Introduction to nudging in IT
- 1. Introduction to behavioral economics in IT - When common sense isn’t enough Christian F. Nissen, CFN Consult RESILIATM, ITIL®, PRINCE2® MSP®, MoP® and MoV® are Registered Trade Marks of AXELOS in the United Kingdom and other countries COBIT® is a registered trademark of the Information Systems Audit and Control Association (ISACA) and the IT Governance Institute (ITGI) TOGAFTM and IT4ITTM are trademarks of The Open Group SIAM® is a registered trademark of EXIN © 2018 of CFN Consult unless otherwise stated
- 2. 2 Agenda 1. Introduction to behavioral economics 2. Nudging in IT 3. From process architect to behavioral architect 4. Conclusion © 2018 Agenda
- 3. 3 Can we trust our intuition? Intuitively, which of the black circles is the largest one? © 2018 Behavioraleconomics
- 4. 4 Can we trust our intuition? What colour do the middle fields have on the cube? © 2018 Behavioraleconomics
- 5. 5 Can we trust our intuition? © 2018 Behavioraleconomics Which table is the longest?
- 6. Rational economics Traditional (neo-classical) theories of economic behavior assume that economic agents apply rational thought to each and every decision to achieve the maximization of personal benefit (utility) or, in the case of producers, the maximization of profits. Rational man (Homo economicus) The rational man or economic man (econ, John Stuart Mill) acts to obtain the highest possible well-being for him given available information about opportunities and constraints on his ability to achieve his predetermined goals. He is fully informed of all circumstances impinging on his decisions and possesses unlimited (brain) capacity to process the information when he makes decisions. Behavioral economics A science that studies how individuals make economic decisions in practice. Behavioral economics attempts to understand the effect of individual psychological processes, including emotions, norms, and habits on individual decision-making in a variety of economic contexts. Human Unlike the rational man, humans sometimes are irrational and predictably err 6 Rational vs behavioral economics © 2018 Behavioraleconomics
- 7. 7 Behavioral economics Selected titles: © 2018 Behavioraleconomics
- 8. 8 We ask to be deceived. . . © 2018 Behavioraleconomics
- 9. 9 Our intuition cheats on us Intuitive and logical errors When we must respond quickly, our intuition tends to take over and override our logical thinking. Example: In a lake, there is a patch of lily pads. Every day, the patch doubles in size. If it takes 48 days for the patch to cover the entire lake, how long would it take for the patch to cover half of the lake? Neglect of probability We overestimate the probabilities of unlikely events and underestimate frequent events. Example: In Denmark, the probability of being struck by lightning is many times greater than that of being hit by terror. Post rationalization We have a well developed ability to find causes and explanations of phenomena that actually are incidental. Example: It turns out that over time luck has greater impact on the success of an average business than the strategy and management team © 2018 Behavioraleconomics
- 10. 10 Our intuition cheats on us Loss aversion Losses bite more than equivalent gains. The disutility of giving up an object is greater than the utility associated with acquiring it. Example: You are offered a gamble on the toss of a coin. If the coin show tails, you loose € 1,000. If the coin shows heads, you win €1,500. Would you accept the game? Mental accounting We may have multiple mental accounts for the same kind of resource Example: We may not enter the time and gasoline we use to run after a special offer on the same mental account as the offer itself. Herd behavior The tendency to do or believe things because many other people do or believe the same. Example: Fashion, likes, number of downloads, menu choice at the restaurant © 2018 Behavioraleconomics
- 11. 11 Our intuition cheats on us Framing We draw different conclusions from the same information, depending on how that information is presented Example: ”90% of the patients who have this operation, are alive after 10 years” versus ”Of 100 patients who have this operation, 10 are dead after 10 years” Menu dependence The choices we make are often highly dependent on the choices available in a certain context, i.e. what alternatives a particular option is presented with. Example: Ballots have to list candidates in some order. One study found that a candidate whose name is listed first gains about 3.5 percent points in the voting. © 2018 Behavioraleconomics
- 12. 12 Our intuition cheats on us Default bias To avoid the discomfort of complex choices, we usually opt for the default supplied to us. Example: The rate of organ donation in a country is highly dependent on the default option. The high-donation countries have an opt-out form. The low- contribution countries have an opt-in form. Availability We overestimate available information and ignore absent information. Including recent experience, access to information, clarity and convenience. Example: A professor at UCLA asked different groups of students to list ways to improve the course, and he varied the required number of improvements. The students who listed more ways to improve the class rated it higher. © 2018 Behavioraleconomics
- 13. 13 Our intuition cheats on us Anchoring and adjustment We start with an anchor, e.g. a number we know, and adjust in the direction we think is appropriate. The bias occurs because the adjustments are typically insufficient. Example: The subjective difference between €99,900 and €100,000 is perceived less than between €100 and €200. Confirmation bias We tend to search for, interpret, focus on and remember information in a way that confirms our preconceptions Example: An employer who believes that a job applicant is highly intelligent may pay attention to only information that is consistent with the belief that the job applicant is highly intelligent © 2018 Behavioraleconomics
- 14. 14 Our intuition cheats on us Optimism bias The tendency to be over-optimistic, overestimating favorable and pleasing outcomes Example: Planning fallacy. We view the world in a more positive light than is justifiable, we overestimate our own abilities, and we consider the goals we set, easier to achieve than they probably are. We are often driven by the desire to get our plans approved, and therefore our estimates are closer to a best-case scenario than to a realistic assessment © 2018 Behavioraleconomics
- 15. 15 Our intuition cheats on us MINDSPACE The UK Institute for Government has summarized our biases in the abbreviation MINDSPACE: Messenger: We are heavily influenced by who communicates information Incentives: Our responses to incentives are shaped by predictable mental shortcuts, such as strongly avoiding losses Norms: We are strongly influenced by what others do. Defaults: We ‘go with the flow’ of pre-set options Salience : Our attention is drawn to what is novel and seems relevant to us Priming: Our acts are often influenced by subconscious cues Affect: Our emotional associations can powerfully shape our actions Commitments: We seek to be consistent with our public promises, and reciprocate acts Ego: We act in ways that make us feel better about ourselves © 2018 Behavioraleconomics
- 16. 16 And much more cognitive biases . . . © 2018 Behavioraleconomics Source: https://en.wikipedia.org/wiki/List_of_cognitive_biases
- 17. The intuitive/automatic system Uncontrolled Effortless Associative (similarity, proximity, causality) Fast Unconscious Skilled The rational/reflective system (Self)Controlled Effortful Deductive Slow Self-aware Rule-following 17 Two systems © 2018 Behavioraleconomics We are predictably irrational When we are faced with incomplete information, when consequences are uncertain and when we have to make fast decisions, we fall back on simple rules of thumb (heuristics) and educated guesses. They are essential to prevent mental meltdown, but they also regularly leads to systematic errors.
- 18. 18 Influencing behavior © 2018 Behavioraleconomics The rational system (slow) The intuitive system (fast) Rational behavior Instinctive and trained behavior Education Knowledge / information / data Nudge Training Disruptions Convince / Post rationalize Manipulate / Stimulate
- 19. 19 Difference natures of practice © 2018 Behavioraleconomics David Snowden, 2002, 2007, 2014 Disorder_ Complex Probe Sense Respond Emergent practice . Complicated Sense Analyze Respond Good practice Chaotic Act Sense Respond Novel practice Simple/obvious Sense Categorize Respond Best practice Complacency Relationship between cause and effect is obvious. Predictability, routine, established practice, entrained thinking. Encourage simplification. Relationship between cause and effect requires investigation and analysis. Unique, non-repeated, multiple right answers, experience/expertise. Encourage analysis. Relationship between cause and effect is impossible to determine. Unpredictability, turbulence, trial and error, responsiveness. Encourage action Relationship between cause and effect can only be perceived in retrospect. Unpredictability, flux, emergence, creativity. Encourage experimentation
- 20. 20 Different natures of practice © 2018 Behavioraleconomics The rational system (slow) The intuitive system (fast) Rational behavior Instinctive and trained behavior Education Knowledge / information / data Nudge Training Disruptions Convince / Post rationalize Manipulate / Stimulate Complex practice Simple practice Complicated practice
- 21. 21 Different natures of knowledge transfer © 2018 Behavioraleconomics Tacit knowledge Tacit knowledge Tacit knowledge Socialisation Externalisation Explicit knowledge Tacit knowledge Internalisation Combination Explicit knowledge Explicit knowledge Explicit knowledge Ikujiro Nonaka & Hirotaka Takeuchi, The Knowledge Creating Company, 1995
- 22. 22 Different natures of knowledge transfer © 2018 Behavioraleconomics The rational system (slow) The intuitive system (fast) Rational behavior Instinctive and trained behavior Education Knowledge / information / data Nudge Training Disruptions Convince / Post rationalize Manipulate / Stimulate Socialization Internalization Externalization Combination
- 23. 23 The limited human The modern workplace has sent the rational system on overtime, leading to stress: We are not rational We don’t have a strong will We do not have infinite cognitive capacity We are not self-managed Humans, unlike Econs, need interventions to make good decisions, and there are informed and un-intrusive ways to provide that help (libertarian paternalism). © 2018 Nudging
- 24. 24 Nudging Sometimes a nudge is needed Nudging explained by Richard H. Thaler © 2018 Nudging
- 25. 25 Nudging The concept nudge comes from elephants crossing the savannah. If one of the baby elephants heads in the wrong direction, it gets a friendly nudge of the older elephants, to help it in the right direction. Nudge can thus be described as a friendly push in the right direction. A definition Any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives. To count as a mere nudge, the intervention must be easy and cheap to avoid. Nudges are not mandates. Good nudges must be transparent and never misleading be easy and cheap to avoid improve the welfare of those who are being nudged © 2018 Nudging
- 26. 26 A good and a more questionable example The World's Deepest Bin Social ad deters pedestrians from crossing at red lights © 2018 Nudging
- 27. 27 Nudges – ITSM examples Neglect of probability State how many times a trivial change resulted in adverse incidents over the past 12 months Begin with end mind Gather all stakeholders and celebrate the completion of a release with cake Loss aversion Introduce levels of authority in the ITSM tool, that may be lost if your quality decreases Mental accounting Introduce an artificial currency for estimation and resource management - for example checkers, function points or story points Divide resources in pools (e.g. development, maintenance, error correction). When a pool is empty, there are no more resources for that specific type of task without management approval © 2018 Nudging
- 28. 28 Nudges – ITSM examples Herd behavior Indicate how many times a knowledge article has been used Number of ’likes’ on workarounds Inform what x other / x% typically have done in a similar situation Framing Use alarms / notifications from real life (traffic lights, stop signs, etc.) Make physical changes when the work situation changes (rooms, draw a circle, physical staging, war room, standing meetings, daily morning meetings, kaizen meetings, coffee tables, etc.) Use Kanban boards (signaling) Introduce a baton for major incidents - the person with the baton is accountable for the management of the major incident Show a picture of the person who will receive the case after you, to make you more diligent with the details © 2018 Nudging Remind the participants what is a Nudge for good: It is a smart and simple initiative that influences consumers’ behaviour in order to help them achieve their own goals It is ethically designed (means-end goal / legitimate originator) It acts in favour of people’s own interest and that of the community (or the planet) It preserves freedom of choice and existing options It is based on observational insights of individuals, recorded in their local environment and community It leverages unconventional factors revealed by Behavioural Economics, neuroscience and cognitive psychology, along with more conventional concepts (education, information, marketing, communication…) It uses a creative re-design of some situations and interaction points (including branded touchpoints) It doesn’t activate any economic incentive: you shouldn’t pay people to change their behaviours, although you may offer them a symbolic reward The output here is to pinpoint which ideas are potential Golden/Revolution Nudges! NUDGING FO R GO O D 9
- 29. 29 Nudges – ITSM examples Menu dependence Consciously order of the services or categories, you have to assign to a case – e.g. by placing the services or categories that are often neglected at the top Use the "frequency of use", "likes", "added date”, "last updated" etc. as sort criteria in ITSM tool lists rather than alphabetical order Default Priority is by default set to 3 © 2018 Nudging
- 30. 30 Nudges – ITSM examples Availability When creating a new change, highlight the last 10 changes that went really wrong Traverse the configuration management system to draw attention to possible impacts Every employee from respectively 1st and 2nd level must minimum twice every week contact the other function with minimum one positive feedback and minimum improvement item. 1st level keeps records Support with statistics and artificial intelligence – e.g. this change is similar to these five previously implemented changes Automatically notify if there is no attachment to a mail containing the word ’enclosed', 'attachment' etc. Ease the access to help: "If you are in doubt about how this field is to be filled, then call 12 34 56 78" or "Watch this video on how to fill the field" Set up the best coffee machines where you want people to meet © 2018 Nudging
- 31. 31 Nudges – ITSM examples Anchoring and adjustment The completion of one field can prime the user in filling in a subsequent field "The average development estimate for similar changes is nnn hours" Status quo First month is free, then you pay a monthly subscription unless you cancel your subscription Rules of thumb "You need to estimate the same amount of time for testing and training, as for the design, development and deployment” Representativeness "Have you checked these similar cases: # 1, # 2, # 3” Confirmation bias "Before you assign the incident to the network department, we want you to consider once again whether there is another department, that might be more obvious" © 2018 Nudging
- 32. 32 Nudges – ITSM examples Reward long-term gain Give more and more freedom in the tools, the more experienced a person gets Social contracts "Dear Colleague: If you register your change here, we take full responsibility from here." Applies both to technicians and customers "Avoid extra work: Be aware that you avoid ??? if you have ???" or "Be aware that you are prompted for ??? if you have not ???" "I hereby declare that I have attached the following documents: a, b” "You now only have eight fields left" (The implication: Hold on and continue to work carefully) © 2018 Nudging
- 33. 33 Nudges – ITSM examples Gamification Use of gaming elements: Point, badges, scoreboards, progress indicators, levels, rewards, challenges, etc. Assign points for contacts with the user during the lifecycle of a case Earn points through work (e.g. number of cases weighted by complexity) Credits for the number cases you have related to other cases in Service Desk Rating of cases you receive from other groups User satisfaction team score boards © 2018 Nudging
- 34. 34 Nudges – ITSM examples DataCenter 2000 The physical work was arranged into four zones: Customer zone Support zone Operations zone Maintenance zone Each zone had its own unique office environment. A workplace in DataCenter 2000 was not a place you owned, but a place you had to move to for a certain type of work. The surroundings supported the work method of the individual zone. In the support zone, as we see here, all workplaces for example supported rotation. Furthermore, there was big screens were mounted to give a status the of operations and support situation. The workplaces were designed to encouraged dialogue and knowledge sharing. © 2018 Nudging
- 35. 35 Nudging Use 5 minutes with your neighbor to suggest additional examples of IT service management nudges © 2018 Nudging
- 36. 36 Bureaucracy ”Because adherence to standard operating procedures is difficult to second-guess, decision makers who expect to have their decisions scrutinized with hindsight are driven to bureaucratic solutions – and to an extreme reluctance to take risks.” Kahnemann © 2018 Fromprocesstobehavioralarchitect
- 37. 37 When to nudge? People will need nudges for decisions that are difficult and rare, for which they do not get prompt feedback, and when they have trouble translating aspects of the situation into terms that they can easily understand. Situations where people are least likely to make good choices: benefits now - costs later (situations that test my capacity for self-control) degree of difficulty frequency - first time no or bad feedback not knowing what you like © 2018 Fromprocesstobehavioralarchitect
- 38. 38 How to nudge? iNcentives While humans respond to nudges, they also respond to incentives. Make sure users have the right incentives. Make the incentives salient (or prominent) so that people don’t miss them Understand mappings People make better choices when they have help in understanding what the various choices means in terms of their welfare (health, satisfaction, happiness, and well being) Defaults – Padding the paths of least resistance Every choice situation has a default choice, whether it is made explicit or not. The default is what you get when you choose nothing Give feedback The best way to help Humans improve their performance is to provide feedback. Well-designed systems tell people when they are doing well and when they are making mistakes. Expect error Humans makes mistakes. A well-designed system expects its users to err and is as forgiving as possible. Structure complex choices When people need to choose one item from a long list and evaluate each item by another long list of criteria, people need structuring of choices, e.g. elimination by aspects © 2018 Fromprocesstobehavioralarchitect
- 39. 39 How to nudge? 1. Map Describe the desired behavior. Be specific, both in terms of desired behavior and expected results. Use the 'video test'! Observe and describe current behavior. How does it differ specifically from the desired behavior and what are the consequences? Describe the involved actors. Who is closest to the desired behavior and who is furthest from? Are there actors who try to achieve the desired behavior, but do not achieve it? Are there patterns (persona / archetypes)? Describe the desired behavioral change. Describe both the desired change and how it can be measured. 2. Analyze Identify the nature of irrational behavior (biases). What is the reason for the undesired behavior? Identify barriers (physical, mental and social) for the desired behavior and triggers of irrational behavior. Ask and observe. © 2018 Fromprocesstobehavioralarchitect
- 40. 40 How to nudge? 3. Design Brainstorm on specific nudges. How can the desired behavior be activated? Start with the nature of irrational behavior. Take advantage of your knowledge about bias, barriers and triggers. Identify any ethical challenges of the identified nudges Select the most promising nudges with the least unwanted side effects (risk) 4. Evaluate Test for effect and unwanted side effects. Use prototypes, test on different subjects, test in the 'laboratory’ as wells as in the field. Compare achieved behavior (test group) with current behavior (control group). Optimize if possible. 5. Implement Deploy Monitor and follow-up © 2018 Fromprocesstobehavioralarchitect
- 41. 41 How to nudge? © 2018 Fromprocesstobehavioralarchitect 1. Map 2.Analyze 3. Design 4.Evaluate–5.Implement
- 42. 42 The 21 drivers of influence © 2018 Fromprocesstobehavioralarchitect THE 21 DRIVERS O F INFLUENCE RANSMITTERT ABITSH GOE EFAULTD ECIPROCITYR NCENTIVESI ALUEV MOTIONE EWARDSR ALIENCES Choose the right messenger to reinforce the message Promote the development of new habits through new triggers and appropriate rewards Give value to the action by rewarding with recognition Encourage a behaviour with money, goodies or social reward Create a default choice sequence that leads to the desired behaviour Engage in a logic of reciprocity by creating a social debt Generate an emotional response through images, visuals and embodied stories Highlight the scope of work realised to justify the price Reward with positive feedback Attract attention by making an item salient O RAMINGF MMEDIACY I OSTALGIAN AIRNESSF OSSAVERSIONL PPER/ LOWER ANCHORINGU ASINESSE ORMSN OMPARTMENTALIZEC NGAGEMENTE Create a choice context that encourages the desired behaviour Make the task seem easier by going on step at time Generate favourable associations and reference points Activate the feeling of nostalgia Show the fairness of the targeted behaviour Mention the loss provoked by not adopting the desired behaviour Encourage the need for conformity Simplify the desired behaviour Postpone constraints and efforts in the future while immediately providing advantages Materialise invisible flows (like efforts, expenses) with objects you can visually handle Generate commitments to promote consistency NESTEPATA TIME 7 Eric Singler, Nudge Marketing – Winning a behavioral change, 2015
- 43. 43 Conclusion © 2018 “Don’t build a machine and expect it to behave with any intelligence or creativity. Don’t employ people and then try and make them behave like machines.” - Rob England Conclusion
- 45. Contact 45 Christian F. Nissen [email protected] +45 40 19 41 45 CFN Consult ApS Linde Allé 1 DK-2600 Glostrup CVR: 39 36 47 86 © 2018