk before it (to a 50–75% rise) | Start the poke test after about | Typical proof | Upper end |
|---|---|---|---|---|
| 18°C (64°F) | about 9 h | 3¼ h | about 4½ h | 6¾ h |
| 20°C (68°F) | about 7½ h | 2½ h | about 3¾ h | 5½ h |
| 22°C (72°F) | about 6 h | 2¼ h | about 3 h | 4½ h |
| 24°C (75°F) | about 5 h | 1¾ h | about 2½ h | 3¾ h |
| 26°C (79°F) | about 4 h | 1½ h | about 2 h | 3 h |
How the model gets there: the final proof is about half of the bulk time the model predicts at the temperature where the loaf proofs, counted with 20% levain (with less levain, the proof is taken as for 20%). Start checking at a little over a third of that bulk time, and the upper end is about three quarters of it. It is the calculator's rule of thumb for a proof after a full bulk, not a measurement of your dough. It is anchored mainly on The Perfect Loaf's guide to proofing, which gives 1½–3 hours at warm temperatures for most of its recipes and 2–4 hours in a 23–24°C (74–76°F) kitchen. Flour, starter strength, loaf size and shaping all move it, so the table gives a window and a place to start checking, not a timer setting.
The model also assumes the room holds steady. A kitchen that cools in the evening slows the proof, and a sunny window speeds it up; measure the temperature where the basket sits. In a cool kitchen the proof gets long: at 18°C (64°F) the model predicts an upper end of about 6¾ h. A room-temperature proof that long tends to slacken the dough. The calculator suggests a cold retard instead when the typical proof it plans is over 5 hours, which with 20% levain happens below about 17°C (63°F); see below.
Why other pages say 4–6 hours, or 1–2
Proof times vary a lot between recipes, partly because recipes count the time differently. Check two things before comparing a figure with this table:
- Where the clock starts. Some recipes give one time for the whole rise from mixing to the oven, bulk included. For a 22°C (72°F) kitchen with 20% levain, the model predicts about 6 h of bulk and then about 3 h of proof. A single figure that covers both is not a proof time.
- How far bulk went. The split between bulk and proof is not fixed. A recipe that ends bulk early, at a small rise, moves part of the fermentation into the proof, and its proof runs longer than this table. A bulk pushed close to doubling leaves less for the proof, so the loaf is ready sooner. The table is for the middle case: a 50–75% rise.
Temperature and starter amount shift both stages too, so a page that says "4–6 hours" and one that says "1–2 hours" can both be right for their own dough.
With 10% or 30% levain
Less levain makes bulk longer, but the calculator plans the proof as for 20%. At 22°C (72°F) with 10% levain, the model predicts about 8 h of bulk and then a proof of about 3 h, the same as with 20%. The reasoning: bulk ends at a rise, not at a time, so the dough goes into the basket equally ripe either way, and the proof depends far less on the starting levain than bulk does. That is an assumption, not a measurement.
More levain shortens bulk, and the model shortens the proof with it. At 22°C (72°F) with 30% levain it predicts about 5¼ h of bulk and a proof of about 2½ h, with the first check after about 1¾ h. The model's levain term is fitted to 5–20% levain and matches that data best at 10–20%, so outside 10–20%, as with the 30% figures here, its times are an extrapolation. Above about 30% the model likely underestimates the proof: the calculator plans it from the shorter bulk, while recipes with that much levain can proof longer. Keep checking with the poke test rather than baking on time.
The poke test decides
Start at the time in the table, earlier if bulk went to the top of the 50–75% range or the dough feels warm. Flour a fingertip and press it about 1 cm into the dough, then watch the dent:
- It springs back quickly and completely: under-proofed. Wait, and test again a little later.
- It springs back slowly and only partly, leaving a shallow dent: ready. Bake.
- It stays as a hole, or the dough sighs and sinks: over-proofed. Bake it right away; waiting makes it worse.
Look at the loaf as well as the dent. A ready loaf looks noticeably bigger and softer than when it went into the basket, but not doubled: puffy rather than tight, with a surface that feels airy under a light touch. The poke test is the best guide you have, but it is not exact; sourdough troubleshooting covers where it gets confusing and what an under- or over-proofed loaf looks like after the bake.
Turn the oven on about an hour before the start-checking time, so a 60-minute preheat (common practice; see Dutch oven vs. baking stone) is done when the loaf might be ready. If your oven only bakes with the fan, see baking sourdough in a fan or convection oven. If the loaf is ready before the oven is hot, put the basket in the fridge until it is. The cold slows it while you wait.
When a room-temperature proof does not fit the day
If the table puts the proof at a bad time, or your kitchen is cool enough that the model's typical proof runs past 5 hours, retard the shaped loaf in the fridge instead. That is a different decision with its own timing, and the calculator's model does not cover the fridge. Cold retard vs. same-day explains when to retard, for how long, and how to tell a cold loaf is ready, since the poke test reads cold dough as under-proofed. For a warm kitchen, see summer sourdough.
Frequently asked questions
Can I proof sourdough for too long at room temperature?
Yes. Once the dent no longer springs back, the dough has used up its strength and gas, and it spreads in the oven. Start checking at the time in the table and bake when the poke test says so, even if that is earlier than the typical time.
Does the bench rest count as part of the final proof?
Not in this table. The times start when the loaf is shaped and in the basket. The pre-shape and bench rest come before it, and the calculator plans them separately.
Why is my proof much faster than the table?
One likely reason is that bulk went further than a 50–75% rise; another is that the room or the dough was warmer than you assumed. Next time, end bulk a little earlier or measure the temperature where the dough sits.
Plan the proof with your numbers
Put your kitchen temperature and levain amount into the bake schedule calculator and set the cold retard to zero. It lays out bulk, shaping and a room-temperature proof with the time to start the poke test, so you can see when the loaf will be ready for the oven before you mix.