In this issue we look at the ethical responsibilities of AI programming, the cognitive traps of hindsight bias, A/B testing and more.
The main thing
McKinsey’s developer productivity report continues to cause some uproar. People working in product development know that any metrics like number of designs produced or the number of lines of code written are pointless because they don't factor in quality and can actually lead to negative behaviours. So why do companies keep trying to re-implement them?
With manual work, you can correlate the work someone did with the output they created so individual productivity measurements are straight-forward. But those jobs are typically repeatable processes, like on an assembly line whereas product development is a creative process that is always building something new. The corporate pressure for standardisation, predictability, and productivity metrics hurts creative processes because no variability means no innovation.
So what can we do?
From individuals to teams
We shift productivity from the individual to the team level. Break down the silos in the development process and treat the end-to-end process as a single entity. Get the team excited about the product they are building. Give them the context to know what problems to solve, and ambitious targets to hit. Implement DORA metrics for tracking the ability of a team to deliver code to production. And when teams can finally deliver to production autonomously, hold them accountable for their results.
There will still be individuals who aren’t as skilled or motivated, but the team knows who is performing much better than the managers. If a team doesn’t hit its targets it should be disbanded. This outcome creates a motivation for team members to hold each other accountable.
Change the system, not the people
Ever noticed how some people are rockstars in one team and underperforming in another? A bad system beats a good person every time. We need to change the system itself to motivate teams, enable them to deliver autonomously and hold them accountable. But this is hard; requires wide-ranging organisational changes and is slow to implement.
If that hard work doesn’t sound appealing, we could just create a new individual productivity metric and pretend it works?
The (subtle) plug
From June 2025 on the European Accessibility Act (EAA) will be enforced within the EU member states. That means that websites must comply with the Web Content Accessibility Guidelines (WCAG) guidelines level AA. Let's just say, there are a lot of websites that do not meet these standards at the moment!
One particularly tough area is around data visualisations. Fortunately though Kent Eisenhuth, Staff UX Designer at Google has been tackling this challenge for the past few years and he is sharing his insights at UXDX EMEA in four weeks time. This is a talk that you do not want to miss.
This weeks updates
Enabling the Team
Productivity Is About Your Systems, Not Your People by Daniel Markovitz
Daniel argues that productivity depends not only on individual efforts, but on organisational systems and recommends four ways to improve them.
Measuring Developer Productivity? A Response To McKinsey, Part 2 by Gergely Orosz and Kent Beck
A continuation of the response from Gergely and Kent on McKinsey's methodology for measuring developer productivity.
Product Direction
The Human Decisions That Shape Generative AI: Who Is Accountable For What? by Stefan Baack
Find out how Stefan discusses the human influence on generative AI, focusing on responsibility and ethical considerations in its development.
Hooked: How To Improve Mobile App Engagement (+ Examples) by Euan Harkness
Explore how industry leaders like Instagram & Duolingo have used user testing successfully to improve mobile app engagement.
Continuous Discovery
Learning to Interview Continuously by Melissa Suzuno
A case study on how Welena Villasoto, has adopted continuous interviewing practices to better understand and serve her community.
Hindsight Bias: Why You Make Terrible Life Choices by Nir Eyal
Nir discusses the cognitive traps of hindsight bias, explains how it is affecting your decision-making and offers strategies to mitigate it.
Continuous Design
The Ultimate Guide To A/B Testing by Lenny's Podcast
A podcast discussion between Lenny and Ronny Kohavi talking about the importance of a/b testing, sharing his experiences at Microsoft and Airbnb.
Leaning into Business as In-House Designers by Kyle Lambert
As designers, we love storytelling and speak in the narrative. But if our audience responds to numbers and facts, our pitch may not be persuasive.
Continuous Delivery
AI, ML, and Data Engineering InfoQ Trends Report - September 2023 by Roland Meertens and Srini Penchikala
This annual report looks at the state of AI, ML and data engineering sharing the key trends that are taking place in the industry.
Don’t Build Microservices, Pursue Loose Couplingby Chris Bertinato.
Teams that embrace loose coupling are better equipped to respond to challenges, gain a competitive advantage and adapt to a constantly changing tech landscape.
Want to get involved?
Attend
Speak
Do you have a great case study that highlights how you have adopted a new practice, tool or technique?
Do you want to share the challenges as well as the successes?
UXDX
13A Belgrave Road, Rathmines, Dublin D06 E9H9, Ireland
You've received it because you've subscribed to our newsletter.
|