Baking Projects,  Copycat Recipes

Levain Bakery Copycat Recipe Showdown

The gooey inside of a Levain Bakery dark chocolate peanut butter chip cookie

The Premise

*UPDATE 10/5/20* I remade the Bravetart dough and rebaked the top 4 cookies in order to accurately rerank them

Pre-quarantine my partner and I were watching Cupcake Jemma’s “NYC cookie” video and they off-handedly remarked that we should roundup all the copycat Levain Bakery recipes and try them all. You can’t just say things like that and not expect me to take it seriously so I got to work gathering 7 recipes and created a spreadsheet of all the main ingredients (flours, sugars, butter, eggs, chocolate, and walnuts). I also added the number of cookies, the size of the cookie dough balls (if stated), chill time, oven temperature and bake time to the sheet to help me keep track of everything at a quick glance.

Spreadsheet of details for the Copycat Levain showdown

I’m an average baker—I can follow a recipe and end up with a great final product but I don’t necessarily understand the science behind what I’m doing. I chose these 7 different-enough recipes to give me a good variety of flours, weights, chill times, etc. The Levain cookie is a very specific cookie and I wanted to know if all these differing factors would, like, cancel themselves out and we’d end up with 7 true dupes for the classic cookie or if we’d end up with 7 distinctly different cookies.

I’ve been waiting until the supply chains seemed to have steadied because I didn’t want to waltz into the grocery store and clear them out of all the ingredients everyone’s scrambling to get their hands on and last week I felt comfortable enough to go ahead. I used all Good & Gather ingredients from Target except for the flours and chocolate. 

The flours were Gold Medal AP, Softasilk Cake Flour, Pillsbury Best Bread Flour and Martha White Self-Rising flour. The chocolate was Ghirardelli 60% chips for all the recipes that called for dark chocolate chips. I used Ghirardelli 60% bars to chop for Joshua Weissman’s recipe, Ghirardelli milk chips for Cupcake Jemma’s, and a mix of Market Pantry semi-sweet mini chips, Ghirardelli Grand semi-sweet chips and Guittard extra dark chips for Bravetart’s.

The Lineup

Here were my contenders: Joshua Weissman (JW), Cupcake Jemma (CJ), Broma Bakery (Broma), Brown Eyed Baker (BEB), Modern Honey (MH), Bravetart, and Delish. I chose this assortment because some of these bakers I already follow and also because they gave me a wide assortment of variables to “test.”

Note 1: I made Modern Honey’s recipe last year and very much enjoyed it. I was interested in repeating her recipe and comparing it to the others to see where it fell in the rankings.

Note 2: Broma Bakery and Brown Eyed Baker’s recipes are basically the same. Broma attributes her recipe to BEB and with a few tweaks included. Most notably Broma calls for AP flour instead of bread flour and no walnuts. Broma also recommends chilling for 30 minutes to 12 hours while BEB gives a hard and fast 30 minute chill time. I chose to do these two basically identical recipes to mostly see the differences, if any, between the two flours.

There are several, I think, major variables:

  • Types of flours: bread, cake, AP, and self-rising all make an appearance. BEB uses only bread, Bravetart uses on AP, CJ uses a combo of AP + self-rising flours, the other 4 use a combo of AP + cake flours.
  • Butter: 5 recipes called for cold cubed butter while JW calls for melted and Bravetart calls for softened.
  • Eggs: 6 recipes call for 2 eggs, JW calls for 2 whole + 3 yolks.
  • Chocolate: 6 recipes call for chips–JW calls for chopped, 4 recipes call for dark chocolate–CJ calls for milk, Bravetart an assortment and MH doesn’t specify.
  • Walnuts: Broma leaves them out entirely, Bravetart calls for untoasted walnuts or toasted pecans, MH doesn’t call for toasting.
  • Chill Time: MH requires no chill time, Bravetart calls for at least 12 hours and everyone else is somewhere between 30 minutes to an hour and a half.
  • Oven Temp: 3 at 375, 1 at 365, 1 at 350, 1 at 410 and 1 at 425.
  • Bake Time: everything from 9 to 26 minutes.

Each recipe uses a combo of white and light brown sugar. Delish, CJ, and Bravetart (close enough) use a 1:1 white to light brown sugar combo—Bravetart recommends toasted white sugar which I did. Broma, BEB, JW, and MH use more brown than white sugar.

JW and MH add cornstarch and skip baking powder. The other 5 recipes use a combo of baking powder and baking soda. Bravetart adds a pinch of nutmeg.

It’s widely rumored that Levain doesn’t use vanilla extract in their dough but Bravetart, BEB, and Broma call for it.

4 recipes call for 1 cup (2 sticks) of butter (CJ is close enough to 1 cup for me). Bravetart calls for ½ cup (1 stick), and JW 1.25 cups (1 ½ sticks).

The Doughs

5 of the doughs were your standard soft-ish cookie dough. Delish and CJ save the addition of the eggs for the last step and create a pie dough/wet sand dough that clumps together when smooshed in your hand.

Bravetart, Delish and JW have you weigh out 6oz dough balls. For Bravtetart and Delish that yielded 8 cookies while JW yielded 11. JW also specifically mentions using a 4oz cookie scoop to get the desired cookie dough ball weight which I had and used. CJ’s dough balls weigh 125g or about 4.5 oz and that yielded 13. Broma and BEB have you divide the dough into 12 balls while MH calls for 8 balls. I did attempt to keep the dough balls a little rough in hopes of helping create that kind of craggly look/texture. Using the scoop for JW kind of negated that and some of the others I got carried away and rolled them smoother. Other than size differences, all the dough balls looked pretty similar other than JW’s because of the chunks and shards of chopped chocolate.

The Method

I created a schedule around how long each dough needed to chill before baking.

Thursday night:

  • Measure and label all flours, sugars, chocolate and nuts (toasting if required).
  • Make Bravetart dough and chill in fridge. (Bravetart calls for at least a 12 hour chill time so it made sense to do this one the night before)

Friday:

  • Make and chill Broma dough in fridge. (Broma calls for 30 minutes to 12 hours of chill time so getting it into the fridge first would allow me to fit it in anywhere should the schedule change)
  • Make and freeze CJ dough (90 minutes in freezer)
  • Make and freeze Delish dough (90 minutes in freezer)
  • Make and fridge JW dough (45+ minutes initially, then scoop it and fridge again for 25 minutes)
  • Bake CJ dough
  • Bake Delish dough
  • Bake JW dough
  • Make and fridge BEB dough
  • Bake Broma dough
  • Make and bake MH dough
  • Bake Bravetart dough

Here’s how the schedule actually turned out as we went through the day:

  • Make and chill Broma dough in fridge
  • Make and freeze CJ dough
  • Make and freeze Delish dough
  • Make and fridge JW dough
  • Bake Broma dough (chilled for 95 minutes) (baked for ~18 minutes)
  • Bake CJ dough (chilled for 95 minutes) (baked for 24 minutes)
  • Scoop JW dough and fridge again
  • Bake Delish dough (chilled for 95 minutes) (baked for 24 minutes)
  • Bake JW dough (chilled for 85 minutes total) (baked for 14 minutes)
  • Make and fridge BEB dough
  • Bake Bravetart dough (chilled for 22 hours) (baked for 22 minutes)
  • Bake BEB dough (chilled for 35 minutes) (baked for 17 minutes)
  • Make and bake MH dough (no chill) (baked for 15 minutes)

Thursday night’s prep work took about 3 hours total between toasting/chopping walnuts, portioning all the dry ingredients, and making Bravetart’s dough. Friday I got started at 8:30AM, took about an hour break for lunch and was done by 3:30PM—so about 6 hours of work for the day and 9 hours in total.

I have 2 bowls and 2 paddles for my stand mixer so I was easily able to go from one dough to the next with plenty of time for my partner or I to wash the dirtied dishes and have them ready for the next batch. I had a sticky note on a cabinet door where I kept track of when a dough went into the fridge/freezer and also when it went into and came out of the oven so I could compare actual cook times to what the recipes stated. It was honestly a very low stress day thanks to having done as much prep as possible the night before.

I baked 3 cookies from each batch and then froze and bagged the remaining cookie dough balls to hoard in my freezer.

The Results

Did we end up with 7 Levain cookie dupes? No.

In appearance, Delish, CJ, BEB, Bravetart, and MH either nail it or get pretty dang close—Delish and CJ being the closest. Broma and JW spread and flattened and smoothed out more and looked like a standard chocolate chip cookie.

Several of the cookies had that crisp scone-like shell with the slightly gooey center—Delish, CJ, Bravetart, and MH.

While the BEB cookie looks very similar to the Levain, the inside was more fluffy/cakey than the Levain cookie which is not the texture we were after and not a texture I personally desire from a cookie. This could have perhaps been improved by a slightly shorter bake time. The edges did spread a bit and browned quicker than the rest of the cookie but it wasn’t enough to really notice a difference when eaten.

Broma and JW’s were soft and dense throughout like you would likely want from a traditional chocolate chip cookie. Because JW uses chopped chocolate, the main texture and flavor you get from the cookie is puddly chocolate and it obscures any potential gooeyness or flavor of the dough. It’s a delicious cookie but there’s no distinction between the chocolate and the dough like you would get with chips that retain more of their structure. The Broma cookie spread quickly in the oven so the thinner edges browned very quickly, similarly to the BEB cookie.

As mentioned, the Broma and BEB recipes are essentially the same—BEB uses all bread flour while Broma uses AP and BEB calls for a half hour chill time while Broma calls for 30 minutes up to 12 hours. I thought it was interesting that the Broma dough was chilled for about 90 minutes and the BEB only for 30 and yet the Broma dough still spread and flattened out while the BEB dough maintained more height. I’m assuming this, along with the texture differences, is because of BEB using all bread flour?

Bravetart has a 22 minute bake time which is what I did and although they were still nice and gooey in the center, they got a little extra brown on the outside and I thought I’d overbaked them. Looking back at her video however, her cookies are very brown so that may be just a byproduct of using toasted white sugar. Bravetart’s dough had a very unexpected “warmth” that we didn’t like. I assumed the culprit was probably that my pinch of nutmeg was likely bigger than Bravetart’s. Again, looking back to her video, she says if you can taste the nutmeg you’ve done too much so I need to work on my pinching. It was a great cookie that absolutely warrants a redo, either with the tiniest pinch or just eliminating it entirely.

The Delish cookie was a perfect balance of crispy shell and gooey center. It had a very balanced flavor overall and wasn’t intensely sweet. It nailed that super thick chonkiness that I think we all associate with the Levain cookie. It was just a great cookie and I think it truly did nail its goal.

The CJ cookie was the sweetest cookie of the bunch, I’m guessing because of the use of milk chocolate chips in tandem with the 1:1 ratio of the white and brown sugars. It was the only cookie that I scuttled over to my partner and was like “You have GOT to try this.” I have *quite* a large sweet tooth and am a lover of milk chocolate so it’s not surprising the extra sweetness appealed to me.

I noticed no difference between the recipes that called for untoasted walnuts (Bravetart, MH) and the recipes that didn’t. I did notice the lack of walnuts in the Broma cookie, obvs.

The Rankings

So there are two ways you can rank these cookies—how closely they came to copycatting Levain and just how good of a chocolate chip cookie they are. *UPDATE 10/5/20* I remade the Bravetart dough and rebaked the top 4 cookies in order to accurately rerank them.

For the Levain Copycat:

  • Delish
  • Cupcake Jemma
  • Modern Honey
  • Bravetart
  • Brown Eyed Baker
  • Joshua Weissman
  • Broma Bakery

Delish just absolutely nails it while CJ is a close second with it’s slightly smaller cookie and sweeter dough. MH and Bravetart are really good ringers while BEB achieves the look but falls short on the interior texture. The chocolate puddly-ness of JW’s cookie pushes it farther down on the list while Broma’s lack of walnuts combined with the flatness, thin edges, and uniformly denser texture puts it at the bottom.

I’ll be interested to see how the JW and Broma cookies change once I bake them from frozen dough balls—I’m hoping that’ll help give them more of that Levain-y height.

For a good ol’ chocolate chip cookie:

  • Cupcake Jemma
  • Modern Honey
  • Delish
  • Bravetart
  • Broma Bakery
  • Joshua Weissman
  • Brown Eyed Baker

JW’s cookie is a GOOD cookie, there’s no debating that. But calling for melting the butter and using 3 additional egg yolks, for me personally, puts it low on the list. If I’m going through the effort of dirtying an additional thing to melt butter, why not brown it? And, as a general rule, I am annoyed by recipes that use only part of an egg. I know, I know, make an omelet or something, yeah yeah—that omelet’s never getting made. Just realistically I know I’m not going to reach for a cookie recipe that has partial ingredients leftover. The Broma recipe yields a very similar cookie without the extra yolks or step of melting butter which is why it’s a slot higher on the list.

The BEB cookie was the most “fine” of the bunch. I took a bite of my sample half and didn’t feel the need to eat any more. For me it was just too cakey and I don’t love a cakey anything other than an actual cake. Again, the interior texture could possibly be altered by baking for less time but for this particular experiment it’s at the bottom of my list.

Despite the warm flavor of Bravetart’s cookie from my too-large pinch of nutmeg, I know that once that is tweaked it’s a standout chocolate chip cookie. I *don’t* know that toasting the white sugar did much though. The MH cookie is slightly sweeter than Delish which I personally enjoy and why it’s higher on this list. The only way I want to tweak the CJ cookie is by making them bigger to get even more goodness from a single cookie.

It’s worth noting that none of these cookies are bad cookies, regardless of where they fall on either list. They just offer different things and I wouldn’t turn any of them down.

The Re-Bake

There were 4 cookies I wanted to redo for one reason or another. JW and Broma both spread even after chilling the dough for the suggested amount of times and so I wanted to rebake them from frozen to see if that would help them attain their Levain-y height. BEB I wanted to rebake for less time because I didn’t like the cakey interior texture of the cookie from the original bake. And the Bravetart recipe warrants a total do-over because, man, that nutmeg sitch.

More than a week after the original showdown, I rebaked the JW, Broma, and BEB dough balls that had been frozen—Bravetart will have to wait for another day. Because I was baking frozen dough, I played the bake times by ear, just keeping an eye on them until they started turning golden brown and generally seeming “done.”

My first attempt at the JW cookie resulted in a very brown crust but a straight up raw interior. I baked another dough ball at 400 degrees instead of the original 425 for a couple of minutes longer and I think I was able to hit that sweet spot. It had more height and less spread than the original bake and had some of that cragglyness it had been missing. The Broma cookie held its shape better than the initial bake although it still spread and got some dark brown spots on the edges. The BEB dough looked very much the same as the original bake, spread less with no brown spots on the edges.

The cookie I was most interested in rebaking was the BEB cookie because I was hoping I’d be able to achieve that slightly gooey center and I’m very happy to report I got that. The outer shell was crisp and the interior much less cakey than the original bake. However, I still found the cookie overall just okay when compared to the others. There was no standout flavor even though the texture was much better.

The rebakes of the JW and Broma cookies were pretty much identical to their original counterparts. The big pools of chocolate in the JW cookie obscured any flavor or texture of the dough. The Broma cookie ~experience~ was improved by a more even thickness but was otherwise exactly how I remembered it.

After tasting the 3 rebakes and revisiting my initial rankings, I wouldn’t make any changes to the placements on either list.

The Re-Re-Bake

Once Erika did her Levain copycat bake-off and declared Bravetart the winner I decided it was finally time to remake that overly nutmeged dough. On September 24th I made my dough, the only changes being MUCH less nutmeg, not toasting my sugar, and using a combo of pecans and walnuts. I ran out of walnuts and the recipe actually calls for either so I thought this was fine. Two days later on the 26th I baked 3 of the dough balls. Interestingly they took several minutes longer to bake this time and didn’t get *quite* so golden brown this time which I’m assuming is related to not having toasted the sugar. The cookies still didn’t get that classic Levain height or the craggley texture and they weren’t uniformly crisp on the outside but they did have that delicious almost gooey center. I would definitely say stick to walnuts over pecans because the pecan flavor stands out more and that could be polarizing.

When I tried to think back to the original bake-off so I could but Bravetart’s cookie in its proper place, I began to doubt my memory. I mean, almost 3 months had past. So then I decided the best thing to do would be to take the dough balls of my top 4 cookies in both ranks—Delish, Bravetart, Cupcake Jemma, and Modern Honey—and do a rebake and tasting to see where the new and improved (ie: correctly done) Bravetart cookie fit in. And that’s exactly what I did on October 4th. I baked all the cookies from frozen, one at a time, and tasted them after the last cookie had cooled.

When it came time to rank the 4 cookies, I was, uh, not expecting to get the results I did.

Original ranking for the best Levain copycat:

  • Delish
  • Cupcake Jemma
  • Modern Honey
  • Bravetart

10/4/20 ranking for the best Levain copycat:

  • Delish
  • Modern Honey
  • Cupcake Jemma
  • Bravetart

Original ranking for a good ol’ chocolate chip cookie:

  • Cupcake Jemma
  • Modern Honey
  • Delish
  • Bravetart

10/4/20 ranking for a good ol’ chocolate chip cookie:

  • Modern Honey
  • Bravetart
  • Cupcake Jemma
  • Delish

So, some changes obviously! Here’s the thing I need to reiterate: I think that if you choose any of these 4 cookies, you’re going to have a great cookie experience!

As far as the copycat list rankings go, I think Delish still wins on being the most Levain-like dupe. Upon further reflection, (and biases aside because I love the CJ cookie) I think Cupcake Jemma’s use of milk chocolate and its smaller size bumps it down below Modern Honey. The Bravetart cookie didn’t get that tall height or craggleyness so I think that keeps it in 4th place.

Setting aside the need for the cookie to be a copycat brings us to the good ol’ chocolate chip cookie ranking. And that’s where things went nuts and I had to sit down for a moment and think about my life. Obviously Bravetart really jumped the line on this list. It’s a great tasting chocolate chip cookie *especially* if you take her advice of sprinkling it with salt before baking. Salting your chocolate chip cookie always gives it an edge. Delish is a good cookie, it’s impressive to look at with it’s crazy height, it’s very tasty, you can’t go wrong. I love the CJ cookie and I always will. She is delicious and sweet and a great size for everyday cookie-ing.

But. This brings me to the cookie I think everyone should make—Modern Honey. It’s universally beloved (Erika and I don’t agree on the best Levain copycat but we *do* agree on our love for MH), requires NO CHILLING, takes 15 minutes from start to finish, and you get an AMAZING cookie. It’s got a lot of the Levain factors—especially if you bake it from frozen to get that amazing height—so you can still impress all your friends (or yourself) and it’s a damn fine cookie.

Again, you won’t go wrong with any of these cookies. Really the rankings just be First, Second First, Third First, and Fourth First because they’re all just good. Pick a cookie, make a cookie, be happy!

Final (Short) Thoughts

This was an incredibly fun endeavor, even though my brain is still fried from the actual baking and dissecting my thoughts about cookies. I’m already thinking of another copycat recipe showdown I could take on because I guess that’s just where I am in my life now. (Starbucks Lemon Loaf??)

This post was originally a google doc that I wrote after the initial bake-off which happened on July 3rd.