Chapter 18. Previewing the Experience and a Hundred First Impressions

If the first screen betrays the person brought in by an ad, the FTUE ends the moment it begins.

The character in the ad was smiling, all soft and round. A person drawn in by that smile opens the app, only for the first screen to thrust a fierce face holding a sword at them. In the half second when the cuteness promised by the ad clashes with the menace returned by the first screen, the person's first thought is to wonder where the game they saw has gone.

So far, we have gone deep inside the first screen, following a person action by action as they press, receive feedback, get stuck, and make mistakes. In this chapter, we take one step back from that screen. We zoom out to an earlier point in time and a place farther outside: the ad and store images the person saw before ever reaching the first screen. However precisely we construct the inside of a screen, that precision is never used if the person turns away somewhere else before reaching it. The images and words that let them anticipate the first screen meet them before any design inside that screen does.

Long before opening the first screen, a person has already met us once—in an ad, in a friend's remark, or in one small store image. From that meeting, they infer what kind of experience we offer and almost decide then and there whether to open it. The first experience therefore begins not on the first screen, but in the images and words that allowed the person to anticipate it.

We call this advance showing a preview. A preview is not merely an advertisement; it is a promise. The user sees it and forms an expectation—“So this is the kind of experience it will be.” If the first screen keeps that expectation, they feel reassured; if it breaks it, they feel deceived. Designing a preview is therefore not a matter of deciding which image to show, but which experience to promise. We must promise only what we can deliver.

The Trailer Was a Comedy, but the Feature Was a Drama

Let us enter an imaginary game. It is a mobile game about collecting characters, talking with them, and watching short videos, and its marketing team has produced a striking trailer. Quick cuts, funny dialogue, and short, lively exchanges between characters make the video bright and lighthearted. It makes viewers laugh, performs well as an ad, and sends people rushing to install the game.

Then they open the first screen, and the atmosphere is entirely different. The round, cute character from the ad now glares with a sword and a fierce expression. A long, weighty explanation of the world scrolls by over serious music, and the characters never make the light jokes seen in the trailer. The user pauses. Is this really the game I saw? The trailer was a comedy, but the feature is a drama: promise and reality have diverged. The game is not bad, and its serious world may be well made in its own right. The problem is that the user came to laugh and was told to prepare to cry. Feeling deceived, they leave to find the lightness they expected elsewhere. The better a preview is at drawing people in, the greater the disappointment when it fails to match. Someone brought in with the wrong expectation turns away faster than someone brought in empty-handed.

Such mismatches have even led to regulatory action. For a time, mobile puzzle ads were flooded with scenes of pulling pins, yet those scenes barely appeared in the installed games. The UK's advertising regulator is known to have sanctioned some of these ads as misleading. The play promised by the ad was absent from the first screen altogether, making this a precise example of the betrayal between preview and first screen discussed here. At a larger scale, similar gaps have shaken whole works. A space exploration game reportedly faced fierce backlash because the landscapes shown and features promised before launch differed from the game people actually received. A highly anticipated blockbuster RPG also suffered because of the gap between its pre-release demonstrations and its real performance. In the space exploration game's case, however, the disappointment concerned the gap between the preview and the whole product, revealed over dozens of hours rather than on the first screen. In the blockbuster RPG, the gap appeared from the first launch on older consoles. The principle is the same—the more successfully a preview raises expectations, the greater the disappointment of the gap—but the pin-pulling ad is the more exact mirror of a promised form of play missing from the first screen itself.

For MEJE Aidong World, a single preview image is already a promise. A fan who enters after seeing a brightly smiling puppy Aidong expects an experience just as warm and bright. If the first screen returns the warmth that image promised, the fan feels reassured. If image and screen diverge, the first thing the fan does is look for the Aidong they thought they had seen.

Show the Big Picture, but Save the Ending

If a preview is a promise, what should it show and what should it conceal? At one end lies the big picture: showing in advance what the experience is, what makes it enjoyable, and what awaits at the end. People struggle to take the first step when they do not know where the path leads; they begin walking only when they know what they are moving toward. A preview therefore needs to show at least one big-picture view of the experience. This is something I do with friends. This is about caring for a character every day. This is played in short, self-contained rounds.

At the other end lies the spoiler. If the preview goes beyond the big picture and reveals the ending and the substance of every best moment, the user has nothing new left to receive from the work itself. Who wants to watch a movie they have already seen in the trailer? A preview must show the big picture while saving the ending, and this boundary is the heart of preview design. What is a sufficiently large picture to make a promise, and what is the substance that should be saved? Show too little and people do not come; show everything and they have nothing left to receive when they do. The preview must find its place somewhere between those extremes.

The Goal of an Application Is Not the Job but the Interview

At this point, there are two ways of looking at a preview. The first is being chosen: capturing someone's mind in this single exposure and making them choose us. It means catching the eye in the first few seconds, clearly showing how we differ from the alternatives, and prompting a click now. The second is durability: creating a first impression that does not wear out even after repeated exposure, rather than staking everything on one encounter. People do not see us only once. They encounter us repeatedly over several days—in ads, through friends, and in the store. An ad that was funny at first becomes tiresome with repetition, and an image that first looked impressive becomes stale when seen too often. If the same first impression wears out quickly under repeated exposure, showing it more often only hurts us.

Durability comes from not showing exactly the same thing every time. Even good things quickly become unremarkable when repeated without change, because the pleasure delivered by the same stimulus fades with each encounter. To keep a first impression from wearing out, preserve one unchanging major promise but layer small variations on top. Show the same character with a different expression today; shift the tone of the same line slightly depending on where it appears; occasionally add a small surprise. The major promise reassures people while each small difference catches their eye again. Durability is the quality that makes them look once more each time, even after encountering the same thing repeatedly.

The two approaches ask different questions. Being chosen asks, “Can this one frame make them choose us?” Durability asks, “Will they still enjoy this after seeing it many times?” A powerful first impression that stakes everything on one encounter gets chosen easily but wears out quickly. A quieter impression that lasts may not seize the heart at once, but it grows familiar and likable through repeated meetings. Which should come first depends on whether we meet a person once or many times. In a channel they pass only once, stake more on being chosen; in a channel they encounter repeatedly over several days, prioritize durability. We can go one step further. The games addressed by this book—experiences designed to bring people back daily through fandom and relationships—structurally sit atop a funnel of repeated exposure. Durability is therefore the default, while being chosen is a tactic reserved for exceptional, fleeting channels. K-pop's release pattern—publishing a comeback scheduler first, then revealing concept photos in varied styles over several days—is a textbook example of that default.

Now we must stop one of the most common misconceptions: the belief that the preview's goal is acceptance. We hope that a single preview will make users fall deeply in love with us, pay, and settle in for good, but that is like expecting a résumé screening to result immediately in employment. Just as the goal of a job application is not the job but the interview, a preview's job is not to complete the user; it is to bring them to the first screen. The preview leads to the first screen, the first screen to the first action, and the first action to the first reward. The goal of each stage is not to finish everything there but to move the person one step into the next stage. Watching where people drop out along that step-by-step path reveals which promise failed to work (→ Appendix).

There Is Not One First Impression but a Hundred

By now, our view of a preview has changed from a single image or video. First impressions differ by person, meeting place, and moment. The impression received from a short-form ad differs from one received through a friend's recommendation. The same ad is read differently by someone waiting for a story and someone looking for something to do with friends. Even the same person receives it differently on the first passing encounter and when seeing it again days later. People do not cross these entry thresholds only once; they move between T0 and T2 repeatedly, and durability is precisely the property that withstands that repetition. A first impression is not one image but the many combinations produced where channel, audience layer, and moment intersect.

When designing first impressions, then, spread out the combinations instead of searching for one perfect image. Where do we meet the person? Which layer of the audience are we addressing—the eight layers established in Chapter 6 form this axis? Is this the first rumor, first exposure, or first expectation—in the thresholds of Chapter 3, T0, T1, or T2? Crossing these three axes produces dozens of first impressions. Give a first rumor to a short-form viewer waiting for a story; give a first exposure to a store visitor looking for something to do with friends; fulfill the first expectation of someone who arrived through a recommendation and already anticipates something. Each combination calls for a different promise. Once a hundred first impressions are laid out, we can see where we promised too little to one person and made the wrong promise to another. This act of spreading them out leads to a detailed worksheet (→ Appendix).

One knot that may seem contradictory to the previous section remains. If every combination calls for a different promise, how can a single first screen keep a hundred promises at once? The answer already appeared in the discussion of durability: preserve one unchanging major promise and place variation only around its edges. The hundred first impressions may differ in expression, outcome, and manner of address, but the core promise at their center must remain one and the same across every channel, audience layer, and moment. Follow the rule “one core promise, variation at the edges,” and we can spread out a hundred first impressions while telling zero lies. This is why the first field in the appendix's first-impression worksheet asks for the common core shared by every combination—the single core promise that never changes anywhere.

Among these touchpoints, an app store product page is where the first impression is decided in the narrowest span of time. Analyses commonly suggest that people stay on such a page for only a few seconds and largely decide whether to install within that brief interval, so the first few screenshots are considered decisive for conversion. People make their decision without swiping beyond those opening images. What works here is said to be not a screen listing features, but one that shows at a glance “what I will get to do with this.” The idea that a first screen splits into a hundred is not abstract: even for the same game, the first image on a particular page lets one person in and sends another away.

The Game-Preview Conventions of Spectacular Cinematics and Feature Lists

Since this chapter addresses previews, an old convention of game previews now arrives at the checkpoint: the approach long followed by game ads and store pages. Grand cinematic videos, dense lists of specifications, and abbreviated genre labels. In one phrase, it is the convention of introducing games in the language of gamers.

First, uncover the premise beneath this convention. Game previews were refined for gamers over many years. Spectacular cinematics prove a game's scale and polish at a glance; feature lists explain what can be bought; genre abbreviations communicate an entire gameplay grammar in one word. Where did this approach grow? It belongs to an era when people judged whether to buy a game by its specifications, when more content and higher requirements were treated as value in themselves. To a gamer, this kind of preview is information. They read graphical quality from the spectacular video, estimate volume from the feature list, and instantly judge from the genre abbreviation whether this is a kind of game they know how to play.

Newcomers do not share that premise. To them, a spectacular cinematic does not answer the question, “So what do I do with this?” The video looks impressive, but they cannot see what they themselves will actually do. A feature list is even more forbidding. A list of systems, currencies, and modes is a manual written in a foreign language to someone who has never played games. Genre abbreviations are exclusion disguised as helpfulness: to those who know the abbreviation, one word says everything; to those who do not, it signals that the message is not for them. Someone who wandered in from reading webtoons cannot find the story they might love in the spectacle. Someone who arrived from filling a shopping cart in a commerce app cannot read what they will receive from the feature list. Gamers read a cinematic as “evidence of scale”; newcomers read it as “a video that does not tell me what this is for.”

The price this convention charges ordinary people is the cost of comprehension and a sense of exclusion. The spectacular video looks good, but the work of discovering what they will do is left to the user. The feature list demands concentration to decipher, while genre abbreviations whisper to the uninitiated, “This is not a place for you.” Once the preview has filtered people out in this way, the ordinary newcomers it should have brought to the first screen are already gone. So change the shell and preserve the substance. Instead of proving spectacle, show in one frame and in the person's own language what this experience is. Instead of listing features, show one scene of “what you will get to do.” Instead of using a genre abbreviation, compare the experience to something they already know. Tell them they will wait for the next episode every day as with a webtoon, move through short rounds as with short-form video, or raise something together with friends. The familiarity we borrow comes from an experience they already possess outside games; what we newly give them is an unfamiliar pleasure wearing those familiar clothes. Spectacle is something people who already love games have always received, so there is no reason to put it at the front of the first promise.

▶ Three Things to Apply to Your Screen

Does the first screen keep exactly what our preview promised, or is the ad light while the first screen is heavy?

Do the first three impressions a person receives point in the same direction as the preview's promise?

Have we treated the preview only as advertising that brings people in, or as a promise the first screen must keep? If even one is misaligned, the more people the preview attracts, the faster they will turn away at the first screen.

Therefore, Begin with What We Can Promise

Once we see a preview as a promise, the order of designing a first impression changes. Instead of first asking which image looks most impressive, we first decide what the first screen can keep. A preview must promise only what the first screen can deliver. If it promises a lightness that the first screen cannot provide, people turn away at the mismatch. Promises should therefore be ordered not by what looks impressive but by what can be kept. First choose one experience the first screen can reliably return, and let the preview promise that experience. When promise and reality match, the user feels reassured at the first screen and moves one step farther in.

We can also verify with numbers whether this promise was kept. Whether a promise works is revealed by whether people who enter after seeing the preview stay at the first screen or leave. If ad performance is strong but many people exit immediately after arriving, that signals a promise the product could not keep. Looking at which channels send people who leave faster reveals which first impression failed. Placing the impression that brought them in beside the experience that made them stay shows which of our promises was false.

Reference Content

Examples from other media that reveal the concepts in this chapter.

Video Games

No Man's Sky: It is known to have faced fierce backlash when features discussed in pre-release videos and interviews, including multiplayer, were absent from the released game. It offers a reference for how the gap between the experience promised by a preview and the product as a whole destroys trust. It is an adjacent example of a promise diverging from a long-term experience rather than a betrayal by the first screen. Live-Action TV / Drama
Netflix thumbnail personalization: The same work presents different first images to different people. It is said that, for Pulp Fiction, someone who likes Travolta may see a thumbnail featuring him while someone who likes Thurman sees one featuring her. This illustrates how first impressions branch into a hundred paths by channel, audience layer, and moment.

Music

K-pop comeback schedulers and concept photos: A scheduler with the release timetable appears first, followed over several days by concept photos—dark version, floral version, and so on—and 15-to-30-second previews of the title track. This demonstrates durability design: preserving the unchanging major promise of the comeback while layering on a different texture each time so repeated exposure does not wear it out. General Apps
The first app-store screenshot: Many users are said to decide whether to install from the first few images without swiping past the third. This shows that conversion requires presenting “what I get to do with this”—its value and outcome—within two seconds rather than listing features.
Fake “pull the pin” ads for mobile puzzle games: Pin-pulling and rescue scenes shown in ads barely appeared in the real games, and the UK's Advertising Standards Authority is known to have sanctioned some such ads as misleading. This is a classic case of a mismatch between the expectation that brought a person in and the first screen.

The remaining reference content is collected in the “Chapter 18 Appendix” at the end of this chapter (grouped as Appendix D in the book edition).

Design Note ▶ Try It Yourself

Write in one sentence the experience our preview promises: “A person who sees this preview opens the first screen expecting an experience of (　　　).” Beside it, write in one sentence the experience the first screen actually returns, and compare whether the two sentences match. If they diverge, revise one of them. Lower the preview's promise to something the first screen can provide, or raise the first screen to the level the preview promised.

Next, list three touchpoints where we meet people—an ad, a recommendation, a store, or something similar. For each touchpoint, write one line describing who we meet there and what experience they bring from outside games. Then mark whether the first impression offered to them makes “what this is for” immediately visible, or whether it is merely spectacular while concealing what they will do. Wherever it is not visible, rewrite the first impression by comparing it to an experience that person knows outside games. The detailed task of spreading first impressions across channel, audience layer, and moment continues in the appendix's first-impression worksheet. Its first field is the common core shared by every combination: the one core promise that never changes anywhere (→ Appendix).

When the two sentences point in the same direction, the first screen no longer betrays the people brought in by the preview.

In One Line: The first experience begins not on the first screen but in the preview that allowed a person to anticipate it. A preview is not an ad but a promise, and it must promise only what the first screen can keep. Show the big picture but save the substance; make the preview's goal the interview, not acceptance. A first impression is not one image but the many combinations formed where channel, audience layer, and moment intersect. Spectacular cinematics, feature lists, and genre abbreviations are information to gamers, but to newcomers they form a first impression that fails to show “what this is for.” Instead of leading with spectacle, promise what the experience is in one frame and in the newcomer's own language. Next Chapter: Once a preview has defined the promise and brought a person to the first screen, multiple promises and elements collide there at once. Direction that seeks to reveal the world and guidance that seeks to teach controls compete for space, as do immersion and purchase guidance. What should appear first, and what should wait? The next chapter turns to collisions between experiences and to the layers that direct their traffic.

Chapter 18 Appendix: Reference Content Collection

These references, gathered from film, games, publishing, music, and everyday displays and grouped under media-specific headings, illuminate this chapter's argument that the first experience begins not on the first screen but in the preview that allowed a person to anticipate it. Each item ends with a “What to examine” note. Keep the chapter's framework in mind: a preview is a promise and must promise only what can be kept; it should show the big picture while saving the substance; and being chosen in one encounter differs from the durability to withstand repeated viewing. Then each case reveals what it promised, whether the promise was kept, and where any mismatch occurred.

Live-Action Film

Movie trailers: The original form of the preview, showing the big picture while saving the ending and offering only a promise worth coming to see. What to examine: Observe the restraint that establishes tone, genre, and expectation within a few minutes while preserving the substance. A trailer's skill lies in what it ultimately refuses to show.
Trailers that reveal all the feature's best scenes: By exposing the substance in advance, they become spoilers that leave nothing to receive from the work itself. What to examine: Treat this as an example of the boundary between big picture and substance collapsing. Fixating only on being chosen spends the experience of the feature before it begins.
Burn After Reading (2008): This Coen brothers film is often cited as having been marketed as a lively comedy although the film itself was much darker and drier. What to examine: A promise of tone is as powerful as a promise of plot. People feel deceived faster when the texture is wrong than when factual content differs.
Drive (2011): A viewer drawn in by a trailer cut like a fast chase movie reportedly sued for a refund over the feature's slower mood; the court found that the trailer was not inaccurate. What to examine: Even if a trailer is not factually false, it is read as a promise. Legal fault and user disappointment occur on different layers.
Teaser trailers: They conceal the ending and show only one restrained piece of the big picture. What to examine: Find the range in which reducing information increases curiosity. Conceal too much and people cannot even tell what the work is; a teaser still needs a minimum amount of big picture.

Video Games

Game trailers: Because a preview is a promise, backlash is strong when the experience promised by a spectacular video differs from the feel of actual play. What to examine: Evaluate production quality and accuracy of promise separately. A good preview is not the most impressive one but one that can be kept.
Cyberpunk 2077 (2020): Soon after launch, frequent crashes and severe performance problems made the game effectively unplayable on older consoles such as PS4 and Xbox One. It was eventually removed from the PlayStation Store for a time and refunds were offered. What to examine: The greater the expectation gathered by a preview, the more the gap at first launch becomes an event. The size of a promise is the leverage of disappointment.
Aliens: Colonial Marines (2013): Differences in graphics and polish between demonstration footage and the released game are known to have led even to a false-advertising lawsuit. What to examine: Observe the endpoint where a preview's mismatch with reality becomes not merely disappointment but legal conflict. A promise is not just a marketing asset; it is an utterance that carries responsibility.
The Killzone 2 E3 2005 video: A pre-rendered “target video,” rather than real-time footage, was reportedly presented like actual gameplay and caused controversy. What to examine: Watch the gap between what is shown and what people receive accumulate into distrust. One exaggeration makes every later preview suspect.
Trailers labeled “actual gameplay footage”: As gaps between staged footage and real screens repeatedly caused controversy, labels such as “actual game-play footage” became an industry convention for clarifying whether the current image is presentation or product. What to examine: This prevents mismatch in advance by specifying the unit of the promise. Separate what is promised from what merely creates atmosphere, and the same video no longer becomes a lie.
The first screenshots and short video on a Steam store page: Most installation decisions are settled in the narrow passage formed by the first few images. What to examine: Determine whether those images list features or show “what I will get to do with this.” The narrowest passage should carry only one clear promise.
Arcade attract mode: Before a coin is inserted, an idle machine plays its own demonstration, shows a high-score table, and flashes “INSERT COIN” to draw passersby. What to examine: Observe a structure in which the idle screen itself becomes a promise. A demo is the game's most honest self-made preview because what it shows is what the player will do.
Game demos and trial versions: They put one piece of the big picture in the player's hands while saving the full work, helping them decide whether to buy. What to examine: Choosing which piece to give away is itself promise design. To avoid mismatch, it should represent the work rather than merely being its most spectacular fragment.

Animated Film

Pixar shorts shown before features: A short work delivers tone and promise ahead of the feature. What to examine: A separate work can calibrate expectations for the main one. A preview need not be a fragment of the feature; a matching texture is enough to make a promise.

Live-Action TV / Drama

Next-episode previews: They conceal the outcome and promise in one shot to make the viewer return next time. What to examine: This preview addresses someone who has already seen the main work. For users with accumulated trust, a promise can be formed with less information.

Online Video

YouTube clickbait thumbnails and watch duration: An exaggerated thumbnail wins clicks, but when viewers immediately leave content that differs from the promise, watch duration falls. Recommendation is known to value watch duration, so exposure itself ultimately declines. What to examine: See how an unkeepable promise returns through measurement. The chapter's signal—strong ad performance paired with many immediate exits—has been institutionalized on video platforms.

Literature

Book covers and back-cover copy: Without revealing everything, they use a brief first impression to decide whether someone buys. What to examine: A cover promises genre and tone. Look at the reception of books whose cover texture clashes with their prose to see the weight of that promise.
E-book “Preview/Look Inside”: The opening 10–20 percent, usually the introduction, is offered free so readers can decide whether the style and beginning fit them. What to examine: The boundary between offering one piece of the big picture and saving the substance for the work itself is drawn in percentages. Deciding where to stop is preview design.

Music

Pre-release album singles: One song creates expectation without revealing the whole album. What to examine: Ask whether the single represents the album's overall texture. If the most popular song is placed first but the album feels different, the expectation drawn in by that song diverges inside the album.
Thirty-second previews: A short exposure helps someone decide whether to buy. What to examine: Observe which thirty seconds are chosen. Selecting the highlight is the same as selecting the promise the song makes.

Comics / Comic Books

Magazine covers and next-issue preview panels: One image promises to make readers wait for the next installment. What to examine: This promise repeats every week. Because there is another week after each mismatch, trust is accumulated or eroded episode by episode in a long campaign.

Webtoons

Work thumbnails and one-line descriptions: In a crowded list, a first impression alone decides whether someone clicks. What to examine: First impressions work through relative evaluation against competing works in the same list. They do not compete in a vacuum but on a display shelf.

Web Fiction

Titles, covers, and descriptions: At the farthest point before episode one, they promise to filter readers before entry. What to examine: This market offers an extreme in which the title becomes the full text of the promise. Its speed of promise fulfillment is measured by whether the development named in the title appears immediately near the beginning.

Advertising / Marketing

Ad fatigue and creative rotation: Repeatedly showing the same ad to the same person lowers clicks and response, so periodically rotating creative is an industry commonplace. What to examine: Treat it as evidence that the same first impression wears out under repetition and needs variation. The durability described in the chapter—preserving the major promise while changing only the edges—is routine work in advertising operations.

Devices / Consumer-Electronics UX

Product-package photographs and the real object: Trust depends on whether the item revealed on opening matches the appearance promised by its packaging. What to examine: Unboxing is the moment a promise is tested. Whether the product is devalued by the expectation raised by the package or fulfills it determines repeat purchase.
Driving scenes in car ads: A real test drive must return the experience that was promised. What to examine: This is a market in which a verification stage—the test drive—has been institutionalized. Placing an experience between promise and fulfillment catches mismatches before purchase.

Offline / Everyday Life

“Serving suggestion” photos on instant-noodle packages: The package shows a bowl piled with garnishes, with the small words “serving suggestion” nearby. Everyone knows the gap between the bowl promised by the photograph and the actual cooked product. What to examine: A disclaimer cannot prevent disappointment. The promise is made not by the small print but by the largest image, and people build expectations from that image.
Restaurant food models and photographic menus: A display-window model shows “what you will receive” more powerfully than a written menu. Japanese restaurant districts are famous for their culture of models made to resemble the real food. What to examine: Treat this as a preview that shows the result in a newcomer's language. It matches the chapter's prescription to show one scene of what they will receive instead of listing specifications.
Supermarket sampling counters: They offer one bite while keeping the full product packaged. What to examine: This is the offline origin of demos and trials. Drawing the line between what to taste and what to save is preview design itself.
Apartment model homes: A full-size sample home is furnished before units are sold, and people often say that furniture placement and mirrors make it look larger than it really is. What to examine: The farther fulfillment lies in the future, the stronger the temptation to inflate a promise and the later its verification. The longer the delay between promise and fulfillment, the more heavily the promise's honesty matters.