Enriched breads with the liquid sweet starter

In my interview with Ian Lowe about his (stiff) sweet starter methods, he mentioned two bakers who were using a variation on it, the liquid sweet starter:

One variation on the liquid sweet starter, created and used quite successfully in a real-world bakery production, is a young sweet starter. This permutation still uses 25% sugar to total flour weight. It was first discovered and refined by Maaryasha Werdiger of Zelda Bakery in Melbourne, Australia, and used to leaven her awesome babka. This same approach is also used by the extremely talented Em Hart at Damsel Bakery in Crieff, Scotland. Em makes a wonderful bun dough, lightly enriched with butter and egg, that contains more than 50% high-extraction flour. The final product has no perceptible sourness.

This approach, using a young sweet starter, is best for doughs that contain low degrees of enrichment, particularly solid fat. I'm guessing you want to use less than 20% solid fat to total flour weight, as well as pushing egg level up, to allow greater extensibility, elasticity, and air-bubble stabilization. This approach also allows you to decrease sugar levels in the final dough—Maaryasha and Em both use less than 10% to final-dough flour weight.

One note: A young liquid sweet starter can likely not be used for days-refrigerated storage of the shaped loaves.

I'll be sharing a post soon that collates all my understanding of the liquid sweet starter, but the most important thing to understand is that it is as effective as the stiff one, but the two are not interchangeable; context matters. With high-sugar (15% and up) and high-enrichment (20% and higher) doughs, the liquid sweet starter will result in slow-to-proof breads; only the stiff starter will work here. With less-sweet, less-enriched doughs, the stiff one will produce sourness, while the liquid one will work great. (In Breaducation, I am including only two SS recipes: an LSS shokupan—with 10% sugar and 13% butter—and a SSS choreg, with 22% butter and 25% sugar. I figure anyone who uses them successfully will be able to adapt the other enriched formulas as well.)

I recently reached out to both Maaryasha and Em to get more intel on how they put the liquid sweet starter to use in their bakeries, and they were very generous with what they were willing to share.

—Andrew

Maaryasha Werdiger, Zelda Bakery

Maaryasha: I want to start by explaining that the reason I started doing this was part of the process of reducing sourness in the babka. I went through all the stages of the process and did what I could at each stage to control the acidity. While I have read lots of the science behind all this, I really can't seem to process much of it and am more led by smells, touch, and visuals. I want to add that my mother starter is fairly strong and active. 

Andrew: From Ian’s description it’s a liquid starter (100%-hydration), with 25% sugar. How much seed do you use in the liquid sweet starter?

Maaryasha: 25%. For our overnight liquid mature starter for doughs with different needs, our seed is 12.5%, the sugar is still 25% though. Our stiff sweet starter is 25% seed too, but takes much longer to mature.

Andrew: How long and at what temperature do you ferment it before you use it in a dough? Is there is a particular visual you look for for ripeness?

Maaryasha: We feed it at 26–28˚C (78–82˚F) and keep it at that temperature, and it peaks after about 4 hours. It should look similar to any ripe liquid starter: We look at the height, the distribution of bubbles—there are bubbles of various sizes all over the dough, comparable to pancake batter—we are looking for even distribution all over the dough of various size bubbles, as well as a flatness rather than a dome or indent (before shaking or using). Does that make sense? haha. It's hard to explain!  

Andrew: Ian suggested using 12.5–17.5% prefermented flour in his sweet starter method for best results. How much do you use in your babka dough?

Our babka dough has 25% starter so whatever PFF that is... [Andrew: 12.5%]

Andrew: How long is the bulk fermentation of the dough, and what is the visual for readiness for shaping or retarding in bulk? In Ian’s process, he says you shouldn’t see visible expansion before shaping begins, though I’ve found that to be problematic, especially for inexperienced bakers, because if you don’t let the dough bulk long enough, you can have problems with the final fermentation. I’ve heard from others they look for a small amount of activity to know the dough is fully active.

Maaryasha: It bulks WARM for about 4 hours. it does expand and feels more aerated, but it isn't comparable to a bread loaf, that's true.  

Andrew: Do you ferment both bulk and shaped loaves at elevated temperatures (~30˚C/86˚F) like Ian does?

Maaryasha: Absolutely. This is very important from my experience.  

Andrew: Ian suggests above that the liquid sweet starter is not amenable to long retarding, but from your Grainz Conference video, it sounds like you do retard the dough in bulk. How long do/can you hold it in bulk before shaping?

Maaryasha: From my experience, when working with sweet sourdough, the most important things are a strong starter and a proper warm bulk (28–30˚C/82–86˚F). If you do not have either of those things, your proof will take forever (sometimes it will never even get there), and your final product will be more dense, chewy, and sour. 

It is true that the young starter does not work for everything. Due to the sugar being under 10% in this particular recipe, it meant the bulk was short and we were able to then manage a young leaven and the bulk all in one day production. 

We then cut the fermentation by traying up the babka and cooling it quickly—we do not retard it in a large dough quantity. It stays in the fridge on trays for 24 hours, gets shaped and goes back in the fridge for another 15 hours, then goes into the proofer at 30˚C (86˚F) for the proof, and takes about 5–6 hours to get ready for baking. As long as the starter was good and the bulk was good, there is minimal sourness. 

This starter process does not work for our other sweet sourdough recipes and I think it's mostly about the amount of sugar, but I'm not 100% sure. Our sourdough hot cross bun works best with an overnight stiff sweet starter, probably due to high sugar in the dough with all the fruit. We trialled every starter for it and got best results with the stiff overnight in terms of strength, flavor, and speed.

Hope this helps.

Emily Hart, Damsel Bakery

Emily: Despite having worked as a microbiologist before I got into baking I'm afraid I'm not as technically-minded as Ian, so I probably won't have the right terminology, but will do my best to give you the right details. I'm a bit more of a "smells good, feels right" kind of baker.

Firstly to say that my bun dough is more like 30% high-extraction flour, not 50% [as Ian said]. I have fallen into the trap of settling for something that works, so while I do want to see what happens if I increase it, it's just not something I've got around to trialling yet.

A quick bit of background on Damsel to help set the scene for how our schedule works. We're a small, neighborhood bakery in central Scotland doing bread, pastries, and coffee for our local community. We're a two-person team (I bake, David serves and makes coffees), so the idea is that the mix/bake schedule is manageable for one person. We're also a small space with a pretty rudimentary set-up, so the doughs and schedules are also intended to be reasonably forgiving in the context of nothing being ideal (no luxury of climate control, laminators, or retarder/provers, etc., etc.).

We retard overnight and then bake in the morning (I know ambient/straight bakes are wonderful but I love serving people fresh, warm bread in the morning and I'm not prepared to work through the night, so that's a compromise the business is based on), so I use short-build starters for both bread and sweet doughs as I get the bake out of the way first, before I start mixing. 

A rough rundown of the process is that we start at 5am, I get some bread loaded in the oven, mix starters (sweet starter roughly 28˚C/82˚F and sit it somewhere warm if needed) and shape baguettes, then start baking pastries around 6am. 

I get started mixing around 9am doing bread doughs and then sweet, so the croissant and buns doughs are getting mixed around 10:30-11:30am (so the starter has a minimum of 5 hours before use). They have a DDT around 24–26C (75–78˚F), and will bulk somewhere warm (high up or on top of the bread oven—nothing controlled) with a few folds for roughly 5–6hrs before shaping. 

The croissant dough gets divided into blocks and refrigerated overnight before hand-laminating the next day; the bun dough is divided, shaped, and then set to proof overnight.

I have a proofer that's probably best referred to as a holding cabinet—it has temperature and humidity control, but is designed for short use. (If I put it on 20˚C and the lowest humidity setting, it runs roughly 25˚C and is pretty sweaty so it's not reliable to leave on overnight.) During cold weather, I'll have it on from when I've loaded (croissants from roughly 1:30pm onwards, buns around 3:30pm) to when we leave (1-3 hours). They'll sit at room temperature overnight, and then I've got the cabinet timed to come on about an hour before we arrive, to give them a little morning boost.

The sweet doughs are a blend of roller-milled white flour, sifted stoneground (85% extraction), and stoneground wholemeal flours, which I assume is why I get away with a slightly lower DDT and shorter bulk times than you'd expect. Roughly 30% stoneground for the bun dough and 20% stoneground for the croissants. I'm sure Ian would find a few tweaks that would improve them immeasurably, but for our situation the dough is reliable, the buns eat well, and I suspect many of our customers don't realize that they're naturally-leavened, which tells me there's little to no sourness from their perspective as well.

Hopefully that gives you a rough idea of what we do. Any specific questions please ask away.

Andrew: Is Ian’s description of your starter correct? 100% hydration, with 25% sugar?

Emily: Yes.

Andrew: Do you use white flour exclusively in the liquid sweet starter?

Emily: Yes.

Andrew: How much seed do you use in the LSS?

Emily: In baker's percentages, my sweet starter is 50% white starter [50% seed/25% prefermented flour], 100% roller-milled white flour, 25% caster sugar, and 100% water. My white starter is a not-quite liquid white—roughly 82% starter, 80% white flour, 20% stoneground wholemeal flour, and 80% water.

What does your liquid sweet starter look like when you use it?

Nice and bubbly, but it still has texture to it—not liquidy, if that makes sense. It's markedly increased in volume and smells lactic (yoghurty) and of sweet fruit.

Andrew: Do you do anything special to prepare the levain for the sweet starter, other than use it at maturity? In Ian’s method, he gives it multiple short builds to flush out LAB, but it doesn’t sound like you or Maaryasha do that for yours, which makes it a more approachable method.

Emily: No. I work with white, rye, and sweet starters for my bread and sweet doughs. The sweet starter is mixed in the morning using the white starter as seed. My white and rye starters are maintained at room temperature and fed twice-daily (roughly 12 hours apart, give or take). 

The evening feed is small but at the same ratios, in order to have enough for 82% seed for the morning white starter, 50% seed for the sweet starter, and 90% seed for the rye starter. The morning feed is larger to bulk up for mixing later that morning. So yes, nothing special with regards to the white starter used for seeding the sweet starter, just used at maturity.

Andrew: Ian suggested 12.5-17.5% prefermented flour in final doughs with his stiff sweet starter for best results. How much do you use in your doughs, and if it varies, why?

Sorry, this is where I'll struggle to answer you properly as I don't structure my formulas like Ian or look at total percentages that factor in the starter. If the prefermented flour is the amount of flour in the starter as a percentage of the total flour then my amount would be lower, around 8.7%. Only reasoning would be that I don't think about how much prefermented flour I use. I also scald/cook the stoneground wholemeal flour that I add to the bun dough, so I guess that changes the behavior of the dough slightly.

Andrew: It sounds like your bulk is on the order of 5–6 hours at warm ambient. What do you look for—if anything—when determining when to shape? I've found that judging adequate bulk proof is challenging with sweet starter methods. Do you see much activity in it at the time of shaping?

Yeah, bulk is anywhere from 4–6 hours. Shorter than what I'd do for an all-white-flour dough as I feel things move a bit quicker in the final proof given the high-extraction/wholemeal component of the doughs.

The dough volume doesn't change significantly, I'm just looking for the dough to feel light and like it's got a bit of life. You do see activity—when you tip the dough out to start scaling there's air to push out. I also do a quick preshape, so when you get back to final shape the ball volume has increased and you need to de-gas them a bit. (This is especially noticeable on my sugar bun dough where I've folded through candied peel and orange blossom water so the dough is slightly wetter.)

Andrew: Maaryasha retards her bulk dough (in small portions) and her shaped loaves, but it doesn’t sound like you do at any stage. Is your dough firm enough to shape at ambient?

No, I don't retard the dough at any stage. We did trial this once out of curiosity but the results were as expected—the buns struggled with activity on the final proof so they had poor volume, and didn't taste great. It's definitely a dough that's designed for our circumstance and workflow. Perfectly firm and fine to shape at ambient.

The croissant dough gets divided into blocks and retarded overnight before laminating, but the percentage of sifted stoneground and stoneground wholemeal flour in this dough is lower. I do keep the croissant scraps in the fridge and bake them off in tins at the end of the week as a crude brioche loaf. There can be a little bit of sourness to these, but they're used for bostock so it's not really noticeable once you've covered the slices in sugar syrup, frangipane, and poached fruit.

Andrew: What is the fat, egg, and sugar % in your doughs (roughly)?

For the bun dough I use 25% butter, 28% whole egg, and 8.2% light muscovado sugar. (These are bakers percentages for the final dough, I just factor in the starter as "starter" rather than including its individual components, so the sugar might be slightly higher.)

The hydration is mostly whole milk (30.3%), but I also use water (22%) to cook the stoneground wholemeal flour that's added to the dough. Flour is a blend of white (69.5%), sifted stoneground (23.2%), and stoneground wholemeal (7.3%). I use the same starter for our croissant and hot cross bun doughs as well, so it also works well in other contexts.

Hopefully that answers your questions. Thanks for getting me to think about what we do. It's very easy to get caught up in your own little world, so it's both refreshing and nice to talk about it.