I was never a contender in Scrabble until Words With Friends allowed me to play asynchronously and my lack of spelling skills could be masked by the private, “…is not an acceptable word” message.
But I do love words.
I’m a big fan of A Word A Day – a short little email that reminds me I am smart (because I know the word), that I am not (when I don’t), and then leaves me feeling better for being more knowledgeable than I was.
Last year, I wrote about consilience (“the linking together of principles from different disciplines especially when forming a comprehensive theory”) in a ClickZ column called “Consilience – The Intrinsic Value of Big Data.”
Great word. I don’t use it much, but it informs a lot of my thinking.
Today’s word, class, is homologous, (huh-MOL-uh-guhs), which A Word A Day defines as an adjective, meaning:
- Exhibiting a degree of correspondence or similarity.
- Corresponding in structure and evolutionary origin, but not necessarily in function.
For example, human arm, dog foreleg, bird wing, and whale flipper are homologous.
Consilience makes all the pieces fall into place to arrive at insight. It’s that ah-ha! moment.
Things that are anomalous stand out and draw our attention as I mentioned in the column “The New Funnel“:
All the reports and benchmarks in the world are useful if – and only if – they raise a flag. When the analytics system is set up properly, it watches your metrics and alerts you to anomalies. It lets you know if something is out of bounds, off the hook, or headed in an unexpected direction.
But we sometimes forget that things that are homologous should attract our attention as well.
When different parts of our marketing efforts are acting too similar and mimic each other too closely numerically speaking, there may be cause for concern. It may be that their similarity is an unlikely anomaly.
If all tests are bringing back the same results…
If every social media promotion generates the same level of interest…
If each mobile ad results in the same behavior…
…then there might be something wrong with data capture, the number crunching, or the reporting.
But the other side of the homologous coin is the one that leads to correlation and, we hope, causation.
If these two customer segments behave a lot alike…
If the response to certain social promotions has the same pick-up and drop-off curves…
If the sales growth rate for this product category has been the same for months and months…
… maybe some money can be saved on chunking them together.
… maybe time-of-day or day-of-week can be leveraged to increase virality.
… maybe it’s time to shake things up a bit.
Sometimes things that don’t go bump in the night should draw your attention. Wouldn’t you be intrigued if you came across two identical snowflakes?