(p41) User habits are a kind of corporate competitiveness
User habits can be called a kind of corporate competitiveness. (...)
Above all, you have to break the long-standing habits consumers have practiced, and all the more so if most users have continuously been using a competitor's product.
And through this book I came to know Professor John Gourville of Harvard Business School, who teaches marketing, and looked up related material. Among them, the most famous sentence is the one below, also introduced in the book.
"Many innovative products fail in the market because
consumers irrationally overvalue existing products,
and companies irrationally overvalue their new products."
And there is a post that introduces his theory in detail — that for new entrants to seize a chance at success, they must offer not a merely better product but one about 9x more outstanding. That theory is The 9X Effect.
https://www.intercom.com/blog/overcoming-customer-inertia/
I also share his 2006 Harvard Business Review post on understanding the psychology of new-product adoption and a behavioral framework for overcoming it.
https://hbr.org/2006/06/eager-sellers-and-stony-buyers-understanding-the-psychology-of-new-product-adoption
(p49) Two factors that create user habits
When companies judge their product's potential to form user habits, they can use two factors. One is frequency (how often that behavior occurs); the other is perceived utility (the degree users perceive its utility over competitors' products).
Behaviors that occur with sufficient frequency and whose utility is perceived enter the "zone of habit" and become default actions. If either factor is lacking and the behavior stays below the threshold, it is unlikely to become a habit.
(52) Products equivalent to painkillers reduce specific pain, address clearly surfaced needs, and have a quantifiable market. Think of Tylenol — a perfect solution for a problem people are willing to pay for. Conversely, products equivalent to vitamins do not necessarily solve a clearly surfaced problem, so they often appeal to the user's psychological side rather than functional. We take a multivitamin every morning, but we cannot be sure it actually makes us healthier. Some studies even show multivitamins may do more harm than good.
"If you feel a little pain when you do not perform some behavior, that behavior can be considered a habit."
Designing a Reward System
148
Designing a Reward System 1. Cash rewards do not last
Mahalo.com's management believed that if they paid users cash, they would repeatedly participate in activity on their site — who in the world dislikes money? Unfortunately, however, Mahalo.com did not properly understand what drives user participation.
First, the person submitting the question posts a prize in the form of a virtual currency called "Mahalo Dollars." Then other users write answers, and whoever is chosen as the best answer receives a prize exchangeable for real currency. Mahalo.com's founders firmly believed cash rewards would drive user participation and form new online user habits.
In the end, Mahalo.com learned that the reason people participate is not to earn money. If the trigger that drew people to the site had been the desire for cash rewards, Mahalo users would have spent more time there for an hourly wage. And had those prizes taken a game-like slot-machine form, reward frequency would have been too low and amounts too small to matter.
Quora's success showed that reinforcing recognition and praise from other users — social rewards — is a far more important motivator in raising participation frequency. Quora introduced voting to indicate satisfaction with answers and continuously provided feedback. This proved Quora's social rewards were more attractive than Mahalo's monetary rewards.
You must precisely understand what users actually value in order to properly pair variable rewards with the intended behaviors.
In many success stories you frequently find gamification (actively introducing game-related elements into situations far from gaming). Points, badges, and leaderboards can be effective, but only when they scratch a real itch for users. If the customer's problem and the company's proposed solution don't match, adding any number of game elements cannot drive user participation. Likewise, if the user's desire does not persist — for example, if they enter a site for the first time and find no special appeal, feeling no need for repeat visits — gamification is bound to fail. Because the original interest in that specific product or service has vanished. In other words, gamification is not a cure-all for driving user participation.
Variable rewards are not magic dust that a product designer can sprinkle on the spot to make a product more attractive. Rewards can succeed only when they align with the reason for using the product and harmonize with the user's internal triggers and motivations.
150
Designing a Reward System 2. Preserve autonomy
Quora is a case of well-linking appropriate rewards to the behavior of asking and answering. But in August 2012, the company made a huge mistake. This incident taught another key point to consider when using variable rewards.
To boost participation, Quora introduced a feature called "Views," which identified who had seen a particular question or answer. This feedback — telling you who had viewed the content you posted — strongly stimulated user curiosity. Now a user could know whether a celebrity or prominent VC had seen what they wrote, and when.
But this feature backfired. The problem was that users were automatically opted in without being warned that their browsing history on the site could be exposed to others. As a result, asking or answering personal, embarrassing, or private questions — or simply viewing them — caused users to lose precious anonymity. This policy met strong user backlash, and Quora changed it a few weeks later so that the feature was used only with clear prior consent.
When autonomy is threatened, we feel pressured by an inability to choose and experience psychological resistance to new behaviors. To drive repeat engagement, you must give users autonomy.
Case
For a few days, I diligently entered info about every food I ate into the diet program the app suggested. Had I been someone who meticulously kept a food journal with pen and paper, MyFitnessPal would have been a welcome improvement.
But before using MyFitnessPal, I had never eaten food while counting calories. Using the app was novel at first, but I soon grew tired of being dragged along by it. Keeping a food journal was far from my everyday activity, and even further from the activity I wanted to do through this app.
What I wanted was to lose weight, and the app was telling me I could do so through the strict method of tracking calories consumed and burned. But soon I realized that if I skipped entering even one meal, that day's program became meaningless. So for the rest of such days I ate freely.
Eventually I began feeling I had to confess my mealtime transgressions to my phone. The app increasingly pushed me into its own pit of suffering. Yes, I was the one who chose to install it. But no matter how well-intentioned the start, as my motivation waned, using the app became a chore. The unfamiliar behavior of tracking and counting calories began to feel not like what I wanted but what I had to do. My only choices were to continue or to quit — and I boldly quit.
But another diet app, Fitocracy, took a completely different approach to guiding behavior change. The app's goal was similar to other competitors — providing people with a better diet plan and helping them exercise regularly. The difference was that it actively leveraged the familiar behaviors users already wanted to do, rather than the behaviors they had to do.
Early on, Fitocracy's suggested activity was not very different from other diet apps. New members were asked to track meals and exercise. But what set it apart was recognizing that, if you cannot leverage the user's unconscious habitual behaviors, they will — as with MyFitnessPal — snap back to old eating habits in no time.
Before the warning sirens of psychological reactance could sound inside me, I began receiving praise from other members of the site. It was when I started jogging for the first time. Curious who had sent me encouraging messages in the virtual world, I logged in. On entering the site, a question appeared from a woman whose handle was "mrosplock5," asking for advice on what to do when her knees hurt while jogging. Having had a similar problem years before, I immediately replied.
I didn't use Fitocracy for long, but I quickly saw how it captures people. Fitocracy was a kind of online community. What pulled me in was that, as in the real world, friends at the gym could chat with each other.
Social acceptance is an important need everyone craves. Fitocracy leveraged humans' universal desire for connection as an on-ramp to health, and devised new ways for users to form new habits. Users could choose between old behavior patterns and Fitocracy's proposed solutions to build health habits.
Of course, MyFitnessPal did not fail to use member-to-member connection features to drive sustained engagement, but compared with Fitocracy, the moment users experience the benefits of community interaction is too late.
Among many new health apps and products, we cannot predict which will succeed, but the most successful advanced technologies — those that changed the daily behaviors of billions — never forced anything on us. This much I can say plainly: the appeal of briefly popping into Facebook or checking scores on ESPN may lie in returning from your boss's orders into a time of completely free will.
Unfortunately, many companies dive into product development convinced users will do what they want, not what users want. A company that neither tries to make its service a fun activity, nor makes the user's daily life more convenient, yet forces them to learn a very unfamiliar behavior, can never change user behavior.
Companies that have successfully changed user behavior clearly give users a choice between the old way and the unfamiliar-but-more-convenient way of meeting their needs. Products that keep offering the user freedom of choice can prompt new habits and lead user behavior in desirable directions.
Like Quora users being automatically opted into "Views," people feel pressure and resist when something unintended is forced on them, when they're made to get used to strange calorie tracking as with MyFitnessPal, or when they feel their autonomy is under threat. To successfully drive behavior change, make users feel they are in control of their behavior. Rather than pressing them to use a product, make them want to use it on their own.
Designing a Reward System 3. Variable rewards are powerful behavior-inducing devices
At its core, a variable reward system must satisfy user needs while driving sustained engagement. Most habit-forming products use at least one of tribe rewards, hunt rewards, or ego rewards.
With limited variability, the use experience becomes predictable and users lose interest. With unlimited variability, the product sustains user interest by continuing that variability.
For example, email uses all three reward types. Why do we check email unconsciously? First, because of anxiety that someone may have sent an email. People feel obliged to reply and want to look kind (tribe reward). It may also be curiosity about information inside the email — perhaps something important about our career or business is waiting. Second, checking email can inform us of material possessions, livelihood opportunities, or threats (hunt reward). Third, checking email is a task in itself. We must sort and classify so that unread mail does not pile up. We feel a compulsion to check and organize our inbox constantly, never knowing when it will fill again (ego reward).
As Dr. Skinner discovered decades ago, variable rewards are a powerful driver of repeated behavior. If you can pinpoint what drives users to habitually keep using a product, designers can build products that realize their intent.
Simply giving users what they want does not create a habit-forming product. That's because a key final step is missing from the Hook model — after Trigger → Action → Variable Reward, the feedback loop needs one more step. The next chapter will show that repeated use occurs only when users invest their time, effort, or social capital in the product.
Only the crucial final step remains. For a connection that activates automatic behavior to form in the user's brain, the user must invest in the product.
Psychology determines attitude
In Chapter 1, I introduced a dental flossing study by researchers in London. They found that how often a new behavior occurs plays a crucial role in forming a new habit. They also found that the next most important factor in habit formation was change in the participants' attitudes toward the behavior. The findings align with the "zone of habit" graph from Chapter 1, which implies a behavior needs considerable frequency and perceived utility to become routine. Attitude change can appear as movement along the perceived-utility axis until the behavior enters the zone of habit.
For attitude change to occur, perception of the behavior must change. In this chapter I'll unpack how tiny investment activities change our perception and turn unfamiliar behaviors into everyday habits.
It turns out the brain does all sorts of odd things because of a psychological phenomenon called "escalation of commitment." Some people will play video games until they collapse and die — that's how powerful commitment is! Commitment is also used to influence people to donate more to charities. It is even said to be used to force POWs to switch their loyalties. Commitment has an enormous impact on behavior, purchases, and habits.
People tend to place more importance on a product or service the more time and effort they invest in it. Plenty of evidence suggests our tiny efforts can turn into affection.
We rate a product by the effort we put in — IKEA
We want to stay consistent with our past behavior
People who agreed to place a small sign in their window were mostly also willing to put up a big, ugly sign in their front yard. This suggests our desire to stay consistent with past behavior greatly affects our current behavior. That is, a small investment like sticking a small sign on a window can become a major factor in future behavior change.
We avoid cognitive dissonance
The fox desperately wanted the grapes but no matter how it tried, it couldn't reach them. Disappointed, the fox decided to think "the grapes must be too sour anyway."
In this story, the fox consoled itself by changing its own perception of the grapes.
Taken together, this human tendency leads to the mental process of "rationalization" — changing attitudes and beliefs to psychologically fit a situation. Rationalization gives us reasons for what we do, even when those reasons were intended by someone else.
How "a little effort" works
In a general feedback loop, the cue → action → reward cycle can change the actions we take right away. A radar sign, for instance, showing speed limit vs current speed is very effective at getting drivers to slow down immediately.
But when forming habits around a product, this pattern shifts. The Hook model is not a one-time behavior-change activity. It is a design pattern that continuously elicits voluntary engagement to connect the user's problem with the designer's solution. Building that loop of voluntary engagement requires more than a three-stage feedback loop.
The fourth stage of the Hook model, "Investment," asks a bit of activity from the user. Here the user puts something of value into the product or service, raising the likelihood that they will use it again and the Hook cycle will repeat.
Unlike the "Action" stage in Chapter 3, what matters in the Investment stage is not immediate satisfaction but expectation of longer-term rewards.
For example, on Twitter, following another user is the investment activity. Following someone carries no immediate reward. You do not earn stars or badges for it. Following is a kind of investment in the service, raising the chance that the user will return to check Twitter.
Opposite the action stage, friction expands in the investment stage. This clearly conflicts with the traditional idea that every user experience must be as "effortless and easy" as possible. My advice from the action stage — that the intended behavior should be as simple and easy as possible — remains largely valid in the investment stage too. But in the investment stage, you can ask for a little effort only after providing a variable reward. Before that, never. That's how important the timing of asking for investment is. Asking for investment after a variable reward takes advantage of a crucial feature of human behavior.
185
LinkedIn's online résumé is a good example of stored value saved as data. Each time job seekers use the service, LinkedIn can add more information. LinkedIn found that the more info users put in, the higher their commitment to the site. Josh Elman, a senior product exec during the company's founding era, put it this way:
"If you can just get users to enter a little information, the chance they come back to the site goes up sharply."
A little investment in the form of offering a bit of your information becomes a strong driver for returning to the service.
188
Because it gave Twitter tremendous value and served as the core driver of continued user attraction.
For followers, the more interesting people on the following list, the better Twitter's service at delivering engaging content. Users investing time and effort in following the right people see more relevant and interesting content in their feed, which also raises Twitter's value.
From the perspective of a Twitter user who wants more followers, the more followers you have, the more valuable the service to you. People who post on Twitter want their messages reaching as many people as possible. The only legitimate way to gain new followers is to tweet things people find interesting so they follow you. So content creators must invest a lot of time and effort into writing more and better tweets to gain more followers.
This virtuous cycle increases the value of the service to both sides as service usage frequency rises. Switching to another competitor's service means, for most users, giving up years of investment activity and starting again. No one wants to re-gather loyal followers that take enormous effort to acquire and manage.
189
Good reputation and bad reputation
Reputation is the stored value users trust most. On online marketplaces like eBay, TaskRabbit, Yelp, and Airbnb, people with negative reviews are treated completely differently from those with great reputations. A reputation increases the likelihood of service use. Both buyers and sellers are more likely to stick with a specific service into which they have invested time and effort, and they must, in order to keep their high rating.
my-data after sale, if I could re-take that related use myself.. and check the data distribution history
A product is valuable when you know how to use it
According to one mobile analytics firm, 26% of mobile apps are used only once after download. Other analyses also showed that people use more apps than before but with lower frequency of continued use.
Any.do is a mobile app that helps you record and manage simple to-do lists — pick up dry cleaning, buy milk for the fridge, call mom. Understanding how hard it is to hold onto fickle mobile users, the company designed a service that drives initial investment. It explains usage to first-time users in a very refined way. Here the trigger appears as "easy-to-understand, clear instructions." The next action is following what the app tells you to do. Congratulatory messages and the sense of fully learning the app's usage deliver a variable reward to the user.
Next comes the investment. Any.do asks new users to connect the app to their calendar. This means the user permits the app to send a notification after their next meeting. This external trigger prompts users to open the app to record follow-up tasks from the meeting they just attended. In Any.do's scenario, the app sends the external trigger when anxiety about forgetting post-meeting tasks (an internal trigger) peaks. Any.do helps the user succeed by expecting them to see the app's necessity.
We've now covered the four steps of the Hook model. Time to answer five fundamental questions for using it effectively.
1. What does the user really want? What pain of the user is your product relieving? (Internal trigger)
2. What makes users come find your service? (External trigger)
3. What is the simplest behavior users take in expectation of a reward, and how can you simplify the product to make it easier? (Action)
4. Are users satisfied with the rewards you offer, or do they still crave more? (Variable reward)
5. What "a little effort" does the user invest in your product? Have you set up the next trigger from it? And what stored value improves your product the more it is used?
Manipulation Matrix
The role of the facilitator is to faithfully fulfill the company's moral obligation while developing products they themselves use, ones that also materially improve other people's lives. If you have proper guidelines to help users with unhealthy addictive symptoms, there is no need for a pang of conscience. To paraphrase Mahatma Gandhi, facilitators are "those who create the change they wish to see in this world."
How the paid version quickly captured users
Since I was researching cutting-edge tech that forms user habits, I decided to start reading the Bible. Browsing many plans, the one titled "Addiction" seemed best for me.
This plan offers structure and a roadmap for people without a Bible-reading habit.
"Some parts of the Bible are hard to keep reading. But a reading plan that lets you read bits of many parts each day can keep readers from giving up mid-way."
Gruenewald said. The plan's feature is to chop large sections of the Bible into smaller pieces and arrange them appropriately. By breaking content into small enough chunks to understand and connect with, readers can focus on a small activity in front of them and shed the burden of reading the whole book.
231
Holy triggers
Gruenewald said triggers play the most important role in every reading plan.
In my own Bible reading plan, a notification comes to my phone every day. A possession trigger, so to speak. It simply says:
"Don't forget to check your addiction reading plan."
Given that the addiction I'm currently trying to treat is dependency on digital devices, it's quite ironic. But oh well. Just this once, I decided to let it slide.
If I ignore or skip the first notification, a tiny icon shaped like a Bible with a red badge is sent to my phone, signaling me again. If I forget and fail to execute on the plan's first day, a message comes suggesting I try another, easier reading plan. I can also receive these messages via email. If I accidentally skip a few days, it emails me to tell me to start reading again.
This Bible app has a kind of virtual congregation too. Members of the site exchange encouraging messages, conveying various triggers. A YouVersion PR person said:
"Emails the members exchange with each other are a stimulus that draws them to open our app."
External triggers formed through relationships with people are positioned throughout the Bible app and play a key role in driving sustained engagement.
Even users who have not registered a YouVersion account contribute to the app's growth. In fact, social media is abuzz every day with some 200,000 pieces of content pulled from the app.
A crucial factor in the app's widespread reach is that a new Bible verse greets readers on the home screen every day. Beneath the verse, a large blue button reads:
"Share today's verse."
One click shoots the day's verse onto Facebook or Twitter.
I haven't investigated every appeal of recent Bible readings in depth, but the reward of showing yourself in a positive light likely contributed to the Bible app's popularity. In short, "humble-bragging." A Harvard meta-analysis titled "Humans derive essential satisfaction from disclosing information about themselves" also confirmed such behaviors "involve neural and cognitive mechanisms related to reward." One study found sharing something actually feels so good that people will give up money to actively engage in self-disclosing activity.
There are many opportunities to share Bible verses inside YouVersion. But the company's most effective distribution channel is not online community activity — it is the long church pews where congregants sit shoulder to shoulder every Sunday.
"When people use the Bible app at church, those around them ask about it and naturally start talking about the app," Gruenewald explains. Just look at how new downloads spike every Sunday. People share more about the app that day, so word of mouth spreads more.
What few know is that Gruenewald's Bible app came to dominate the market only when some pastors actively adopted it. When religious leaders loaded their sermon content into the app, congregants no longer had to flip through Bible pages and could find verses in real time. As church leaders were captivated, the congregants naturally followed.
Using the Bible app at church helped the company grow. It also deepened congregants' faith. Every time users highlight verses, add comments, bookmark, or share app content on other sites, they are investing time and effort in the app.
Dan Ariely, Michael Norton, and Daniel Mochon's research from the previous chapter already showed us that "a little effort" greatly affects value perception. This IKEA effect plainly shows the correlation between effort and perceived value.
As small investments by readers accumulate in the Bible app, each user's personal history of devotion stacks up beside it. Like a well-worn book with dog-eared pages and notes of wisdom throughout, the app becomes, to the user, a precious asset they can never delete.
Things to remember and share
- YouVersion's Bible had very low user engagement as a desktop-based website. But shifting to a mobile app, enabling frequent "triggers," dramatically raised accessibility and engagement.
- The Bible app greatly boosted users' "action" initiation by offering interesting content on the home screen and adding an audio feature so users can listen with their ears instead of read with their eyes.
- Breaking the Bible's content into small paragraph units made it much easier for users to read a little every day. Inducing curiosity about tomorrow's verse also delivers a "variable reward."
- Every "investment" activity — using comments, bookmarks, and highlights — is saved in the app's data, and as that value accumulates, users read the Bible more eagerly.
To measure how effective your product is at forming user habits, how should you apply this book's concepts? Gathering insights from discussions with entrepreneurs who succeeded at creating user habits and various studies, I removed the unnecessary parts and created "Habit Testing." It was inspired by the Build → Measure → Learn loop that Lean Startup advocates. Habit Testing provides the data and insight you need to design products that form user habits. You'll understand who your most passionate users are, which parts of your product form user habits, and why those aspects change user habits.
Habit Testing does not always apply only to products already live in real-time. But without a broad grasp of how people use your product, it is hard to reach clear conclusions. The activities below assume you have a product, users, and meaningful data worth exploring.
Step 1. Identify
The first question Habit Testing asks is, "Who habitually uses the product?" Remember: the more people use a product, the more likely user habits form.
First, you need to define what "enthusiastic user" means. How "often" do they need to use your product to be called enthusiastic? The answer to this question matters greatly and can reshape your perspective. Public data on similar products or solutions may help you set your target users and engagement. If data is hard to come by, use existing hypotheses. But be realistic and honest in setting them.
Step 2. Standardize
Say you've identified some users who fit the habitual user criteria. How many users are enough? In my experience, at least 5%. The share of enthusiastic users will need to be far higher for the business to be sustainable, but in the early stage 5% is a reasonable threshold.
If, contrary to your expectation, fewer than 5% of users evaluate your product's use value highly, there's a problem. You may have misidentified users, or maybe you must restart everything from scratch. But if you cross the 5% line and have properly identified habitual users, the next step is to standardize the usage steps of your product to pinpoint the hooks that capture users.
Of course, users use products in slightly different ways. Even where a standardized engagement system exists, the way each user uses the product has unique features. Background info, decisions made during registration, number of friends using the service — these are just a few behaviors that create identifiable patterns. To find similarities, you must comb through relevant data and look for the "pattern that leads to habit." In other words, look for a set of similar behaviors common to your most loyal users.
One example: early Twitter found that when a new user's following count hit 30, the likelihood of continuing to use the site spiked — a tipping point.
The series of actions taken by committed users differs from product to product. The goal of finding habit-forming patterns is to identify the step that matters most in creating committed users, so you can modify the use experience to further drive that behavior.
Step 3. Modify
Once you have enough new insight, re-examine your product and find ways to apply the habit patterns you discovered in committed users to new users. Streamlining registration, changing content, removing features, or emphasizing existing features all qualify. Twitter used the insights gained in the previous step to revise its registration flow, making new users immediately follow others.
Habit Testing is a process you can keep running every time a new feature appears or the product is modified. Tracking user cohorts and comparing their activity to that of habitual users will guide the direction of product evolution and improvement.
