The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

Sourdough Ciabatta - Firm Starter Version

bwraith's picture

Sourdough Ciabatta - Firm Starter Version

Sourdough Ciabatta - Second Try

Sourdough Ciabatta - Firm Starter Version (loaves)Sourdough Ciabatta - Firm Starter Version (loaves)

Sourdough Ciabatta - Firm Starter Version (crumb and loaf)Sourdough Ciabatta - Firm Starter Version (crumb and loaf)

Sourdough Ciabatta - Firm Starter Version (crumb)Sourdough Ciabatta - Firm Starter Version (crumb)

Sourdough Ciabatta - Firm Starter Version (slice)Sourdough Ciabatta - Firm Starter Version (slice)

Sourdough Ciabatta - Firm Starter Version

One of the favorite family breads seems to be ciabatta, and this sourdough version is clearly preferred (wolfed down) by my kids for its flavor. I've achieved a little better crust and crumb with yeasted versions, particularly the one in Maggie Glezer's "Artisan Baking", but the sourdough flavor is hard to beat, especially with salty grilled left over meats in sandwiches. The recipe is loosely based on the BBA (Reinhart) "Poolish Ciabatta", as well as incorporating ideas from Maggie Glezer's version in "Artisan Baking".

This is a work in progress, but I like the way this one turned out - my second blog entry on this recipe. The flavor is a little mild, which may mean I need to lengthen and/or retard the fermentation, or maybe use somewhat more ripe starter, an exercise for future attempts.

Many thanks to various contributors to this site as always, and especially in this case to Zolablue, who encouraged me to pay more attention to ciabatta with some just great photos and discussions about how to achieve better holes in ciabatta through hydration, proper handling, and flour choice, all of which were used here, (again) after much feedback from my first blog entry on the topic.

This version has been changed to use a lower percentage - around 22% - of flour contributed by the starter, and the starter itself is a firm starter, instead of the usual 100% hydration starter I had been using. My theory here was to try the flavors from the firm starter, as well as lower the percentage of starter in the recipe. The higher percentage seems to cause problems with the texture of the dough before it gets a chance to rise. Zolablue and I have both had good success with a "hybrid" method where some yeast is added to the dough to compensate. However, I was curious to see if I could find a percentage that might work with a pure sourdough raise. The result was a higher rise, but I seem to have made the hydration a little high, as the holes are a little too extreme for my tastes. However that should be solved by simply lowering the hydration on my next try. Also, I ran out of KA AP flour, so I substituted some "sifted red whole wheat flour" from Heartland Mills, which is called "Golden Bufallo". They say it is a "high extraction" flour, i.e. the germ and endosperm is in the flour, but most of the bran has been sifted out. I have found it to be a very good flour for rustic breads. It gives the crumb a darker color and there is a nuttier flavor than with KA. Well, I wish I had just stuck with KA organic AP in this case, as the result was a little more rustic than I had in mind. I suppose there are those who would treasure this result and call it something on the way from white ciabatta to "ciabatta integrale".

Photos of process have been posted for this ciabatta using a firm "recipe starter". Don't worry, you can build this starter from any consistency starter you may have by just setting the hydration to 65%, letting it rise by double and refrigerating overnight. A spreadsheet is also posted showing weights in ounces or grams.

Recipe Starter:

  • 2 oz 100% hydration starter (Use whatever starter you like. The intention is a 65% hydration firm starter)
  • 4.5 oz bread flour (I used KA Bread Flour)
  • 2.5 oz water

The day before this bread was baked, I took my "BBA style barm", a 100% hydration starter fed with KA Bread Flour, out of the refrigerator. It had been fed within the last couple of days and was fresh and strong at the time it was refrigerated. I mixed the starter with KA Bread Flour and water in the amounts above and kneaded it for a couple of minutes to form a dough. I then put the dough in a container sprayed with a little oil and left it to rise by double - about 4 hours. Once it had risen by double, I refrigerated it to be used the next day in the dough. Note that it is not necessarily a good thing to let the dough rise by more than double or become overly ripe, as it may affect the consistency of the dough the next day.


  • 9 oz of recipe starter from above
  • 12.5 oz AP Flour (I used KA organic AP flour
  • 5 oz high extraction red wheat flour (like Golden Buffalo from Heartland Mills, use AP for a less rustic result)
  • 2 oz KA Rye Blend Flour
  • 16.5 oz water
  • 0.5 oz salt (14 grams)
  • 1/2 tsp diastatic malted barley flour


Mix all flours but 5 oz of AP flour, diastatic malted barley flour, and water together and mix on low speed just enough to get a well mixed batter. Let this sit 20 minutes. I'm trying to simplify the processing relative to the last version by using my mixer for a change.


Cut up the starter into small pieces and mix it in along with the salt. Run the mixer for a minute or so to get a good mix of ingredients in what should still be a thick batter consistency. Then, add in the remaining 5 oz of flour as you run the mixer on a low speed. It should turn into a very slack dough after another couple of minutes. I tried to show a picture of it hanging off the mixer dough hook, so check out the photos of the process linked up above.

Bulk Fermentation and Folding: (about 4 hours)

Make a fairly thick bed of flour on the counter about 12 inches square. Using a dough scraper, pour the dough out into the middle of the bed of flour. Allow it to rest for a few minutes. Then, fold the dough by flouring or wetting your hands, then grabbing one side of the dough and lifting and stretching it, folding it over itself like a letter. Do this for all 4 sides. Brush flour off the dough as you fold over the sides that were in contact with the bed of flour. You don't want to incorporate much flour into the dough as you fold. After folding, shape it gently back into a rectangle or square, turn it upside down and push the seams underneath. Place it in a rising bucket or other rising container of your choice. I spray mine with oil to make it easy to remove the dough. I also spray the top of the dough and then dust lightly with flour. Cover the container and allow to rise. Repeat the folds approximately every hour three more times. You just turn it back out on a light bed of flour - you need less flour once the dough is folded a couple of times. Always brush off flour as you fold it to avoid incorporating flour in the dough. If the dough seems very resistant to stretching, only fold it from two directions instead of four. You don't want the dough to get really stiff from too much folding. The amount of folding you will need will be more if you have more water and less if you have less water. Note that even an ounce can make a very big difference in the consistency of the dough. The dough should rise by double over a total of 4 hours. My dough was sitting at about 76F, so you may have to wait a little longer in a cooler room temperature environment. Stop folding if the dough gets too stiff. This means you didn't get as slack a dough as I had here, probably because of less water or maybe because of a different style of flour. No problem, just fold when it is stretchy, but let it rest and rise if not.

This is the first time I used Zolablue's suggestion to put the dough in a container as in Glezer's Artisan Baking, as opposed to doing it on the counter as in the BBA by Reinhart, and as done in my previous blog entry on sourdough ciabatta. I felt it worked well and had a few advantages. First, you can tell easily how much the dough has risen. That's a good thing because the poke test is hard to interpret with this very slack dough. Second, flour doesn't get stuck to the bottom because of the wet dough sitting on the counter in a bed of flour during the rise. I found that the crust was easier to control. Before I was having trouble with big chunks of wet flour paste getting stuck on the dough. I think that can actually look nice and be a characteristic of ciabatta crust, but it seemed to be a little too much that way with the wetter doughs I've been using to make these ciabattas. The last thing is that you can control the bulk fermentation temperature much more easily with the dough in a container that you can move to a good spot for rising.


Divide the dough into four pieces of equal size, roll them in the bed of flour to dust the cut ends, and let them rest a few minutes. To shape, take one of the four pieces, stretch it out and roll it or fold it over itself very gently. With ciabatta this amounts to a gently stretch and fold like a letter. You want to create some tension in the surface of the dough by folding it over itself that way. I sometimes have to roll it up a little more than just a letter fold to give it some tension. Then if you place the dough folds down on a couche, it will seal up the seams. Use the couche to create folds for the ciabatta and then nestle the folds between supports, such as bags of flour or whatever system you may have similar to what you might do for baguettes. I usually also roll the ends underneath a little to get some tension in the ends of the dough.

Final Proof:

Let them rise in the couche for about 1.5 hours, until they are puffy and have increased significantly in volume.

I baked two loaves at a time, so I proofed two loaves for 1.5 hours and two loaves for 2 hours. Da Crumb Bum - you may be right about these SD ciabatta recipes being on the edge of overproofing. I found that the first batch had darker, harder crust, whereas the second was a little bit on the pale side, even though I baked both for the same amount of time in an oven that I know was fully heated and at the same temperature. Also, the oven spring was much better on the first batch. I believe the steam conditions, temperature, and handling were very, very close to identical for both bakes.

Prepare to Bake:

Preheat oven to 500F (yes, you can probably do it without preheating, as mentioned elsewhere on the site, but it's not what I did this time). While that is going on, take each loaf out of the couche, gently stretch it in one direction by about double, lay it on a peel, maybe with parchment paper underneath, maybe sprinkled with corn meal or similar, and use your fingertips to flatten out and dimple the loaf. You can press down fairly firmly to feel the peel underneath. It sounds crazy, but the loaf will bounce back just fine in the oven if it is not overproofed. This step is important to avoid "separation of crust and crumb" or "one gigantic hole" instead of many holes. It also evens out the loaf so it has a nicer shape after baking.


Place loaves in the oven and lower temperature to 450F. Bake for about 15 minutes, until the internal temperature is around 209F (I'm near sea level), rotating them after about 9 minutes. You can bake them longer to get a darker, harder crust, or less to get a lighter softer crust. If you bake them in a dry oven for shorter times, you will get a softer crumb and a crispy, thin, lighter crust that is very good if you intend to use these for sandwiches. If you want a chewier, drier bread with a tastier, harder crust, then bake them more and use steam. Then it is a great bread to just dip in olive oil or use like french bread with dinner. The loaves should spring up from their "flattening" with your fingertips, such that not much evidence is left of the dimples you made with your fingers.

The oven spring on these was better than in my first version. The dough also seemed handle more easily. It seemed to stretch without tearing during the folds and didn't seem as sticky. I was happier with the lower percentage of starter in this dough. I wish I had the KA organic AP for the recipe, just for comparison, and also because I liked the soft white crumb, but this one is good for a more rustic effect. The size of these loaves was about 10in x 5in x 2.5in. This is a better oven spring than last time. They were about 3 inches high right after baking, but they shrunk back down to 2.5 inches after cooling off. That's still quite a bit better than the previous version, which was about 2 inches tall. It's probably true that the best of both worlds would be hybrid method, i.e. add about 1/2 to 3/4 tsp of instant yeast to either this recipe or my previous version and get a little bit faster rise and a softer, lighter crumb, as well as getting the sourdough flavors in there.


Let bread completely cool, if you can stand to wait.

This bread is especially good for sandwiches, sliced in half and then sliced along the "flat" direction to open up like a hamburger bun. It is great for burgers, steak sandwiches, ham, or just with olive oil and pepper.


chuppy's picture


I read that you used a flour that you may have otherwise not used if KA AP was on hand. Do you think that KA AP would have given you different results and how?


bwraith's picture


I think the Golden Bufallo is either higher protein than I thought, or maybe it's just that it is closer to whole grain than I thought. I have to go back and check the stuff on their web site. It seemed like a good idea at the time...

The crumb is a little darker and nuttier tasting than I had in mind making this bread. It tastes very good, but it seems slightly tough or chewy compared to what I had with all AP flour. If it had been the day I was making a more rustic ciabatta, I might have been thrilled with it. I like the KA organic AP (they call it Artisan Select) for this type of bread. All told, it's great tasting bread, and it did rise well. However, I think I would have had the good rise anyway with all AP flour, as I believe the better rise was due to using a smaller amount of "recipe starter" compared to last time, rather than any small variations in flour. The dough had a good feel to it compared to other versions I've done with a higher percentage of preferment, more like when I do the yeast preferment versions.


zolablue's picture

Wow, what is wrong with that!  Really, that is good looking bread.  I know you don't like some of the hole placements (hehe) but I really love the rustic look and the color is great.  Plus you really got it to spring.  Awesome – now I’m going to have to try again.  :o)


I was going to mention something else to you since I think the proofing is so critical and I’m having trouble understanding sourdough in that regard.  I have started using those tubs that have increment marks and it is so much easier now.  I only use a bowl, at times now, just for initial folds if I have an extremely slack dough but even then it ends up going into a tub for the remainder of the fermentation.  I know I was overproofing in the bowl because I simply had no way to determine how far it had gone.

bwraith's picture


The water should be 16.5 oz in this recipe. Sorry, I don't know where that came from. It was correct in the spreadsheet, but I somehow failed when transcribing it to the blog.

I used a rising bucket this time, too. It does make it much easier. Then you at least have the indication of the amount of rise. I think it's possible that the poke test would be more consistent if you have the dough supported consistently and poke the same distance from the edge of the container each time.


dulke's picture

Ha! I tried creating a sourdough ciabatta just last weekend, was only marginally successful, so I am very interested to see this post - looking forward to trying it out. Photos look great!

bwraith's picture

HI dulke,

I hope these blog entries help. Note I did another one previous to this one. It worked OK, and had a more uniform, softer crumb, but not the oven spring. It does come out, although not quite as fine and soft a crumb as I'm used to with the yeast preferment methods, such as in Glezer or the BBA. I would still say this may be better with all AP flour. After looking up the Golden Buffalo flour on the Heartland Mills web site, I realize it has about 14% protein. So, maybe this flour is more of a contributor to the handling quality of the dough itselft and to the texture of the crumb, which was more chewy than I expected.

If you do the recipe, I'd be interested to hear how it went. I'm still trying to figure out the best way to do this.


zolablue's picture

Bill, just to clarify, are you saying to use 9 ounces of firm starter?  But you mean firm levain or "recipe starter", right?  Not as in "firm sourdough starter" as I keep mine.  Gosh, I hope I'm not sounding all confusing. (

But I could not imagine using 9 ounces of my starter, is why I'm asking.  The most I ever make is 90 grams and most of the recipes I've seen that are firm-starter recipes call for 25 - 35 grams in Glezer and up to 58 or so grams in Hamelman.  I always feel I'm cheating when I use 75 grams of discarded starter to a recipe but I want to use it.

bwraith's picture

Hi Zolablue,

Right, I'm suggesting you build a 9 oz 65% hydration dough from about 1.5 oz of your firm starter and additional flour and water. To be very similar to what I did, you would use 1.5 oz of your firm starter, which would have about 1 oz of flour and 1/2 oz of water in it, roughly. So, you could then add 4.5 oz flour and 3 oz water to it. You would then have 3.5 oz water and 5.5 oz flour for 9 oz total for close to 65% hydration, and the time to rise that recipe starter by double would be around 4 hours plus or minus, since it's roughly a 5x feeding.

On another note, I've been thinking that since a lower proportion of preferment, or something similar which is using young or not very ripe preferment, seems to work better so far, you could go all the way to effectively no preferment, i.e. it may be that it's easier and better just using 1.5 oz of your firm starter or 2 oz of my liquid starter and just make a 80% hydration final dough, let it rise by double, which will take a while with the much lower amount of starter, maybe retard in the refrigerator for convenience and possibly flavor, fold occasionally, maybe more toward the end just before shaping, like the no-knead SD style I've read on some posts recently, and shape like ciabatta at the end would work well, too. I may try that approach when I get another bake day. I guess that would be my first try at something similar to the no-knead techniques I've been reading about.


gianfornaio's picture

I found that the first batch had darker, harder crust, whereas the second was a little bit on the pale side, even though I baked both for the same amount of time in an oven that I know was fully heated and at the same temperature. Also, the oven spring was much better on the first batch. I believe the steam conditions, temperature, and handling were very, very close to identical for both bakes.

Not to trouble you about this, but I'm curious-- when you take the first loaves out, do you let the oven heat for much (or any) time before you put the next batch in? It seems like the loaves would necessarily reduce the temperature of the stone baking surface, possibly even to the point where the stone would be cooler with a hot loaf freshly removed than it had been before the cold loaf was dropped on it. How does one test the actual stone surface temperature?

Would overproofing hinder caramelization of the starch that much? My (amateur) inclination is that an imperceptibly cooler stone is a more likely factor in the oven-spring and browning difference.  Actually, regarding overproofing, the slackened dough would spread more, wouldn't it? So the holes would be more horizontal in cross-section, not just smaller. Were they? Just curious.


bwraith's picture


Maybe so. I guess the first ones baked for about 15 minutes, and then I prepared the second ones, so they were in the oven about 1/2 hour later. During that 15 minutes after the first batch came out, I set the oven back to 500F. I guess it's possible that isn't enough time for the stone to heat back up.

It seems to me the holes were a bit spread more and not necessarily a lot smaller.

Unfortunately, I just haven't done all these things enough to be able to tell the difference.