Menu
Search
Menu
Close

Cocoa Powder

Published May 2018

How we tested

When we want big chocolate flavor in everything from cookies and cakes to puddings and pies, we turn to cocoa powder. It has a higher proportion of flavorful cocoa solids than any other form of chocolate, so ounce for ounce, it tastes more intensely chocolaty. It's made in two styles—Dutch-processed and natural—and there's fierce debate in the baking world about which is best. Both styles have staunch supporters who are convinced that using the wrong type will ruin a dessert. For years, we also viewed Dutched and natural cocoa powders as distinctly different products. But when we last evaluated cocoa powder, something surprising happened: A natural powder won, a Dutched powder came in second, and the rest of the lineup was a jumble.

In the years since, we've remained curious about cocoa powder. Some of our test cooks prefer the dark color of Dutched powder and swear that it has richer, deeper chocolate flavor to match. Are they onto something? Is choosing between Dutched and natural the most important decision you can make when buying cocoa powder, or is there more to it than that?

To find out, we sampled eight nationally available cocoa powders (priced from $0.34 to $1.70 per ounce): four Dutched and four natural. To zero in on how much Dutch processing matters, we carefully selected recipes for testing: two different sheet cake recipes—one that calls for natural cocoa powder and another that uses Dutched—and a cookie recipe that doesn't specify which style to use.

The results were mixed. While some desserts were simply acceptable, others were excellent. The good-enough cakes and cookies were tall and “airy” with a “crumbly” structure but a little “dry.” Across the board, we preferred “moist” and “fudgy” desserts. Our favorite cakes had a “plush” texture, and cookies toed the line between chewy and tender. As for flavor, samples ranged from “mild” and “slightly fruity” to “intense,” “complex,” and “earthy,” with the slight bitterness of good espresso or dark chocolate. Why had some desserts been dry, mild, and lean, while others were so rich, flavorful, and decadent?

From Pod to Powder

Cocoa powder—and all real chocolate—starts with cacao pods, the fruit of the tropical evergreen tree Theobroma cacao. Each pod contains between 20 and 50 beans (also called seeds). The beans generally taste bitter and are surrounded by a fruity-tasting, milky-white pulp, according to Gregory Ziegler, a chocolate expert and professor of food science at Penn State University. The beans are fermented, a critical process that develops their dark brown color, before being roasted. The fermented beans are either roasted whole or are shelled and roasted as nibs. Next, the nibs are ground into a paste called chocolate liquor, which contains a mix of cocoa solids and cocoa butter. Some of the chocolate liquor is used to make candy and chocolate products. The rest is pressed to remove most of the cocoa butter, which is also used to make chocolate. The cocoa solids that remain are ground into small particles and become cocoa powder.

In the 19th century, a Dutch chemist and chocolatier named Conrad Van Houten developed an optional step for the above process, known as Dutching, Dutch processing, or alkalizing. Chocolate is naturally slightly acidic, and so is cocoa powder. Treating the cocoa with an alkalizing agent neutralizes the acid, raising the powder's pH from about 5 to about 7. Natural cocoa powder is usually sandy brown with a reddish tint and tastes bright and fruity; Dutch processing darkens the color to velvety brown or near-black and mellows the cocoa's more astringent notes so that its deeper, earthy notes come to the forefront.

Dutching is not a one-size-fits-all process. Ziegler told us that manufacturers use a variety of alkalizing agents, such as potassium carbonate or sodium carbonate. They can also adjust the temperature and time of the process and may opt to alkalize the nibs, the cocoa liquor, or the final pressed powder.

Given the potential variation in processing, we were curious about how our powders compared with each other, so we asked an independent laboratory to measure the pH of each cocoa. The lab reported that the pH of the natural powders ranged from 5.36 to 5.73 and the pH of the Dutched powders ranged from 6.88 to 7.90. It doesn't sound like much, but one point indicates a tenfold difference in acidity.

When we reviewed the results of our recipe tests, we saw that some trends fell in line with the Dutched versus natural division. The more acidic natural powders produced some of the tallest, airiest, and crumbliest cookies and cakes. On the other hand, most of the Dutched powders produced baked goods that hadn't risen quite as tall. This makes sense: Baking soda, a common chemical leavener that was in all three of the recipes we tested, releases carbon dioxide bubbles when it reacts with acid and moisture; this is one of the reasons that doughs and batters rise in the oven. The acidity level affected how our cocoa powders interacted with the baking soda and seemed to have played a role in how high our baked goods rose.

In general, the tall, airy cakes and cookies made with natural cocoa powder were perceived as much drier. Our tasters preferred the fudgier, moister desserts made with less-acidic Dutched powders. In fact, a Dutch-processed cocoa powder won every tasting—even when used in a recipe that was specifically designed using natural cocoa powder—and Dutched products took the top three spots overall. But one Dutched powder consistently landed at the bottom of the rankings; baked goods made with it were slightly dry instead of tender and rich. Dutching is clearly an important variable, but it wasn't the whole story.

Fat Is Another Major Factor

There's another big divide in the world of cocoa powder: fat content. When the cocoa liquor is pressed, some cocoa butter remains with the solids, so commercial cocoa powders generally contain between 10 and 24 percent fat. While that full range is technically achievable, cocoa powders don't run the full spectrum. Instead, they're manufactured in two levels: low fat and high fat. An independent lab analyzed the samples and reported that three products in our lineup contained about 11 to 12 percent fat; the rest had nearly double that, about 20 to 22 percent.

Suddenly, things started to come into focus. Most of those high-fat powders scored high in our tastings. Why? Fat adds richness and flavor. It can also help ensure that cookies and cakes bake up moist and tender. The flip side is that desserts made with the low-fat powders, though still acceptable, tended to be dry. The only low-fat cocoa powder to land in the top half of our rankings was Hershey's, which may owe its high score to its familiar flavor. Our other favorites contained at least 20 percent fat, for rich, moist, flavorful cakes and cookies.

The Opposite of Fat Is . . . Starch?

Baking with a low-fat cocoa powder means risking dry baked goods—but not just because fat adds richness and helps prevent baked goods from drying out. Starch is a natural component of all chocolate and cocoa powder, and the less fat cocoa powder has, the more starch it contains. These starches are very absorbent; they're able to soak up 100 percent of their weight in moisture. By comparison, flour can absorb 60 percent of its weight. Like excess flour in a recipe, the extra cocoa starch present in low-fat powders traps moisture and makes for dry cakes and cookies. It's especially noticeable when recipes call for a high ratio of cocoa powder to flour, as with one of the chocolate sheet cakes we made.

To isolate the role of starch, we performed a simple experiment with all eight cocoas. We whisked together precise amounts of cocoa powder and water, transferred the slurries to bags and vacuum-sealed them, and heated the mixtures in a sous vide water bath to exactly 180 degrees, the temperature at which the starches in cocoa powder gel, or thicken. The differences were striking. Some were very firm and bouncy, like a memory foam pillow, and others were almost runny.

Of course, none of our desserts had been liquid-y or pillowy, but the results lined up nearly perfectly with the textural differences we'd noticed in the cookies and cakes. The cocoa powders that were the firmest in our experiment had lots of moisture-absorbing starch and made tall, airy cakes that tended to be dry; those that produced the runnier slurries had less starch and resulted in moist, fudgy cakes. The pattern was even more evident when we looked at the cookies. Using high-starch powders gave us cookies that rose and had crumbly, cakey textures. Cookies made with low-starch powders spread more, and the available moisture in the dough helped keep them chewy and fudgy. The difference in how much the cookies spread was dramatic. The cookies with the most starch averaged about 3.2 inches in diameter, compared with almost 3.8 inches for those with the least amount of starch—a difference of more than ½ inch. That's a big variation for a chocolate sugar cookie.

Buying the Best Cocoa Powder

By the end of testing, we realized that the old Dutched versus natural debate wasn't wrong but it also wasn't the whole story. The performance of cocoa powder is determined by a complex system of factors including pH, fat, and starch content. For moist and tender baked goods, we recommend buying a Dutch-processed cocoa powder that's high in fat and therefore low in moisture-absorbing starch. (If the nutrition label is all you have to go by, seek out a product with at least 1 gram of fat per 5-gram serving.)

Our top three scorers fell into this category. Each produced “moist” and “fudgy” cakes and cookies that struck the right balance between “chewy” and “tender.” The best of the bunch was our former runner-up (and longtime favorite Dutched product), Droste Cacao ($9.99 for 8.8 ounces), which has the right combination of factors to ensure decadent chocolate desserts with perfectly moist textures and the “sophisticated,” “complex” flavors of good espresso and fancy chocolate. It's well worth seeking out.

Methodology

We tested eight cocoa powders, including a mix of Dutched and natural products, all available at supermarkets. Panels of 21 tasters sampled them in three blind tastings: in chocolate sugar cookies and two kinds of chocolate sheet cake. An independent laboratory analyzed the fat content and pH of the cocoa powders; a lower pH indicates higher acidity. Prices were paid in Boston-area supermarkets, and products appear in order of preference.

Try All Access Membership Free for 14 Days

Included in your trial membership

  • 25 years of Cook's Illustrated, Cook's Country, and America's Test Kitchen foolproof recipes
  • In-depth videos of recipes and cooking techniques
  • SAVE all your Favorites for easy access
  • Up-to-Date reviews and product buying guides

Get America's Test Kitchen All Access — become the Smartest Cook you know, guaranteed.

Email is required
How we use your email address

The Results

Winner
Recommended

Skippy Peanut Butter

In a contest that hinged on texture, tasters thought this "smooth, "creamy" sample was "swell" and gave it top honors, both plain and baked into cookies. Its rave reviews even compensated for a slightly "weak" nut flavor that didn't come through as well as that of other brands in the pungent satay sauce.

$2.39 for 16.3-oz. jar (15 cents per oz.)*
Recommended

Jif Natural Peanut Butter Spread

The big favorite in satay sauce, this peanut butter's "dark, roasted flavor"—helped by the addition of molasses—stood out particularly well against the other heady ingredients, and it made cookies with "nice sweet-salty balance." Plus, as the top-rated palm oil-based sample, it was "creamy," "thick," and better emulsified than other "natural" contenders.

$2.29 for 18-oz. jar (13 cents per oz.)*

Reese's Peanut Butter

This is what peanut butter should be like, " declared one happy taster, noting specifically this product's "good," "thick" texture and "powerful peanut flavor." In satay sauce, however, some tasters felt that heavier body made for a "pasty" end result.

$2.59 for 18-oz. jar (14 cents per oz.)*

Skippy Natural Peanut Butter Spread

The only other palm oil-based peanut butter to make the "recommended" cut, this contender had a "looser" texture than its winning sibling but still won fans for being "super-smooth." Tasters thought it made an especially "well-balanced," "complex" peanut sauce.

$2.39 for 15-oz. jar (16 cents per oz.)*
Recommended with Reservations

Peanut Butter & Co. No-Stir Natural Smooth Operator

Though it says "no-stir" on the label, this "stiff" palm-oil enriched peanut butter was "weeping oil" and came across as "greasy" to some tasters. However, it turned out a respectable batch of cookies—"chewy in the center, crisp and short at the edge"—and made "perfectly good" satay sauce.

$4.49 for 18-oz. jar (25 cents per oz.)*

Maranatha Organic No Stir Peanut Butter

On the one hand, this organic peanut butter produced cookies that were "soft and sturdy" yet "moist," with "knockout peanut flavor." On the other hand, eating it straight from the jar was nearly impossible; its "loose," "liquid-y," and "dribbly" consistency had one taster wonder if it was "peanut soup."

$5.69 for 16-oz. jar (36 cents per oz.)*
Not Recommended

Smart Balance All Natural Rich Roast Peanut Butter

Besides being unpalatably "tacky" and "sludgy," this "natural" peanut butter suffered from an awful "fishy" flavor with a "weird acidic aftertaste" that tasters noted in all three applications. Our best guess as to the culprit? The inclusion of flax seed oil, an unsaturated fat that's highly susceptible to rancidity.

$3.59 for 16-oz. jar (22 cents per oz.)*

Smucker's Natural Peanut Butter

With its only additive a negligible amount of salt, the only truly natural peanut butter in the lineup elicited comments ranging from mild dissatisfaction ("needs enhancement with salt and sugar") to outright disgust ("slithery," "chalky," "inedible"). Cookies were "dry and crumbly" with a "hockey puck" texture, and the satay sauce was "stiff," "gritty," and "gloopy."

$2.69 for 16-oz. jar (17 cents per oz.)*