Hitchhiker’s Guide to Analytics — Tea
Share and Enjoy
This article is not about making tea.
In fact, it contains no advice whatsoever on making tea or even drinking tea.
I’m writing it because it can be difficult to find decent tea while traveling the galaxy. And that’s ridiculous.
Nutri-Matic
You should know by now that Arthur Dent, the Earthling lucky enough to find himself rescued from the destruction of the Earth, soon after found himself standing on the bridge of the Heart of Gold, a truly amazing ship powered by the Infinite Improbability Drive.
If you didn’t know that, go read the books. I’ll wait.
All set? Right, then…
While the other members of the ship’s makeshift crew stood marveling at the majesty of a dual sunrise over the ancient planet of Magrathea, Arthur decided that he would quite like a cup of tea.
As a state-of-the-art vessel, the Heart of Gold was equipped with a Nutri-Matic drink dispenser. This is an amazing bit of technology designed to produce drinks customized to the tastes and dietary needs of its users.
The way it functioned was very interesting. When the Drink button was pressed, it made an instant but highly detailed examination of the subject’s taste buds and a spectroscopic analysis of the subject’s metabolism, then sent tiny experimental signals down the neural pathways to the taste centers of the subject’s brain to see what was likely to go down well.
However, no one knew quite why it did this because it invariably delivered a cupful of liquid that was almost, but not quite, entirely unlike tea.
It claimed to produce the widest possible range of drinks personally matched to the tastes and metabolism of whoever cared to use it.
When used, however, it unfailingly produced a liquid that was almost, but not quite, entirely unlike tea.
Analysis
The Nutri-Matic drink dispenser was, of course, a product of the Sirius Cybernetics Corporation, whose Complaints Division covers the major land masses of three medium-sized planets and is the only part of the Corporation to have shown a consistent profit in recent years.
After several days of tolerating the consistently unpleasant liquid produced by the drink dispenser, Arthur Dent decided he would pursue the subject further. He engaged the dispenser in conversation and told it, in no uncertain terms, that he wanted a cup of tea.
Arthur sat. He told the Nutri-Matic about India. He told it about China. He told it about Ceylon. He told it about broad leaves drying in the sun. He told it about silver teapots. He told it about summer afternoons on the lawn.
He told it about putting in the milk before the tea so it wouldn’t get scalded. He even told it some of the history of the East India Company, at least the bits involving tea.
“So that’s it, is it?” said the Nutri-Matic when he had finished.
“Yes,” said Arthur, “that is what I want.”
“You want the taste of dried leaves boiled in water?”
“Er, yes. With milk.”
“Squirted out of a cow?”
This information was, somehow, too much. The dispenser needed help. It turned to Eddie the shipboard computer (who, by the way, was intelligent enough to pilot the Heart of Gold through infinite improbability and calculate your personality problems to ten decimal places).
The ensuing processes dragged the ship to a halt at a very bad time and nearly led to the deaths of Arthur, Ford, Trillian, Zaphod, and Marvin.
Relevance
“So,” you might be asking, “why do I care? I have tea at home.”
One of the trends in recent years in the world of Analytics, Business Intelligence, Data Science (and so on… you get the point) is the concept of insight generation. You feed your data into the machine/software/model, and it spits out a series of charts and tables and insights based on that data.
The actual technology is pretty clever. The concept itself is somewhat less so.
Companies are paying incredible amounts of money for products that promise to generate ‘insights’ from data without context. These products and processes are admittedly impressive. They can manage incredible amounts of data, structured and unstructured, and produce results that are consistent with the input. That is no small feat.
But, and this is important, they have no way of knowing what matters. I am impressed by their capacity, but that’s all they have going for them. They will inevitably produce a fairly predictable set of results based on internal guidelines that were defined by people who do not know your business needs.
‘Insights’ delivered by a process with no context is precisely as useful as a drink dispenser that does a detailed analysis of your taste buds, neurologic response to stimuli, and metabolic needs before inevitably producing a drink that is almost, but not quite, entirely unlike tea.
The Nutri-Matic drink dispenser, while a remarkable feat of technology, was pointless for its intended purpose.
Even so, when Arthur made a straightforward request for tea, the Nutri-Matic did understand his request. The taste of dried leaves boiled in water.
It just couldn’t produce it. It was never designed for context.
Conclusion
Arthur did get his tea. It required the intervention of the spirit of Zaphod’s dead great-grandfather, Zaphod Beeblebrox the Fourth, but he got it.
Yes, the Fourth. Accident with a time machine and a contraceptive. Stay on topic.
I’m not saying these products cannot produce meaningful insights. They simply produce a large number of insights. When people sort through that mess and find a few useful bits, they call it a success and move on.
It is much easier to consider what you want before trying to generate it.
And if you don’t know which insights you need? Hire a Data Analyst. Preferably not a former employee of the Sirius Cybernetics Corporation.
For more remarkable insights from the Hitchhiker’s Guide to Analytics, visit…
