I've seen this trope on TV shows from time to time. Well now, I know someone who actually has some bbq sauce that they want to send to a lab to find out what the recipe was.

Is this actually possible, and if so, how would we find a lab that can do it? Or is it just a silly TV trope.

Googling turned up nothing.

  • What is the bbq sauce -- the brand and variety? May 25, 2020 at 2:29
  • I know someone who has a pint of original Flint's BBQ sauce in a freezer. We've come pretty close to reproducing it, but I've tasted the original, and we're not there yet. Two people who remember original Flint's also agree that we're not there yet either. The sauce we've developed is by far the best BBQ sauce I've ever had, but Flint's has a signature taste that's beyond words. It deserves every bit of the reputation it has. May 25, 2020 at 21:23
  • 1
    Send it to this guy
    – mustaccio
    Mar 12, 2021 at 16:14
  • Wow, what an amazing site. There goes my weekend. Mar 12, 2021 at 18:32

3 Answers 3


As long as you have a list of potential ingredients, it would be possible to find out if these ingredients are in the sauce. For example, if you don't know what spices were used, you could start with a list of spices, find information on some signature chemical compounds found in each spice of the list, then tell the lab to find out which of these substances are present in the sauce. This would give you a pretty good list of substances (spices) you could use. You will need a curious chemist-food scientist who has experience with that kind of work and is willing to play a detective, I'm pretty sure you can't get to a general purpose organic chemistry lab and expect them to just plug it in and get a result.

Even that information won't be 100% certain. First, you would have to find substances which are present in one source ingredient but not another - and these are unlikely to be the main aromatic substances, since these tend to be shared between plants, for example eugenol is something you'll find in a lot of herbs. Second, you might have unusual combinations in which the plant may get into the recipe: for example, where the lab suspects the use of inverted sugar and hyssop as an herb, it might turn out that the recipe contained wildflower honey and the bees processed lots of hyssop.

And when you get the information, you still don't have a recipe. Both the ratio and the process are missing. A good cook (or food technician, for industrially produced food) can make educated guesses about possible processes, and with some work, they are likely to create some kind of replica, if the original recipe doesn't include surprising tricks.

So, you decide to do so, it is kinda possible, but it will be a long process involving experts, not a send-the-sample-get-full-answer kind of thing. If there are no businesses offering it as a service, it doesn't seem like a practical proposition.

  • 2
    You'll probably also have to invest into a certain amount of basic food chemistry research: this "detective" task is not only inherently difficult, but also has unusual compared to how food chemistry research usually goes. For a given substance of sufficient scientific interest, you may be able to get information from the literature about some plants that contain (or do not contain) it - but you cannot conclude that no other plants have it. Nor can you conclude that all substances in a particular plant species are known. Even for the plant-substance combinations we know, only a subset will May 24, 2020 at 17:59
  • 1
    ...have information about typical concentration ranges available. So calculating back even from an unprocessed mixture (no reactions due to cooking, frying, pH, time,...) will be very difficult. OTOH, if you add to this information about typical recipes (and maybe experiments that dig down into the differences between your sample and BBQ sauces prepared to compare to), you'll probably get further. May 24, 2020 at 18:03
  • 2
    @cbeleitesunhappywithSX I would go even farther than your suggestion - I doubt that the work of matching chemistry signatures to ingredients is doable by a random person who brushes up on research, the small traps waiting out there are probably unnoticeable even for a graduate who studied food chemistry but has not done this work . That's why I specified that you will need an experienced person involved in the project. And yes, you will also need somebody (probably not the same person) who will make guesses about the broad type of recipe used. And it certainly won't involve trying to...
    – rumtscho
    May 24, 2020 at 19:09
  • 1
    ... recognize all compounds in the recipe (impossible and unnecessary), it is just about having a list of possible components, assuming that it is complete (cook expert comes in here!), then finding 2-3 unique chemicals per component and measuring only a yes/no signal for each, nothing about ratios, browning etc. The naive method I have in mind is already too complicated, if I were a chemistry professor putting a grad student on that research, I would already be somewhat worried about it going well without any "nice" features.
    – rumtscho
    May 24, 2020 at 19:12
  • 2
    yes. I guess I was focusing more on that already "just ...finding 2-3 unique chemicals per component" may be an amount of reserarch that is worthy of a PhD thesis. The more so as smell and taste are very non-linear wrt. the involved chemical substances and their concentrations. May 24, 2020 at 19:59

From the chemical point of view, you could run the food through a spectroscopy device to have an exact substances list. From there maybe ingredients could be "guessed".



You can certainly analyze food for its content (to a degree) but that won't tell you the recipe. For example, when you caramelize or brown foods there is a very complicated chemical reaction creating hundreds of new molecules. Sending food to a lab won't tell you how something was cooked. If you're looking for percent of basics like sugar, water, etc., then a lab can be helpful. I suspect that most BBQ recipes are pretty simple so a lab would get you pretty far toward figuring it out.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.