Future Work/Life is a weekly newsletter that casts a positive eye to the future. I bring you interesting stories and articles, analyse industry trends and offer tips on designing a better work/life. If you enjoy reading it, please share it!
Sometimes it takes a while to change your mind about something that you've not only put into practice but broadcast to the world. Not on this occasion, though. It only took Will Page about two minutes to convince me that NPS (net-promoter-score) is fundamentally flawed.
Last week, I spent a wonderful hour chatting to Page, until recently Spotify's Chief Economist and my first guest on the next series of the podcast (episode to be released during the first week of May). We discussed all sorts, including the economics of the music business, the different lifecycles of a scaling company, Straight No Chaser magazine, NFTs (non-fungible tokens), and the danger of NPS.
I've written plenty about the surface-level issues within organisations related to work/life design and wellbeing - see recent articles about burnout and pacing yourself. However, the primary obstacle to fixing problems like these remains, for many companies, defining 'good work'. In January, I attempted to explain how businesses could improve their approach to flexible work, with one suggestion involving using the (now) infamous NPS score to measure customer and employee experience.
So what's the issue?
As Will Page wrote in his new book, Tarzan Economics;
The NPS score ticks all the boxes of a 'grab and go' business metric, as it's easy to measure and even easier to grasp at a glance, which makes it very useful when you're looking to bolster egos and play that other three-letter acronym game (CYA, cover your ass) that's common in hierarchical organisational turf wars. By this point, you shouldn't be surprised to learn that a single number that claims to accurately represent the health of an organisation is going to cause more problems than it solves. Yet NPS fosters the same quantification fallacy of encouraging leaders to ask no further questions, which risks casualties.
The big issues here relate to context and trust.
In an attempt at simplicity, NPS represents so narrow a view on customer sentiment that it's essentially worthless. It's impossible to capture such a range of human psychology and behaviour in a single number and provides little to no context into the customer's experience. Plus, Will Page argues that its ubiquity means we, as customers, understand how to game the system - responding at either end of the eleven-point scale to ensure we factor in the calculation.
In short, you can't trust it to tell you anything useful.
Trust remains a constant challenge for anyone attempting to use data to make decisions in businesses. To refer back to the articles on the evolution of flexible work once more, this is why the oft misquoted Peter Drucker said:
What gets measured gets managed — even when it's pointless to measure and manage it, and even if it harms the purpose of the organisation to do so.
In this respect, even worse than not collecting data is collecting it entirely wrongly.
Reflecting on the subject over the past few days, along with reading a timely anecdote in Tim Harford's book, triggered a recurrence of my recently suppressed penchant for German culture references (which any original subscribers to Future Work/Life will remember from its' early days). In the form, on this occasion, of thermometers, armpits and a hirsute nineteenth-century scientist from Baden-Württemberg.
For 18 years, Dr Carl Wunderlich collected over a million body temperature measurements taken from more than 25,000 patients.
A MILLION measurements recorded and filed on paper.
This incredible effort led to his discovery that the average body temperature for humans is 98.6 degrees Fahrenheit - this became conventional wisdom, which seems entirely fair given the sheer scale of the undertaking.
The thing is, Wunderlich was wrong. Our bodies are, in fact, about half a degree cooler than 98.6 degrees. But it took us almost a century to realise it. So the question is, for a dataset so large, how did he get it so wrong?
Upon recent inspection of the thermometers, which now live in a medical museum, we found that they were miscalibrated by around 4 degrees Fahrenheit.
But this wasn't the only mistake. While there are a few parts of the body that we now consider an accurate reflection of our body's temperature, armpits are not one of them and guess where Wunderlich always stuck his thermometers?
The lesson here, of course, is that unless you collect data accurately, you can't trust what it's telling you.
In the absence of NPS, how should we think about measurement, whether it relates to customer or employee experience (eNPS)? As I said in the last newsletter, there isn't a simple answer. As food for thought, though, perhaps we could contemplate a counterweight to 'big data' and, as Drucker alluded to, the problem of quantification bias.
In Tarzan Economics, Will Page introduces the work of Tricia Wang, a tech-ethnographer, who has reframed this by introducing a new term to illustrate why "people are the data and not the metrics they produce".
'Thick data' is at the opposite end of the spectrum and aims to capture the most direct, unmediated data from humans and the full context of their emotions and stories by - wait for it – meeting them. This is such a wonderful phrase to counter the hype around ‘big data’. On one side, we've got snake oil merchants selling promises based on millions of points of data; on the other side, we've got deep, patient observations of real humans, listening and talking to each other.
A good point, well made.
Have a lovely weekend.
Ollie
More on flexible work, this time from George Anders, Senior Editor at Large for LinkedIn - How the pandemic changed us: Our fastest rising priority is job flexibility.
You Only Live Once. That’s the attitude for a rising number of people for whom the pandemic has fundamentally changed their outlook of work and life. Read about the YOLO economy in The New York Times (behind a paywall, but you get some free articles every month).
Li Jin is consistently thoughtful and insightful on the growth of the Creator Economy. In this article, she and Lila Shroff put forward a case for Universal Creative Income - ‘In the digital world, user rights are civic rights, and creator rights are worker rights’. Interesting idea.
If you haven’t signed up for Andy Spence’s Workforce Futurist newsletter yet, then check it out. In this week’s edition, he writes about the future of HR, but I’d also suggest digging through his ‘back catalogue’ to read his articles about the decentralisation workforce and tokenisation of work.
And finally, as I’ve mentioned Straight No Chaser, here’s an article that wouldn’t feel out of place in that publication from this week’s Guardian. Jesse McCarthy writes about how moving to Paris in 1992 led to him discovering a love for French Hip-Hop - a great piece covering music, culture and history.