Full disclosure – I am not a kitchen person. I am more at ease in front of an open-flame grill. However, I got into the quarantine baking kick, for similar reasons as to why I perfected my pizza dough recipe in the Peace Corps – sometimes you can’t get what you want.
For me, the need is gluten-free (GF) bread. I am one of those people with allergies, and sadly wheat is a biggie. Unfortunately, during my grocery store runs over the last few months, the GF bread was always sold-out. Needing another option to keep my PB & J habit alive, I took stock of my pantry and decided that sourdough bread made the most sense.
I started with reading regular and GF recipes from my cookbooks and the internet. Each time I used a recipe, I document how I altered it and the resulting bake. This iterative nature of incremental adjustment feels similar to how I collect, clean, and analyze data. It also feels like a homegrown version of “the Great British Baking Off.” I imagine what Paul Hollywood would say if he saw the lack of uniform air bubbles. Fortunately, all the loaves have had crisp crusts and a sweet hollow sound when tapping the bottom, which indicates a good bake 😀.
When I collect data, I look through multiple sources for validation or an official, definitive reference. Without reliable, relevant, and documented sourcing material, the dataset is no better than GF bread without a binder. The kitchen equivalent of unedible mush that wastes valuable resources. GF flours aren’t cheap, and sourdough takes hours to ferment. My first attempt was pretty pathetic, but I learned a lot.
(Baker’s note: When starting, keep it simple until you better understand the data, the steps, and the process.)
Ask for help
Sometimes you ask for help. I called my mother. When I went GF a decade ago, my mother was amazingly supportive. At every holiday meal I came home for, she made sure there was a tasty GF dinner option just for me. She spent weeks perfecting a GF flour combination for a loaf bread so I could have leftover sandwiches. She knows well the drama of making GF bread. I also dug a little deeper and examined my sources. I found more recipes that explained each step’s purpose. It is just as important to understand “why” something went going wrong, as the “what” went wrong.
My second loaf was exponentially better; it was edible. It had a beautiful crust and was light and moist inside. I followed directions to the letter. No deviations, no creative inspirations. Regrettably, it was a little bland and did not have uniform air bubbles throughout the bake. I needed to kick up the flavor and do something to improve the outcome of the bulk fermentation step. Like data storytelling, I had identified the purpose of each step and decoded the intention and expectation. Unfortunately, it was also a pretty small loaf. But at least I did not spend a lot of time and energy, creating something that didn’t work.
(Baker’s note: Be open to learning from others to build confidence in your own processes and methods.)
Iterative learning is vital to forward momentum
Due to high demand and low yields, loaf number three was quickly underway. I adjusted the flour type and proportions to yield a richer, nuttier flavor and created a better environment to let the bread dough ferment. This iterative learning made a difference in the density of the bread, which resulted in a lighter, more substantial loaf.
(Baker’s note: It was almost perfect; but getting something right once, does not automatically mean all future bakes will be equally successful.)
Sometimes the best-laid plans don’t pan out. You encounter an unexpected learning event, and one misstep negates all the earlier hard work. It is even more impactful when its a mistake on something that seemed simple and straightforward. This is what happened with loaf number #4. I did everything the same as loaf #3, but at the last minute, something went very, very wrong.
In a bleary pre-coffee haze, I started the bake at the wrong temperature. The bread never had the chance to reach its full potential. The temperature never got hot enough for the yeasty beasties to do their job.
If you aren’t paying attention all the way through your data project, your data could set you up for an super dense, overbaked, not very satisfying result – no matter how you slice it.
(Baker’s note: It was flavorful and pretty, but best-laid plans can quickly go sideways.)
Seeing through the assumptions brings clarity to a process
My fifth loaf is currently in the works. The well-fed brown rice flour starter sits in a warm, loving environment, eagerly awaiting the moment I have time away from the computer to mix it with all its best friends for a bread party. All these iterations clarify the parts of the process that cannot be altered without seriously undermining the outcome.
Data storytelling is an iterative process. A colleague once commented that it is a “safe to fail” approach. Incrementally learning and adjusting the ingredients and process means you never waste too much time or resources on pieces that don’t meet your needs or expectations. You build on mistakes until you can turn each mistake into a success, and you have a sturdy, verified foundation for your analysis.
(Baker’s note: Over the years, I stocked my pantry with various GF flours and starches – each has a purpose and a best use case. Of them all, I am most looking forward to using the plantain flour. I enjoy the non-conventional, while sometimes a bit trickier, it often pays off and provides more satisfying results.)
Want more data stories? Follow me on Instagram @data_with_devon.
Contact me if you’d like to discuss your data story.
Interested in learning how data impacts our every day lives in real & meaningful ways? Check out the rest of my blog.