Homemade Vanilla Extract – Week 1.

October 9, 2010

After a week, some of my 4 vanilla extract recipes are “officially” ready; Recipe #1 (Chris Kimball’s 2009 Recipe) and Recipe #2 (his 1993 Recipe). During this week I have shaken all four jars daily. Today, per the recipe, I filtered Recipe #1 through cheesecloth which halts the extraction process. It’s flavor will not improve from here on out. Recipe #2 will not be filtered so will continue to mature.  (See Last Week’s Post for all 4 Recipes)


I performed two tastings on each of Recipe #1, #2 and #4: (1) tasting the vanilla extracts straight from the bottle, and (2) mixing each in 1 tablespoon of plain, unflavored yogurt and 1/8 teaspoon sugar. I did not incorporate either into a cooked recipe at this early stage.  All four recipes still have a strong aroma of alcohol. Recipe #1 and #2 smell only vaguely of vanilla, and Recipe #4 smells the most like vanilla. As expected, the amount of vanilla beans directly corresponds to the intensity of vanilla in the aroma.

Week 1 Results:

  • Recipe #1: 0-star. Very weak. Unusable.
  • Recipe #2: 1/2 star. Very weak. Not ready.
  • Recipe #3: Ready on December 1. Not sampled.
  • Recipe #4: 1 star. About 1/3 potency of McCormicks.
  • McCormicks: 4 stars. Classic vanilla flavor, but $4 per ounce. Corn syrup added.
  • Imitation Vanilla: 2 stars. Similar, but off flavor. Made from wood by-products and chemicals. Yuck.

Based upon my tastings, in all cases, 1 week is not nearly enough time to obtain sufficient potency no matter what the concentration of vanilla beans.  Because I halted the extraction process for Recipe #1; I can declare it a failure. I would not include it in any recipe, as it is too weak and taste too boozy (at least the booze would cook off during cooking). So clearly, heating the vodka in Recipe #1 does not compensate for either (a) waiting longer for the extract to be ready, or (b) a reduced number of vanilla beans.

To be honest, I was skeptical about Recipe #1’s potency from the beginning. It uses the fewest beans (only about 20% of the FDA required potency) and the recipe only uses a fraction of the time as my other recipes.

The jury is still out on Recipe #2, though, because the recipe call for it mature further. From here on out I will shake the bottles once a week, and will continue to post and compare the results at intervals over the next 6 months.

More Vanilla Bean Details:

I ordered “Extract Grade” (another name for “Grade B” beans). However upon receipt, it turned out that it would take 240 beans to make one pound, when they were advertised as 140-160 per pound. Eventually, the website sent me more beans labeled “Grade A”. However, they turned out as 160 per pound: the upper range for “Extract Grade”. In the end, I got what I paid for.  But to avoid confusion I would suggest that you contact your seller beforehand to be sure that you are getting what you paid for.

One weakness of all recipes (Except Recipe #4) is that they list the number of beans per cup of vodka. But because they don’t specify which grade of beans. “Grade A” beans are heavier (about 100 beans makes one pound), and I used “Grade B” (150 beans makes one pound). There is 40% variance in potency due to this ambiguity.

%d bloggers like this: