Product #78
Nudge, Group Think and More.
Nudge · Thaler & Sunstein · 2018
Choice architects organise the context in which people make decisions. You can be one without knowing it.Small details can have a major impact on behaviour, so assume everything matters. There is no such thing as a neutral design.
A nudge is when a choice architect alters behaviour in a predictable way without restricting options, or changing economic incentives.
Humans are fallible. We have to make thousands of decisions each day. We can't afford to think deeply about each one, so we take shortcuts. These rules of thumb help, but they can lead us astray, especially in challenging or unfamiliar situations. People are ‘nudgable’ even when making life's most important decisions.
Biases and blunders are how and when people systematically go wrong. There are many: anchoring, availability, representativeness, overconfidence, loss aversion, status quo bias, framing.
We think in two systems, the automatic system (fast and intuitive) and the reflective System (slow and reflective). The automatic system acts rapidly and instinctively, it is associated with oldest parts of the brain. The reflective system is more deliberate and self-conscious. Designing policies that allow people to rely more on their Automatic Systems can improve their lives.
Temptation and mindlessness can contribute to our inconsistency in decision making. We don't forecast self-control problems well because we underestimate the influence of the arousal effect in the moment. Pre-commitments and requirements (seat belts) can help with self-control.
Humans are influenced by other humans, even when they shouldn't be.
We are heavily influenced by others' choices and social norms.
Peer pressure and desire for approval shape our decisions.
We tend to follow established patterns in groups.
Information cascades influence us as other people’s preferences are revealed.
Awareness of others' true thoughts can quickly change our behaviour.
Given we can’t avoid choice architecture we should offer nudges that are most likely to help and least likely to inflict harm.
Nudges are most helpful in situations where choices are difficult, infrequent, require scarce attention, have delayed effects, offer poor feedback, or have an ambiguous relationship between choice and experience.
Checklists are powerful nudges and can empower junior employees to speak up.
Time gaps between choices and consequences cause self control problems. Nudges can help when consequences of choices aren’t immediate:
Investment goods: invest now, get rewards later (exercise, healthy eating)
Temptation goods: enjoy now, face consequences later (smoking, alcohol, Netflix)
If you want to encourage some action or activity, make it easy.
Defaults: If there is a default option, expect a large number of people to end up with that option. If you suggest there’s a normal course of action - people will likely take it. Choose defaults in ways that are self-serving or welfare enhancing. Nudge for good!
Required choice (or mandated choice) is when you remove the default and false a choice. Doing so overcomes inertia, inattention, and procrastination; you can find out what people prefer, without having to guess.
Expect Error. Humans make mistakes, systems should expect error and be as forgiving as possible. Gmail provides prompts to users who mention the word "attachment" but fail to include one. Well-designed systems give feedback, they tell people when they are doing well and when they are making mistakes. E.g. laptops ask us to plug in when low on battery.
Mappings are the relationship between choices and outcomes. Good choice architecture helps people to improve their ability to map choices to outcomes and select options that will make them better off.
When faced with a small number of well-understood alternatives, we’ll examine all the attributes of all the alternatives and then make trade-offs. When faced with many choices, we must use alternative strategies.
Elimination by aspects: we decide on what aspect is most important, establish a cutoff level, then eliminate all options that don’t comply. We then repeat attribute by attribute until either a choice is made or the set is narrowed down enough to switch to another method. BUT options that don’t meet the minimum cutoff level may be eliminated even if they are fabulous on all other dimensions.
Winnow down the choice set to a manageable size if you can.
Help people to learn: so they can later make better choices on their own.
Incentives are important. Ask: Who chooses? Who uses? Who pays? Who profits? Salience is the most important incentive intervention strategy. Make sure choosers notice the incentives they face.
Smart disclosure improves decision making and makes the market more transparent, competitive, and fair. Governments play a role in creating standardised units to facilitate consumer comparisons. Complex information should be disclosed in a format that is easy to understand and in machine-readable formats.
Sludge is using nudging for bad. Any aspect of choice architecture consisting of friction that makes it harder for people to obtain an outcome that will make them better off.
Dark patterns are an assortment of online practices designed to manipulate people. Often they make pricing less transparent.
The unsubscribe trap: creating deliberate asymmetry between the ease of joining and the pain of leaving.
Rebates: A seller offers to return a portion of the sales price to customers, but there’s sludge involved in redeeming the coupon and only 10%-40% will.
Shrouded Attributes: when the headline price of the good understates the true cost to the user because the shrouded attributes, and their costs, are hard to discover. E.g. sell the printer cheap and make the money on the ink.
Competition doesn’t eliminate sludge: ‘Free Bank Accounts’ is better marketing than ‘Bank Accounts, $100 per year, no hidden fees’
Choice architects occasionally need to drop the goal of neutrality and decides to nudge directionally. In saving for retirement humans need help with enrolment, increasing contributions and improving their investment returns.
In the News
At ‘Make with Notion’ the company announces forms, mail, new database layouts, better AI, better template store and more. For those that use Notion everyday, this is exciting · Video
Intercom announce next generation AI agent Fin 2 · Video
Federal Trade Commission announces “Click-to-Cancel” rule making it easier to end recurring subscriptions and memberships. The end of a bunch of dark patterns? · Ruling
Quick Links
AI, Algorithms, and Awful Humans · Paper
21 strategies to maximise user engagement without exploitation · Article
Automating processes with software is hard · Article
When data and anecdotes disagree, the anecdotes are usually right · Article
Apple’s marketing philosophy · Image
Gala on tools to drive smarter decisions · Article
5 tips for staying focused · Article
Group Think · Irving Janis · 1982
Groupthink, a term coined by social psychologist Irving Janis (1972), occurs when a group makes faulty decisions because group pressures lead to a deterioration of “mental efficiency, reality testing, and moral judgment”. Groups affected by groupthink ignore alternatives and tend to take irrational actions that dehumanise other groups. A group is especially vulnerable to groupthink when its members are similar in background, when the group is insulated from outside opinions, and when there are no clear rules for decision making
Eight Symptoms of Group Think:
Illusion of invulnerability: Excessive optimism leading to extreme risks.
Collective rationalization: Discounting warnings, not reconsidering assumptions.
Belief in inherent morality: Ignoring ethical consequences of decisions.
Stereotyped views of out-groups: Dismissing effective conflict responses.
Direct pressure on dissenters: Suppressing arguments against group views.
Self-censorship: Not expressing doubts or deviations from group consensus.
Illusion of unanimity: Assuming majority views are unanimous.
Self-appointed 'mindguards': Protecting group from contradictory information
Remedies to Group Think:
Assign critical evaluator roles: Distribute these roles to all group members.
Leader impartiality: Leaders should avoid stating initial preferences.
External consultation: Members discuss with trusted associates outside the group and report back.
Invite outside experts: Bring in challenging external perspectives to meetings.
Devil's advocate: Appoint someone to question assumptions and play this role.
Scenario analysis: Allocate time to analyse rival intentions and construct various scenarios.
Book Highlights
Status lights are ideal for communicating low-importance, persistent information.
Amber Case · Calm Technology
In most development shops, the most-experienced programmers take responsibility for the most-demanding parts of the program. In return for this effort, they are given some modicum of immunity from having to field annoying technical-support calls. When users of the program call from the field, they are routed to technical-support personnel or to more junior programmers. On the rare occasion that a user gets through to the senior coder, it is because that user has already demonstrated her expertise to the junior programmer or the tech-support person. The result of this filtering process is that the more senior programmers are, the less contact they have with typical, run-of-the-mill users. By extension, they mistakenly assume that "their" users are representative.
Alan Cooper · The Inmates Are Running the Asylum
Model development is an iterative process. After each iteration, you’ll want to compare your model’s performance against its performance in previous iterations and evaluate how suitable this iteration is for production.
Chip Huyen · Designing Machine Learning Systems
Remember our definition of management? A manager’s job is to get better outcomes from a group of people working together through influencing purpose, people, and process.
Julie Zhuo · The Making of a Manager
Quotes & Tweets
As far as I can tell, one of the biggest changes across organisations over the past few years is simply the rise of distraction. The default often appears to be a kind of continuous partial attention. I'm not sure whether it's good or bad. Maybe people were sub-optimally stuck before, and perhaps there are gains to being able to allocate attention more flexibly? Or maybe it's bad because meetings become even less efficient and our ability to focus (already under assault!) withers further. Either way, it's a striking new normal.
Patrick Collison
Only 54% of Americans have read a single book this year.
Andrew Chen