Article Review: Coca Cola’s Orange Juice Algorithm

Orange Juice Algorithm

Coca Cola has been using an algorithm to increase revenue by perfecting the taste and the chain supply of one of it’s up and coming segments, 100% natural orange juice. I happen to be a huge fan of orange juice and Coca Cola’s marketing campaigns so naturally the recent article in BusinessWeek stood out to me. I commend the entire company and especially Coke’s vice president of business acceleration, Doug Bippert, for their dedication to using data driven, quantitative solutions. In fact upon more research it was so impressive it even encouraged me to look into the company’s stock. (Disclaimer I am long on KO)

About the Algorithm “Black Book”

Black Book, the algorithm they describe, was implemented by industry expert Bob Cross at Revenue Analytics. According to the article they use “up to 1 quintillion decision variables”… that is a lot of factors! Some of those include “external factors such as weather patterns, expected crop yields, and cost pressures.”

The article suggested that the model’s response “includes detailed data about the myriad flavors – more than 600 in all – that make up an orange, and customer preferences.” Additional output includes “how to blend batches”, predictions for supply chain out as far as “15 months”, and optimal time to pick the oranges. Finally, not only can they accurately model all these but as they state, “we can quickly replan the business in 5 or 10 minutes just because we’ve mathematically modeled it.”

How to build your own Black Book Algorithm

Here I propose the steps for building your own predictive model, regardless if you choose to use a consultant or implement it yourself.

  1. Define response(s) and what you want to model.
  2. Collect data on all possible factors that might influence your responses and their associated responses as defined previously.
  3. Use a variable reduction technique to reduce your factors (ie. take it down from 1 quintillion to something a little more realistic, say 10000 or less)
  4. Create a model. This takes a little experience and data knowledge but it could be as simple as a linear model however more likely they used a form of boosted forests.
  5. The model will tell you the relative influence from each factor. Here you’ll consider reducing the significant factors even more to probably the top 100 for each response.
  6. Predict the future. Apply the model to future or anticipated values for each of the significant factors. Since you are only looking at the most influential factors this part should easily take less than a few minutes each run.
  7. Optimize future response. Now you can also run simulations to experiment and optimize the future output. Depending on the complexity of the model and the ability to converge on an optimal response this last part could take longer than 20 minutes.

Comments appreciated. If you would like to talk more about implementation details for a project you are working on please feel free to contact me.

Demand the (Right) Right Data

Mark Twain once noted,”The difference between the right word and the ‘almost right word’ is like the difference between lightning and a lightning bug.” So too with data.

Know the data, where it comes from and what it means.

Analytical driven management

“Interestingly, the top performers also turn out to be the organizations most focused on improving their use of analytics and data, despite the fact that they’re already ahead of the adoption curve,” said Michael S. Hopkins, editor-in-chief of MIT Sloan Management Review. “We discovered that the more managers know about analytics-driven management — and see how it can create value — the more they know that they want to know more.”

Dealing with the Data Deluge – MIT News Office

Is it worth going back to school to get an MBA

Over and over I have debated looking for an MBA program that would excite my interest for business and with enough technical emphasis. As an engineer I feel like some of the business math courses would be interesting but mathematically slow.

There is also the debate that taking two years off would be too great of an investment of time and money that may never pay off. Today Yahoo brought to light this debate when discussing a bachelors degree vs just a high school education in their featured article

Your opinion on this debate will be worth hearing. In the mean time I’ll work on a cost benefit analysis between the options.

Breaking the Time Value Curve

We’ve all noticed that there exits a relationship between time and money. Teens learn quickly that society puts a price tag on their time starting at a minimum wage. Depending on the location, additional skills, the industry, and several other variables this minimum wage might differ. At any rate, one hours work might afford them one hamburger at the restaurant, or one movie ticket at the theater, etc.

This illustrates very well that things have an intrinsic value. The hamburger is sold for the price it is because of what it is made of and the cost of the time invested making it into it’s final consumable state plus or minus market factors like supply and demand as well as customer perceived value and so on. However, even though this is the price tag, the price goes beyond that for the consumer because he too must invest the time and other costs to attain the item. I’ve noticed that as time becomes more and more of an valuable to the consumer, then the intrinsic value of any item becomes less important and the cost of acquisition. Expecially when we consider that acquisition includes in many cases travel, decision making, the transaction, consuming or using, and in other cases possessing and maintenance.

So as time becomes more and more valuable to an individual the cost of items to that individual goes up as well and there is only one way to stop the cost from growing. The individual must break the time and money connection. Time cannot be so intimately tied to the amount of money the individual can make thereby determining the amount of items he can acquire. I know of two ways to do this.

The first way to adjust the dependence on time and money is to become educated and skilled in the work that you do. This makes it possible for you to accomblish things faster and to be sought after for what you do. This however only changes the slope of the time value curve. It is what makes the price that you pay for an item like a burger nearly irrelevant to the rest of the time that you spend with that transaction.

There is only one way to break the time value curve and that is to leverage people who have a lower time to money value than you do. For example if a task needs to be done and someone else can do it as good as you for less than your cost of you doing it yourself then do it. Even if they can do it in half the time or twice as long. As long as they can do it as good as you need it done for less than it would have cost you to do it yourself. This is a great justification for having a home cleaning service. If a family maintains the order and organization then it could easily be justified that spending the 2-4 hours on a saturday deep cleaning the home is not worth the price tag of a cleaning crew even if they can do it in 1 hour as long as the price you pay the cleaning crew is less then the total cost to do it yourself. A generic consumer price equation might look like

benefits of having service or item + time saved by not doing it ourselves + other benefits doing things that couldnt be done without it or because it took the pace of – tag price paid – other costs associated with getting – opportunities lost by having = cost to the individual

If the cost is negative then it is always a good idea to get that item or service. If it isn’t a good idea but it is still negative then you aren’t including all the benefits or disadvantages or maybe you are incorrectly valuing your time. For the cleaning example, many families, mine included, find that if we choose to have a cleaning service the cost for cleaning would be too great because we would miss the opportunities to teach our children the value of hard work and we also might have a disconnect from the impact of our actions.

Unfortunately this can be taken to the extreme. For example how do you value your relaxation time? and is time spent with some people more valuable then spent with others? Time at work vs time at home? And ultimately now you’ll have to decide if thinking about this topic and reading this post was even worth your time? Your input, references, and extended evaluation of this topic is important to me and part of the reason that I wrote the post at all however, you’ll have to ask yourself, what value do you place on submitting a comment below?

Update: a really good article

Public Radio

If there is one thing that I enjoy about my 35 minute commute each way to work it is listening to public radio. I am especially loving It doesn’t come on until 5:30 pm so some days I even take a little extra time to finish up things at work so i can time my commute just right so I’m listening to the entire segment for that day.

I’m adding it for the time being to one of my top links to the left so I can review the site’s top articles when I’m just surfing online. Part of what I really enjoy about the radio version is that they embelish the stories to help draw a mental picture. It is a nice change from tv or other similar media that is so visually stimulating. Way to go public radio, you have my support!