Tuesday, September 1, 2009

A semi-scientific approach to homemade vanilla extract

Months ago I saw a post on Chocolate & Zucchini about making your own vanilla extract. After my past adventures in infused vodka (for drinkin'), I was intrigued, especially by Clotilde's comment that using rum instead of vodka as the base results in a tastier extract for baking.

Mad scientist that I am, I decided to test this out. Luckily, Costco sells normally pricey whole vanilla beans for a song. I picked up a pack of 10 for something like $11.

The plan
  1. Make vanilla extracts with vodka, rum and -- why not? -- 190-proof grain alcohol (a.k.a. Everclear).
  2. Judge the deliciousness of each in a series of blind taste tests. 
  3. Take over the world. Give the tastiest to foodie friends for birthday and Christmas presents.

Vodka-based extract after 2 days. Is it just me, or does that bean look like a leech?

The recipe
  • Slice each vanilla bean lengthwise so the alcohol will reach the tasty seeds inside. If necessary, cut up the beans so they will fit in the jars you're using. 
  • In a 1-pint glass jar, combine at least 1.5 beans with 1 cup of rum, vodka, grain alcohol, or other potable solvent of your choice. By the way, these do NOT need to be top shelf alcohols.
  • Cap the jar and shake vigorously. 
  • Store in a cool, dark cupboard for a minimum of 6 weeks. I sort of forgot about mine…they sat for 4 months. 
  • Check for color and flavor periodically, if you remember. Shake the jar to help the infusion process along. 
  • Don't take the beans out, just use it as-is and leave the rest to infuse more and more and more.
Here's what my extracts looked like after 4 months of infusing. The grain alcohol is in the middle. See that difference in color?

The taste tests

I decided to test out the extracts in both this vanilla wafer cookie recipe and homemade whipped cream. I also decided to test them against store-bought vanilla extract, in this case Trader Joe's bourbon vanilla extract. 

To make sure the vanilla really stood out in both recipes, I doubled the amount I'd normally use. However, there was no way for me to use the store-bought and homemade vanillas at the same strength, as I didn't know how much vanilla went into the little jar from TJ's

For the cookies, I made the dough, divided it into four portions, and added the same amount of the different vanillas to each portion, and baked everything for the same amount of time. 

For the whipped cream, I whipped 1 pint of cream with about 1/2 cup sugar, divided it into four bowls, and added the different extracts. 

I labeled the cookies A, B, C, and D and the whipped cream bowls 1, 2, 3, and 4. This made it into a blind taste test because my guests didn't know what was in each. 

The results: whipped cream

Whipped cream tasting samples. 

1. Whipped cream with TJ's vanilla. People said (without knowing which vanilla it was):  
"Nope -- too strong (more than 4)." 
"Wendy-made (could be store)." 
"Good attack, finish kind of disappears. Super vanilla?"
"Strong vanilla, store bought I think. Perfumey."*

2. Whipped cream with homemade vodka vanilla. 
"Dig it. Subtle." 
"Wendy-made, a little sweet"
"Store bought, very sweet."
"Sweet. Second favorite."*

3. Whipped cream with homemade grain alcohol vanilla
"Between 2 & 4." 
"Even taste throughout, sweet, light, least favorite (mild)." 

4. Whipped cream with homemade rum vanilla
"Too strong."
"Wendy-made. Bourbon whiskey? My favorite." 
"Even taste throughout, deeper than #3, eggier."
"Complex. Rum? Favorite."*

The results: cookies

A. Cookies with homemade rum vanilla.
"Halfway between B & C."
"Delicious - nice & buttery, my other favorite."
"Good, tasty."
"Good. Can't tell."*

B. Cookies with homemade vodka vanilla
"Strong vanilla."
"Too sweet for me, but still good."
"Good, mild."

C. Cookies with homemade grain alcohol vanilla
"Better, more buttery."
"Good, mild, least flavorful."
"Spread out less when baked, more tough texture."*

D. Cookies with store-bought vanilla
"Very strong vanilla, but buttery."
"Awesome -- my favorite."
"Good -- most vanilla. (Same as #1?) Store? Favorite -- most intense."

* My comments are singled out because they weren't quite blind (that is, just because I forgot which was which didn't make it an officially blind taste test).

Results summary

Overall, the consensus was that the rum-based homemade vanilla was best in the whipped cream (the vodka one was a close second), while the store-bought vanilla tasted best in the cookies. People expected the homemade extracts to be stronger than the store-bought, but that wasn't the case. The store-bought extract smelled much stronger than the homemade ones, and I think that adding more vanilla beans to my extracts to to beef them up would make them better for baking.

Have you tested your own vanilla extract? Let's hear about it in the comments. 


  1. Great idea! I love the idea of doing tests like this too.

    Also, we've got those same plates and bowls. :)

    (And I was just checking out your blogroll. Lots of similarities; I'll have to check out some of your links.)

  2. Hey thanks for doing this- I've been wanting to make my own vanilla but I was sort of overwhelmed by the number of different methods/booze choices that others have suggested. I like the idea of bourbon, mostly b.c I like to drink bourbon.
    I found this great place online where you can buy a pound of fancy vanilla beans for $20
    they also sell posters- you know if you are dying for new posters along with your stash of vanilla beans...

  3. Wendy -- Gotta love the Ikea plates!

    M.B. -- Wow, that's a lot of vanilla. And I think it's a good idea to make the extract with whatever kind of booze you enjoy drinking.

  4. I just started a replication of your experiment with both Mexican and Madagascar(ian?) vanilla beans from Penzey's -- will let you know my results after Christmas. Thanks!

  5. Rob -- that's awesome. Definitely keep us informed!


Note: Only a member of this blog may post a comment.