UX CORE

The tendency to overestimate the likelihood of events with greater "availability" in memory, which can be influenced by how recent the memories are or how unusual or emotionally charged they may be.

In application development, understanding this bias is necessary for consistent interface design, content design, and user communication. If the action we need the user to do is associated with something negative (especially if it's been covered in the media not so long ago), the likelihood that an action will be taken is greatly reduced. Understanding this allows us to design content (text, images, etc.) so that it is associated only with what we need. This bias allows us to reflect on the current world and market conjuncture to choose a better "tone" of our messages.

Another example: Bitcoins and various kinds of ICOs. The topic of cryptocurrencies was so often negatively boosted in the media that, at some point, investors simply decided to avoid everything connected with it without going into detail. Common users realized that the blindness by the increased volatility of this market did not end well. The hype on this topic eventually fizzled out. Many high-quality blockchain projects have faced severe difficulties in development, due to the highly distorted reputation of everything associated with blockchain, Bitcoin, and crypto in general.

The last example is that I chose the topic of software design and blockchain technologies to describe Availability heuristics. The first topic is obvious to me due to my profession (product manager); the second one simply came to mind with ease when I asked myself, "What stream in IT was full of hype and then quickly died out?."

It refers to how a person's perception is affected by selective factors in their attention. For example, people who think a lot about their Instagram profiles are more likely to pay attention to other Instagram profiles

Understanding this bias gives us a reason to think about what we want our users to pay attention to. If we want to encourage our users to open profiles of other members more often, we should create some value in the user's profile. That value should have special significance (karma points, the total number of "likes," a number of sent messages, etc.). Partially this bias decreased the amount of user-generated content on Facebook. People became so heavily dependent on "likes" and other "sympathy gestures" that the fear of being "not popular" decreased the number of publications. Realizing this, Facebook created the mechanism of "Stories" - the opportunity to unobtrusively share something that "disappears" after 24 hours.

Also, Facebook is testing the possibility to remove "likes" in different countries. If people pay less attention to such elements of "public recognition," it will become easier for them to create content. The more content created - the more valuable the network is, and the more users will be involved. Further, according to Metcalfe's law, the more users involved - the higher the utility of the network and hence more profit.

Also known as "reiteration effect." A tendency to believe in the validity of information if it has been repeated multiple times.

This bias is food for politicians, marketing specialists, and the media. In developing applications, understanding this bias shows us why the terminology used in marketing/promotions of our project should be consistent.

If we say that "Our app is the best work tracker in the world," and this correlates with customer feedbacks and blog reviews, which refer to our app as the "best tracker," the potential user, wondering which tracker to choose, is more likely to remember us. The potential user can even put us on the same level with trackers of multibillion-dollar companies, even if we work locally, only targeting the village he lives in.

This effect can also explain the meaning of slogans. For example, "Apple," with the slogan "Think Different" plays with our need to "be original" and "not like everyone else." They repeatedly use it in all their products and ads to have the "consistent" image of their products in our minds.

The psychological phenomenon by which people tend to develop a preference for things merely because they are familiar with them. The more often one sees someone, the more pleasant and attractive this person seems to become.

It is important to emphasize that it is not only about people but about objects in general. Let us assume we have a successful app we want to promote in Spain. If we use the colors of the Spanish flag on the background of the description of our application and our digital promotional materials (tastefully and not overdoing it, of course), Hispanic readers will subconsciously sense it familiar.

Another example is that we place a faded, monochrome image of "Sagrada Familia" on the background of a white page that describes the benefits of our app in Spanish. If the text is written with grand and pompous words along with adding the background image, it will easily give the effect of "familiarity" and increase the probability of converting potential users into actual ones.

Another example would be to use standard buttons in our application, standard colors, and sizes, making our entire interface "universal." When we need to notify our users of new app features, we can do it unobtrusively by using a standard pop-up with the same familiar buttons used in the other parts of the app. If we need to focus the user's attention so that we can create a feeling of novelty - we move away from our standard system and use a slightly different window color and a slightly different button size in the pop-up. Seeing the new window, the first association that the user will have is the feeling of "novelty," the intensity of which will be proportional to how long the user has been using our application.

It describes the influence of environmental factors on one's perception of a stimulus.

This effect was currently used and studied predominantly in marketing.

For example, research has shown that the comfort level of the floor that shoppers are standing on while reviewing products can affect their assessments of product quality, leading to higher assessments if the floor is comfortable and lower ratings if it is uncomfortable.

In the context of software, the Context effect can be attributed to the user's expectations of what the system will provide. So if on the main page of our website we talk about the non-profit orientation of the project, but at some point after registration we offer the user something to buy (even if not from us), dissonance caused by the contrast of the purchase offer and what the user saw earlier will lead to negative feelings. The purchase offer will fall out of the general context and reduce user satisfaction with the product. This is another reason why it is important to approach the architecture of information very carefully.

Cue-dependent forgetting, or retrieval failure, is the failure to recall information without memory cues. It is an inability to recall memory due to the lack of stimuli or signals that were present during memory coding.

In the context of the application, we can use this bias to "remind" the user what they can do with the system.

I will give a simple example on the online sweepstake where many users make bets. Obviously, the "average" user wins and loses, and in the interests of business, it will be good to "support" such a user at that difficult moment when he has lost everything. Since in the mind of a player who has experienced a series of "defeats," there are only defeats the system can remind him of several wins on a certain pattern, bringing back all that series of good memories he experienced. This can be done indirectly, with a message like "Dear % username %, we just wanted to remind you of your incredible winning streak that has lasted three days in a row at % game _ names % ." This approach is possibly too direct, so we can change the message to "Due to your winning streak in % game_ name%, you are now in the 20% of our top players in %sport_name%" This kind of approach is less aggressive and just shows statistics. Doing this is not morally right, of course. Therefore, betting offices and casinos operating under the licenses of the Malta Gaming Authority (MGA), Curaçao, and others, agree in advance that they will not push players into acute gambling. In any case, this example clearly illustrates how businesses can benefit by knowing such a simple error of our brain.

A phenomenon in which a person tends to recall information congruent with his current mood, i.e., if you are in a good mood, you are more likely to remember happy moments of the past, whereas, if you are sad, you will recall sad and depressing events.

In the context of application development, understanding this phenomenon is important to properly manage user expectations and avoid errors that may cause the cancellation of service. If the user has a bad mood while using our application (no matter why), he tends to call up sad events. Since human memory is associative, if the user recalls sad events while using the app, the feeling will often be related to the application itself. This means that when working with a user who had a bad experience, worsening his mood should be especially avoided.

To avoid being too vague with this bias, a simple example can be given. If our app is an online video game where a user has lost 30% of his game currency over the last 3 days, he understands that it will take him about a week to win it back. Obviously, he cannot be in a good mood and does not benefit us in keeping him in negative emotions. Let us assume that his annual membership is finishing, and we need to request the due payment. In theory, if we do not do anything, the user will get the payment request for the game. As he continually recalls his fresh memory of losing 30% of his game money, he may make an emotional decision and give up playing completely, which is not in our interest. Instead, we can provide him with some kind of temporary bonus, wait for his gaming economic condition to stabilize, and then send a payment request. Once he gets back to the track, he will be able to pay for the game. The game repeatedly brings him back positive emotions, recovering his playing power, and spending his time effectively.

Cognitive bias is when a recently learned information, which repeatedly appears with improbable frequency after a short time, is perceived as extremely often reoccurring information.

This bias highlights the importance of competent content formation in the application. For example, we have an ecosystem consisting of several products, and we need to convey to the user the idea that he is part of our community. To do this, it is important to make sure that the same idea is emphasized in the content of these products, their texts, and images.

It is better if this idea is written absolutely identical, for example, "Best solutions for freelancers!." Of course, this bias can also be used within a single product, however, the less the products and their sizes are, the more difficult it is to use this bias and not be pushy and too recurring.

A hot-cold empathy gap is a cognitive bias in which people underestimate the influences of visceral drives on their attitudes, preferences, and behaviors. The most important aspect of this idea is that human understanding is "state-dependent." For example, when one is angry, it is difficult to understand what it is like to be calm and vice versa.

The best use of this bias can be described by the example of the support service of our application.

We can introduce this bias with our customer service colleagues. First, we can indicate possible types of emotional states of our users. It is then better to try to calm an aggressive user with a soft, sympathetic voice, using mirroring and labeling techniques (e.g., Chris Voss - the book "Never Split the Difference").

Training the customer support service will reduce the emotional stress of our colleagues. It will become clear that the customer's negative is not a personal claim to them, but only a brain error common to all of us.

One of the cognitive biases manifested in people's tendency to underestimate the consequences of inaction compared to action with a similar result. Responsibility for action by people is perceived more than inaction.

Using it in software often goes against moral and ethical standards and continues to be actively used by a wide variety of companies. For example, knowing that inaction is underestimated, we can create an online magazine service with subscriptions in such a way that people would just be lazy to unsubscribe from the service.

In another case, we may need people to unsubscribe from services they don't use. This may be necessary to obtain vital information to analyze our application use. To achieve this, we can send a request to update the list of services that are being used so that users can see that the inaction will be followed by any negative consequences. Therefore the number of users unsubscribing from the unused services will possibly increase. In this case, understanding of this bias will directly affect the text and tone of our message that we are about to convey.

The tendency to ignore general information and focus on only the specific case, even when the general information is more accurate.

For instance, people instantly believe the test results of a rare disease, not taking into account that the disease is a rare one. Another example is the fear of terrorists and flying on a plane. The bottom line is that our brain tends to exaggerate a particular case over statistics.

By understanding this bias, we look closer at developing text content for applications. Thus a message describing a potentially negative outcome in case of any action will be perceived differently by users: "You are about to start the disk defragmentation process. With a 99% probability, the operation will be successful." And another one saying, "You are about to start the disk defragmentation process. There is a 1% chance that the hard drive will be destroyed, and your data will be permanently lost. "

By the way, this is the reason why it is very important to provide quality technical support. A disappointed user who did not find a reasonable explanation for the program error can leave negative feedback about the application. This will harm the product more than it may seem.<br />When people see 50 extremely negative reviews mixed with 15,800 good ones, they tend to consider the product way less valuable, even if the negative reviews are less than 0.1%.

The tendency of bizarre material to be better remembered than common material.

Understanding this effect is the key to creating vivid appeals to the audience. In software development, this effect should be avoided in the vast majority of cases, as bizarre material requires more cognitive load for analysis, and more cognitive load is poor user experience. If the application context allows bizarreness, using this effect is an exception to the rule.

To use this bias, we first need to understand whether our product is suitable for it. For an app that helps you quickly organize a memorial ceremony, bizarreness surely will be an outlier that will affect the app's success rate. Another Facebook app matching your profile picture to mythological heroes and generating texts like "You were born in the same month, under the same stars. You are Apollo!" draws huge masses of users with its bizarreness and the easy principle of work.</p>',

Humorous items are more easily remembered than non-humorous ones, which might be explained by the increased cognitive processing time to understand the humor or the emotional arousal caused by the humor.

The best example of using humor is memes. Plenty of big IT companies use memes to promote their products, and the "Humor effect" is exactly what they bet on.

It is very important to understand that this bias is about remembering humorous things, but not about a positive attitude towards them. If the user working on an important action (form filling, data saving), and suddenly gets to the error page 500 (Internal server error), 502 (Bad gateway), 503 (Service unavailable), 504 (Gateway timeout)), then humor like "Ho ho! Our pirates are working on it, and soon everything will be work again! " will not be appropriate. In this scenario, humor will be noticed, remembered, and most likely will cause anger, and this incident will be remembered better. If such an event occurs several times in a month, according to the Availability heuristics, the user is likely to give a negative rating the next time he thinks about the quality of our product, even if in 99% of cases the application was running smoothly (Base rate fallacy).

In this case, a good user experience is when the company takes the blame, explains the core of the error, and notifies that everything will be restored soon, and the page will be refreshed.

A phenomenon in which images are more likely to be remembered than words.

The debate about this effect continues to this day.

It claims that images have advantages over words with regards to coding and retrieval of stored memory because words are coded more easily.

Since mental effort is a time investment, it turns out that a person is more likely to remember something if the process of analyzing a certain object is longer. And that is why he remembers an image better than text. However, it is not that simple as it highly depends on the image and text itself.

Not to speculate too much on this effect, I will only note that a well-chosen image can replace thousands of words in certain cases. Placing the image in a context that is familiar to the user, and using it as an acute option will be a clever move. The use of this effect depends on an excessively large number of factors, so we will not elaborate more on this bias.

Also known as the "isolation effect." It predicts that when multiple homogeneous objects are presented, the object that differs from the rest is more likely to be remembered

For example, in our website's header, we have a navigation menu with the following items:

| Home | About us | Our services | Our products | Contact |

Now, let us put another item here, naming it, say 420:

| Home | About us | Our services | 420 | Our products | Contact |

Obviously, paragraph 420 will draw too much attention, and even if we do not click on it, it is highly likely that we will hover over it (what if there are categories there?). This effect can be used both to make some content "invisible" (such as "Terms of Service" and "Privacy Policy" items) and vice versa, to draw users' attention to a new CRM filter item, or something else.

A tendency for people to encode information differently depending on the level on which they are implicated in the information.

This effect is widely used daily by marketing specialists. For instance, in advertising, a customer perceives information better if there are people similar to him. They are more likely to remember birthdays that are closer to their own birthday, and so on.

If we have developed an application for the elderly, then using the images of young couples on the home page will be unreasonable.

This effect can be successfully used to compile context links ("Click here if you served in the army between 2005 and 2015"), menu (| Freelance | Private Business | Public Institutions), and more. All this is so intuitive that, in most cases, it is used for its intended purpose. However, a better understanding of this effect will let you move towards creating a simple interface.

Also known as the negativity effect. It is the notion that even when of equal intensity, things of a more negative nature (e.g., unpleasant thoughts, emotions, or social interactions) have a greater effect on one's psychological state and processes than neutral or positive things. In other words, something very positive will generally have less of an impact on a person's behavior and cognition than something equally emotional but negative.

That is why it is necessary to resolve conflicts and any kind of problems with our users as gentle as possible, and erroneous system behavior should be avoided. To erase a negative event memory that has lasted for half an hour or so for the user, in some cases, it will take many days or even weeks of uninterrupted service corresponding to client expectations. If the system fails every so often, and the memory of a previous failure stays fresh, it will push the user to decide that it is full of errors and switch to competitors.

It is important to understand that we can still solve the user's problems 99% of the time, but this will not be enough if we cannot work properly with errors that occur (Base Rate fallacy). As a solution, many companies notify users in advance of possible system errors. Even if the user was offline and wasn't affected by occurred errors, a letter of apology and a progress report is still sent to him. All these details become far more important if the application is for enterprise customers.

Anchoring or focalism is a cognitive bias where an individual depends too heavily on an initial piece of information offered (considered to be the "anchor") to make subsequent judgments during decision making.

Example of use in stores targeted on increasing sales:

• Group item pricing (even if there is no big discount), for example. "Buy 3 bottles for $6."
• Marked limitations (only X pieces per person)
• Mentioning a random number (buy 18 "Snickers")

It is not only a paradox that this method works, but also that its effect remains, even if disproportionately large or small numbers are used as anchors, and even if the consumer is aware of the Anchoring effect.

Understanding this bias is crucial for any kind of e-Commerce business because an easy game with numbers can be converted into large sales. In addition to the above examples, which are successfully used by stores (both physical and online), we can give another example that deserves attention. It is the indication of MSRP (The manufacturer's suggested retail price) for a certain item. The manufacturing company uses this indicator to standardize the prices of its products for different regions of the world. However, if expansion into other regions is not planned, the MSRP can be used more deftly.

For example, you can mark the MSRP for a laptop with a sum of $2849. Then indicate the "store" price $2099 next to the MSRP, crossing out the "$2849", and then put a 20% discount on the product with the "final" price: $1679. Visually it will look perfect. The potential consumer will see a cascade of discounts, and mentally, when evaluating the product will base on the initial price "$2849." Due to that, he believes that he can buy a product with more than a 40% discount! But what if the store originally planned to sell this laptop for $1550, and it was included in the store's strategic plan? This is a brief example of how online stores work. Of course, there are countless methods to work with numbers, yet going back to the Anchoring effect, it is necessary to emphasize that working in any market, not knowing how this effect works, is least to say, naive.

This effect can be used not only in eCommerce. There are many methods of using it outside software development, which is a material for another article.

Conservatism is a bias in human information processing, which refers to the tendency to revise one's belief insufficiently when presented with new evidence. This bias describes human belief revision in which people over-weigh the prior distribution (base rate) and under-weigh new sample evidence when compared to the previous belief-revision. This cognitive bias explains one of the reasons why it is so difficult for people to change their views, values, and beliefs.

From a software perspective, understanding this bias obliges us to be cautious when developing content for our app. Any messages, or "philosophical" speeches even about the advantages of our product, should be reviewed if they can somehow hurt someone's ideological values.

Another utility that can be drawn from this bias is the necessity to consider a value and belief system of our target audience when designing solutions. If, in addition to basic data about potential end-users like "gender, age, race, nationality," etc., we correctly determine the value and belief systems, this will allow us to develop more precise app content. Particularly this refers to promotional materials and various articles that we prepare to promote our product. Often content writers are assigned tasks like "Write an article about this product feature. We are the best on the market, our competitors have nothing of the kind, and thousands of our users have already tested it." With data on our audience values, we can set a task like this: "We need an article describing the %name% feature. Our audience consists of young couples 18-27, most of whom either already have children or are going to. They live in %location_city_name%, and they are concerned about air quality. It is also a common practice in this area to attend Sunday worship services." In the latter described case, a competent content writer will be able to write a far more heartfelt "article" using this material.

Although conservatism is a vast topic, I would like to stress the importance of meeting socio-cultural norms. Regardless of what application we have and how it looks and works if, in the developing process, we don't take into account the socio-cultural factor and the resulting conservative beliefs, we are less likely to succeed.

This is an enhancement of an object's perceived parameters (people, phenomena, and processes) if an immediately previous object was of lesser or greater value in the same dimension.

A couple of examples:A rectangle of a neutral shade of grey will appear lighter or darker than it actually is, due to the dark or light frame around it;A person appears more attractive compared to less attractive people and less attractive if others are more appealing.

Understanding this effect is important for creating consistent interfaces. If, during the process of working with the application the user suddenly gets to the page where engineers have not updated the design (CSS) yet, the overall website will look too faded because of this, and the user will notice the oldness of the design more than it actually is. Unfortunately, engineers often ignore this effect: "it's just CSS, there are more important things!"

However, in comparison with the project engineer, the user will overrespond to this "just CSS." Given that the development of high-quality products is always focused primarily on users, priorities should be adjusted accordingly.

This effect gives food for thought when it comes to working with audiences. Let 's say we are going to launch a dating site. Depending on our audience and how we position ourselves, the images we will use will differ greatly.

If we are doing a project for single people who were previously married, our audience will probably be 30+. It's expected that the priority for these people is "reliable," "kind," "good," "family" partners. As a result, we get images of "average," happy couples who "found each other on our website!"

A different approach is when we focus on students. Obviously, the images of the people mentioned above will not be suitable due to the more impulsiveness of the audience, its hormonal background, and completely different life priorities. Vivid photos of young girls and guys will be more suitable here.

What does the Contrast effect have to do with this? In the first case, if we place images of vivid young people, the confidence of those "who have been divorced," will suffer, and according to this effect, they will consider themselves less attractive. In the second case, if we put "average couples" photos for students, it will contrast with young people's opinions about themselves. After all, they consider themselves quite extraordinary, unusual, and just the best in the world. In other words, they are prone to youthful exuberance.

All these are just examples, of course, and generally, when the criterion for developing something is the "contrast effect," A/B testing will have to be carried out to find out the truth.

The tendency to view two options as more dissimilar when evaluating them simultaneously than when evaluating them separately.

Understanding this bias allows us to take different approaches to developing the information structure of our application.

For example, if we want the user to clearly see the difference between two service plans, we can put the service plans in a row indicating the characteristics and prices of each (you have seen such on many websites in the "Price" section).

But if we need the user to consider our service plans to be "almost the same" (no matter why), we can place them vertically instead of placing them horizontally. Thus he will have to scroll the page, which will not allow him to evaluate differences in the service plans; hence we will be more likely to achieve our goal.

This bias is also one of the reasons why online sweepstakes and various kinds of casinos do not show "Deposit Amount," "Winnings," and "Losses" on the same page. This may be user-friendly, but does not meet the project's business targets, as the user will give more importance to the difference between winning and losing. It does not even matter what he will give more importance to. The fact that the user will have feelings and thoughts beyond control poses risks for the business.

Drawing different conclusions from the same information, depending on how that information is presented. Thus the same statement depending on the wording and semantic accents can be presented negatively or positively as a gain or as a loss.

This is one of the most important biases to understand. I recommend everyone to properly study it, regardless of one's occupation.

Here are the message examples that came up after the user clicked the "Start System Update" button:

"You are about to start the system update process. 99% probability the operation will succeed."

"You are about to start the system update process. The probability of permanent data loss is very low."

If, for example, the probability of data loss is less than 1%, the very fact of mentioning that the "permanent data loss is low" will be perceived by the user much more intensively, disproportionately to the 1% probability.

Or let's study another example. Our mobile app's user receives a message "You got 20 points and beat 70% of our project's users!" Or a message with the same meaning stated differently: "You got 20 points and got into the top 30% of users of our project!." The alert reader will notice that the meaning of the messages is semantically the same, but the arising emotions vary greatly depending on the situation.

In addition to this effect, I recommend reading about the "7-38-55" rule deduced by Albert Mehrabyan, a remarkable Armenian scientist. Although it is about live communication, an attentive reader will be able to find valuable information to use in the product.

Special attention should be paid to the fact that this effect forms the end-user expectations after reading the content—every word, every message, written anywhere in the project matters.

Alas, currently, in most companies, the importance and integrity of the content are highly underestimated, especially if the management team of the company consists of only engineers. For the same reason, most companies are either closed down or sold to other engineers.

Money illusion, or Price illusion, is the name for the human cognitive bias to think of money in nominal rather than real terms. The tendency to perceive money as a material object, while in fact, money is just symbols. For example, if you have $1,000 (10 banknotes with a face value of $100 each), you will assume that if you do not take a single banknote, and no one steals it, then the amount of money will not change in a year. In fact, there will be a smaller amount of money in a year, although neither the number of banknotes nor the figures pictured on them will certainly change. Simply put: the ordinary person does not think about money in terms of inflation.

Understanding this illusion is primarily important in the context of computer game development.

Despite its graphics and playability, many online games are faced with big challenges due to the underestimation of "inflation." So if online game management decides to distribute any "bonuses" to players, understanding the Money illusion and inflation mechanisms is very important. Mainly for determining the optimal size of these "bonuses," which will satisfy the wishes of players and, at the same time in the long term, will not harm the business. This illusion can also be used to develop applications for casinos and various types of sweepstakes.

When you are aware of inflation that hit some currency by 7% in the last 3 months, but the users still think it "costs the same," it allows maneuvering and organizing magnificent bonuses and various kinds of promotions.

It is an empiric psychological law quantifying the perception of change in a given stimulus. It relates to human perception, specifically the relation between the actual change in a physical stimulus and the perceived change.

Could be illustrated as follows:

An eight-light chandelier seems just as bright as a four-light chandelier, just as a four-light chandelier seems brighter than a two-light chandelier.

Accordingly, the number of light bulbs should increase evenly each time to make it appear that the brightness gain is constant. If the absolute gain in brightness (the brightness difference "before" and "after") is constant, it would appear that the absolute gain decreases as the brightness value itself grows. For example, if you add one light bulb to a two-light chandelier, the apparent gain in brightness will be significant. If you add one light bulb to a chandelier of 12 light bulbs, the brightness gain will be almost imperceptible.

In the case of sweepstakes, understanding this law makes it obvious that offering the user $2 when he has just lost $1,000 is mildly said, "not a good idea." If the goal was to "ease out loss effect", then the $2 could be spent in a more useful way. For instance, in online poker rooms, the system gives tournament tickets instead of 1$ bonuses. The cost of the ticket is still $1, but the participation in the tournament leads to more intense, positive feelings.

Confirmation bias is the tendency to search for, interpret, favor, and recall information that confirms or supports one's prior personal beliefs or values.

This is another bias that is important to understand, along with "Conservatism." If, in case of Conservatism, we should avoid conflicts with the long-standing beliefs of our users, then in the case of Confirmation bias, we should place mental "bonuses" in different parts of our application, which will meet our users' expectations and their beliefs.

For example, we have a website that allows us to announce "online auctions" of various secondhand things. There are many such projects, and we stand out by allowing people to trade "anonymously." From the very first day with all our content, we show that our users' privacy and security are primarily essential for us. The platform visitors have a set of expectations with which they register and sell different items.

In such a project, the understanding of Confirmation bias allows us to distinguish two different viewpoints by our users: some of them will consider the project doubtful. Once we make the slightest mistake, they will leave thinking that it was bound to happen. Another percentage of users will think that everything is OK and believe in anonymity and security. Understanding this allows us to create content, newsletters, and new system features, taking into account the plurality of the users' beliefs.

As a result, at some point after gaining our users' trust, this approach can make us one of the world's largest online retailers (more detailed examples are described in the "Amazon Era" book ).

Congruence bias is a type of cognitive bias similar to Confirmation bias. It occurs due to people's overreliance on direct testing of a given hypothesis and neglecting or rejecting indirect testing.

Understanding this bias is important when creating metrics by which we evaluate the results of A / B tests and other success metrics of our product.

Let's assume that the 2500 new users who have registered on our website in the last week are the result of launching our advertising poster on the popular website testing.com 10 days ago. Due to the Congruence bias, we tend to focus our attention on the characteristics of the audience who visited testing.com, without taking into account that the testing.com visitors could ignore our banner; and completely ignoring the group of people who visited our website regardless of the banner on the testing.com.

For example, if our partner at testing.com turned out to be dishonest, realizing in advance that we expect good results from our cooperation, he could use Google Ads before launching our banner, targeting an audience relevant to our banner - the Congruence bias will never let us understand this. Everything will definitely change if we consider such development of the scenario.

Also known as Choice-supportive bias, Post-purchase rationalization is the tendency to retroactively ascribe positive attributes to a subject or action one has selected and/or to demote the forgone options. People are more likely to ascribe the advantages of the chosen option and to amplify the disadvantages of the rejected one.

Knowing the rationalization mechanism allows a wide variety of online retailers to send customers thank-you letters with texts like "Thank you very much for choosing %product_name%! This is the best %product_type% on the market, and we are glad you made such a good choice!"

One of the main goals of such a letter is to give a person more thoughts to rationalize their actions because shops do not want to give a refund.

It is also important to highlight that the intensity of the rationalization material should be proportional to the complexity of the user's decision. Of course, in most cases, when buying products on a website, we do not know whether the purchase was easy or difficult for the user when buying goods regularly. However, for example, if we sell beautiful teacups that cost $49 a piece, it may make sense to spend a few dollars on a beautiful wrapper, a thank-you message that no one reads (and yet they are important!), an email how great it is to be the owner of this limited edition cup, and so on.

I remember I bought a bike saddle made by an English brand, "Brooks." This brand has always been distinguished by high-quality saddles, but in any case, $150 per saddle is expensive. Since the moment of purchase, I somehow rationalized it, mainly focusing on the fact that "They are the best!" Having received a saddle in a beautiful box with nice inscriptions, I continued unpacking. I was surprised to find not only a thank-you letter from the "Director of the Company" but also the issue of the English newspaper, which had a large image of a young couple hiding in a forest as I opened it up. Of course, newspapers never have a direct relation to bicycle saddles, yet, finding all that in the box that day made me close the purchase rationalization issue.

What I mean is, if the user has just purchased from us, this should be the beginning of his journey but not the "final goal." Although a business needs to convert potential buyers into actual ones, however, if we want to have a high-quality name on the market in the long run, our work on UX should continue until the user uses his purchase and gives a good review on the site, preferably also rating the product.

It is the tendency of people to notice those elements that are consistent with their expectations and ignore the rest. An example of this phenomenon is the selective perception of facts from news reports.

Thanks to this form of perception, when we work in MS WORD, and at some point, we need to change the color of any text, we very quickly find the appropriate tool on the panel. Working with different editors made us form a set of expectations of what the editor should be, and where the color panel should be placed. Understanding this when creating our own editor allows us to place many tools on one panel, without worrying that people will get confused. In other words, it may be appropriate to group interface elements with the expectation that Selective perception will allow users to intuitively understand the work of the application. Of course, due to its complexity, testing working hypotheses with selective perception should be very careful and include A/B testing.

Also called the Experimenter-expectancy effect; Expectancy bias. It is a form of reactivity in which a researcher expects a certain result and subconsciously manipulates the experiment or misinterprets the data to detect this result.

Understanding this effect is necessary to correctly read the analytical data on our application. For top management, this is even more important, because middle range managers are more vulnerable to this effect, due to the constant work on both proving their effectiveness and the effectiveness of their decisions. Most often, people fall victim to the effect simply because they wanted to see more numbers. For example, the number of "returned" users of our website is 98, but if you filter in the Google Analytics not from February 18, 2020, but the 17th, then the number will increase to 235. At some point, we can fool ourselves into thinking something vague like "well, there is a time zone difference and all that," and by shifting the filter from the 18th to the 17th, generate a report that will better meet our expectations.

In turn, this report will be given to the management, who will perceive it as accurate and not notice mistakes. Based on incorrect data, the management will make not optimal decision to extend cooperation with the marketing agency since it "seems" to be effective.

It is the tendency to avoid dangerous or negative information, ignoring them.

In a broader sense: avoiding information that can cause psychological discomfort. The best illustration of this effect is everyday expenses.

Having spent a significant amount of money, a person does not hurry to check his bank account, pretending that his balance has not changed.

Understanding the role of this effect in e-Commerce or computer games' logic allows us to create mechanisms for replacing the spent real currency by various bonus points or bonus gaming currency. Psychological discomfort here should be understood in the broadest sense. To help our users avoid this discomfort, in addition to the data gathering on gender, age, race, and nationality, we can add to the "Persona" of our typical user the expectations from the application, as well as expectations from each action performed.

Knowing our users' expectations allows us to assume the type of psychological discomfort from a particular action in the application and act on it. So if we know that by sending a message on our social network, the user may worry if the message has been sent, we can add the "Delivered" notice, which, as we all know, is widely used in most social networks and various kinds of messengers.

This bias is the perception that something is true if a subject's belief demands it to be true. It also assigns perceived connections between coincidences.

This bias is similar to Confirmation Bias. However, if there is a tendency to seek confirmation of one's beliefs in the case of Confirmation bias, here we are talking about distorting the interpretation of the received information. For example, after a large purchase that was emotionally difficult for us, we tend to rationalize the purchase by giving importance to the product's different properties, which we would have evaluated differently if we were outside observers.

Let's say we bought a bike. If it was expensive, and we spent all our savings on it, then we will tend to exaggerate the significance of every compliment received for the bike from both friends and strangers. Our friend tells us, "So cool, it is so lightweight!" And we cling to his phrase willingly agreeing, adding, "Yes, and at the same time it rides really well!"

For working on software, understanding this effect is important to create a good, comfortable "path" after the user makes factual action. Let's say we have an online store of expensive fitness watches. Taking into account the Subjective validation, as well as the need for people to rationalize their purchase (since watches are expensive), immediately after the purchase, we can send an email to the user indicating the serial number of the watch, indicating that there were only %number% similar watches made, emphasizing that it is a limited edition.

In a couple of days, we can send an article from our blog, which describes the benefits of the watch purchased by our user and includes reviews by athletes competing in various sports. A week later, we can send the last "thank-you letter" with a 10% discount bonus coupon, emphasizing that this coupon is offered ONLY in case of purchasing this watch.

Of course, Subjective validation can be used in a variety of ways in an infinite number of cases.

Here is another more complex example. Let's say that when registering on our site, we asked a user whether he wants to participate in charity events protecting the environment. If the user agrees, we will visually emphasize the importance of his choice and "thank" him. Later if we decide to send him a letter asking to donate a couple of dollars to restore the burnt forest in Ukraine, the likelihood of his consent will be higher, because when he agreed to participate in the program, he created a new belief about himself (I am the one who cares about the environment), and thereby the Subjective validation will work in our favor. Such a user will find it emotionally difficult to step back from the created beliefs about himself. In the context of the Internet, this bias will be more difficult to use than in the "real world," where all these experiments have been successfully conducted and documented. Yet, certain business benefits in IT can be derived from this.

This is the tendency to believe previously learned misinformation even after it has been corrected. Misinformation can still influence inferences that one generates after a correction has occurred.

This effect can be interpreted as a consequence of two other cognitive biases: Conservatism and Confirmation bias. The reason for this effect is that our brain protects us from possible cognitive dissonance and correlated emotional trauma. That is why the brain needs time to carry out all the necessary validations before the influence of previously learned misinformation decreases.

In the application logic, understanding this effect is important for proper communication with the user about "external world" events. For example, if we decide to send a message or a support letter to our users about COVID-19 (which has become very trendy for many applications), then it will be risky to use previously dubious information in the selected tips and ideas, which has become "true" only recently.

The Bias blind spot is the failure to notice your own cognitive biases. Therefore the inability to detect biases in oneself is also a cognitive bias. People are more likely to notice erroneous behavior and motives in others than in themselves.

This bias is why it is not necessary to explain to users the reasons for a particular product decision if it is based on psychology and cognitive sciences. For example, if in the release note of the next version of our product, we add information that focuses on psychology, most users will likely perceive this with distrust. Moreover, in some cases, it will be perceived as if we are manipulating them.

Of course, the design of any application consists of manipulating the users' expectations by 100%. The attentive user will certainly notice this, and yet even he will consider himself smarter than others, and to some extent, will be exposed to the Bias blind spot.

Sometimes it is better to seem simpler than you are.