Magic Number for Digital Analysts: 5±2

I got an interesting question from a student: I mention 5 ±2 as a reference to the “magic number 7work by cognitive psychologist George A. Miller in 1956.

The number “7” is often stated as a universal truth without providing the specific context of the experiment… A curse faced by many analysts!

Recent research challenge this notion. For example, the 2012 article “Four is the ‘magic’ number,” on the ScienceDaily says:

“According to psychological lore, when it comes to items of information the mind can cope with before confusion sets in, the “magic” number is seven. But a new analysis challenges this long-held view, suggesting the number might actually be four.”

Maybe because we’re not as smart as our parents and grandparents and we can only remember 4! In fact, no, that’s not the reason…

As an explanation for the magic number 7, we often hear the reference to phone numbers. But this newer study gives the following counter-argument:

“So to remember a seven numeral phone number, say 6458937, we need to break it into four chunks: 64. 58. 93. 7. Basically four is the limit to our perception.”

It’s interesting to note this breakdown of 64. 58. 93. 7. is common in Europe, while in North America we would probably say 645. 8937. See — still a max chunk of four digits. Even with an area code, we would say 486. 645. 8937.

But more seriously, whether it’s 7, 5 or 4, the important point to remember is to ask ourselves if we can reduce and optimize the volume of information we have to deal with. In our field of expertise as digital analysts and marketing specialists, this means thinking very hard about the ideal number of KPIs on a dashboard, the optimal items in our web or apps interfaces, the number of concurrent tasks we can deal with at a given time.

Personally, I have found 5 ± 2 was applicable in so many situations — as Avinash Kaushik once told me after making a bold statement on stage — “I use this number to make a point that sticks.” You should always exercise judgment when someone — whomever it might be — makes a bold statement!

“I use this number to make a point that sticks” — Avinash Kaushik

Here are some additional examples to think about:

  • In marketing, you have the “decoy effect” which uses 3 elements;
  • In user interface design, famous (and sometimes controversial) web usability consultant Jakob Nielsen stated, “Human attention is limited and we are only capable of maintaining around five items in our short-term memory at one time.”;
  • One of the frequent question about dashboard is “how many KPIs should I have?” In the book “Key Performance Indicators”, author David Parmenter says “an organization with over 500 FTE will have about 10 KPIs”, and later add “you may have considerably less”. I try to aim for 5±2 KPIs, with their supporting metrics.
  • In our analysis work, a junior analyst should be able to deal with 3 analysis projects concurrently because each one will be at a different stage of the DMAIC analysis process, might have different levels of complexity, and are likely to be about different topics. Even as an experienced analyst, dealing with more than 7 clients/projects/tasks become unproductive, unreasonable and exhausting (as I’m dealing with these days!). Going beyond 7 tasks is unrealistic as the “switching cost” becomes too high. There are numerous articles about the cost of multitasking.

What do you think of this magic number 5±2 ? I would love to read your thoughts!

Stéphane Hamel is a seasoned independent consultant, teacher and speaker. He shares his passion for digital analytics — be it technical ‘how to’ or assessing organizations’ digital capabilities and maturity.

If you enjoyed this article, you should follow me on Facebook or Twitter, and while you’re at it, why not click the nice little clapping hands on the left!

Featured photo credit: Lindsey McCutchan Photography rlmccutchan on Flickr/Instagram @lindseymccutchanphotography, with permission.

All the world is made of faith, and trust, and pixie dust. Digital marketer & analyst with a strong interest in privacy and the ethical use of data.