About the Cereal Market

What is the 'Cereal Market'?

The Cereal Market is an attempt to rank 'value' of cereals over time using Google search interest as an (imperfect) proxy. Basically, I wanted to see which cereal was most popular based on search interest and that devolved into this pointless exercise. I've split out specific summary pages for the cereals listed at the bottom, complete with 'price'(shown over time) and some 'fundamentals' of each cereal. Pop one of those pages open and you'll probably understand what I'm going for here.

How does the Cereal Market work? (and what are Cheerios (Ch)?)

As mentioned above, I am gauging 'value' of cereals over time using Google search interest as a proxy. Unfortunately, Google Trends has a few limitations:

1. Google Trends does not provide an absolute metric (e.g. # of searches) of interest, it normalizes the item against historical interest in said item across a 0-100 scale. This makes grabbing the Google Trends information for each cereal meaningless, since both could read 100 on the scale but have drastically different search interest.
2. To solve the first issue, we need to compare the cereals directly against each other in the same Google Trends query. However, although Google Trends does let us compare multiple items against each other (solving the lack of an absolute metric issue), it limits the comparison to 5 items at a time.

So, to circumvent the above two issues, I used an unofficial API for Google Trends, PyTrends, and wrote a script to compare each cereal 1:1 with Cheerios. This way, Cheerios acts as a sort normalizing factor, allowing two cereals (e.g. Trix and Lucky Charms) to be compared directly against each other by comparing both to Cheerios separately. For example, if Trix reads 10 against Cheerios while Lucky Charms reads 15 for that same week, I know Lucky Charms has more interest than Trix in that period.

That's pretty much how the 'value' calculation works, I loop through all cereals, comparing 1:1 with Cheerios, then I take those numbers, divide by the Cheerios interest #s, and multiply by 100. I call this currency Cheerios (Ch), as you'll see throughout site. So, if you look through cereals and see a value > 100, that means it had more interest than Cheerios during that particular time frame; conversely if < 100 then less interest in Cheerios.

How often is the cereal market updated?

I'm pulling weekly interest metrics, so more or less will update weekly.