Chapter 8. Customers Within the Customer: From Persona to Immigration Candidate

Anyone who has created a persona knows: once the sheet is finished, it usually goes into a drawer.

A persona drawn from age and gender cannot choose a single button on the first screen.

We give the person a name, write down an age, fill in a job and hobbies, and even attach a plausible portrait. “Kim Ji-eun, a 28-year-old office worker, looks at her smartphone on the subway during her commute.” Hang the sheet on a meeting-room wall and everyone nods. But when it is time to design the first screen, the paper labeled “28-year-old woman” cannot tell us what the first button should be. Whether she is twenty-eight or twenty-nine says nothing about whether to put a dog or a cat here. That is why few people take the sheet out again when designing the first screen. Whether Kim Ji-eun is twenty-eight or twenty-nine does not change what the screen should show first.

A persona ends as decoration because we drew the person as “a point in a demographic.” Age and occupation tell us where someone lives, but not what they expect before our screen or what makes them hesitate. A persona useful for first-experience design begins with different questions: Where has this person been living, and are they willing to move into our world?

Three chapters have progressively narrowed one person. Chapter 6 divided the roles within a person into layers: the several faces inside one person, a customer comparing value at one moment and a player moving a finger at another. Chapter 7 examined how the kind of thing a person does not know differs from one person to another: one does not know the genre, another does not understand payment, and another does not know where to press. Chapter 8 fixes that person as an actual object of design. It gathers both the layers and the kinds of ignorance onto one concrete face, creating one person to keep on the desk while we design the first screen.

We first need to distinguish two things. A persona with many attributes is different from a persona that changes decisions. No matter how densely we fill Kim Ji-eun’s boxes, if none changes a single decision on the first screen, the persona is decoration. Her height, neighborhood, and love of coffee say nothing when we decide what to place first. A useful persona may have few fields, but those few must change a decision. One line saying, “This person dislikes failure,” immediately decides whether to show failure first. A persona’s value is measured not by its number of fields, but by which first-screen decision each field changes. We can turn this standard into a verifiable procedure. Beside every field, write one line titled “the first-screen decision changed by this field,” then delete every field left blank. This one deletion rule turns a persona from a document that only gains weight into one that goes on a diet.

Not One Person, but People Entering Through Different Doors

Let us enter a fictional game. It is a mobile game centered on characters, mixing short-form video, collection, and conversation, with light combat alongside them. Whenever we try to draw one person likely to install it, the image keeps slipping, and a different person emerges each time.

One person is a longtime fan of an idol group. They watch every video of their favorite, collect merchandise, and check the idol’s news every day as part of their routine. If this person installs our game, they expect neither combat nor competition. They want to exchange daily greetings with a beloved character, dress the character, and feel their relationship accumulate. Another person is a webtoon binge-reader, accustomed to swiping through one episode on the commute and waiting for the next. This person expects the next part of a story, the tactile rhythm of vertical swiping, and the sense of completion provided by one episode. A third person scrolls through short-form videos endlessly. If fun is not visible within a very short moment, they swipe away. Loading, introductions, and tutorials are all obstructions to this person.

All three install the same game, but they are not the same person. Show them the same first screen and each falls out of alignment somewhere. A sudden combat tutorial bothers the fan. Freedom to become lost burdens the webtoon reader. An introduction that must be watched before anything begins is too long for the short-form viewer. This is what it means to say there are diverse customers within the customer. Our customers are not one kind, but several people who lived in different neighborhoods and entered through different doors.

In MEJE Aidong World too, the fans who meet and bring home Aidong are not one kind. One fan is deeply attached to the favorite idol’s gentle, soft side. Another is drawn to the idol’s lively, mischievous side and cannot feel satisfied without repeatedly eliciting and checking reactions. A fan drawn to a puppy’s softness and a fan drawn to a frog’s bouncy responses stop their eyes at different places even on the same first screen.

Give the Persona the Grain of a Game

To draw a useful persona, first fill the fields found in a general product persona, then add fields unique to games.

Begin with questions asked for any product. What is this person trying to achieve? What makes them uncomfortable? What do they like and dislike? In what environment do they look at the screen: a shaking subway during the commute, a bed before sleep, or a living room while sitting beside a child? These are fields we fill for any app.

Games make people experience failure, set them in competition, and let them own things. They touch the mind more deeply than general apps, so we add another layer. Does this person want to be seen by others, or quietly enjoy themselves alone? What gives them pride, and what makes them ashamed? Do they want to show off, or to belong somewhere? Do they pour attachment into a favorite character, or find satisfaction in operating a system efficiently? Only when these fields are filled does the persona provide decisions. Do not show a proud person a last-place ranking on the first screen. Do not make a deeply attached person treat a character like a tool too quickly.

Then go one step farther. The medium of origin examined in Chapter 5 is the persona’s most important information. An idol fan brings daily companionship with a favorite and dislikes combat and failure. A film or drama viewer expects narrative, direction, and an ending, then becomes flustered when suddenly asked to operate something. A webtoon reader brings a sense of episodes and vertical scrolling, while disliking getting lost or failing. These people bring not the grammar of games, but the grammar of the worlds where they lived. Design the first screen without that grammar, and we disable the actions familiar to their hands while confronting them with something they dislike at the entrance.

We can see traces of real games attending to this point. Solo Leveling: ARISE adapted Solo Leveling, which first gathered a large fan base as a web novel and webtoon, on the assumption that many players would arrive after reading the original. Reviews of the game therefore often praise cutscenes that revive scenes from the original through comic-panel direction. The reason becomes clear when we identify what the original reader brought. They brought not combat controls, but “the next scene in that story.” When the first screen greets them with a familiar scene, they receive a familiar signal from where they lived before learning unfamiliar controls. The opposite treatment of the same origin is common too. Draw in original readers, then open with the same combat tutorial as any other RPG, and people who followed the story feel that what they expected is absent and turn back at the entrance. What separates the two is not visual splendor, but whether the first screen knows where the person came from and what they brought.

Slots and Stats: Promises We Never Stopped to Question

At this point, we must place one long-inherited game convention at a checkpoint: the habit of rewriting personas as “character slots” and “stats.”

Game designers unconsciously use the language of games when classifying people. This character is a tank, that one a damage dealer. This user is payment tier X, that user has combat power Y. Even personas are divided into slots such as “warrior-type user” and “collector-type user.” This convention came from party composition in RPGs, where roles are distinct: the tank blocks the front, the damage dealer attacks from behind, and the healer restores health. It is a neat framework of divided labor and works well inside a game.

The problem is that ordinary people do not share this agreement. Tell an idol fan, “You are the collector type,” and they may feel insulted. They do not think they “collect” something; they think they spend time with someone they love. A gamer calls the same action collection, while a fan calls it affection. The moment we place this person in a slot, we see them in language different from the way they see themselves. The first screen is then designed in that wrong language, presenting collection progress bars, assigning grades, and calculating efficiency. What the person actually wanted was a greeting and attachment.

The fee this convention charges ordinary people is exclusion. A person classified in language that is not theirs feels that the game is not meant for them. Keep slots and stats in internal design documents, but do not use them as the first step in understanding someone. Instead of asking, “What type is this person?” ask, “Where did this person live, what did they bring, and what bothers them?” A slot is a tool to take out when operating the system after understanding the person, not when drawing them for the first time. There is another reason to go a step farther: internal language always leaks. Call a fan a “collector type” in internal documents, and one day “collection progress” leaks into the UI, because the words on the screen ultimately come from the words of its makers. Keep system values internal if necessary, but it is safer to change at least the terms for people in internal documents to the words those people use for themselves in advance.

Grouping people into slots misses something else. Even when people appear to share an origin, their participation temperatures differ, as we saw in Chapter 7’s spectator category. Even among webtoon readers, someone who only reads and someone who comments after every episode want different things on the first screen. Our immigration candidates contain this temperature difference too. Write the person’s temperature on the candidate card in one line and we can avoid pushing creation tools at everyone on the first screen.

▶ Three Questions to Apply to My Screen

Does our persona narrow the choices on the first screen, or does it contain only demographics?

Do we see this person as an “immigration candidate,” or stop at a portrait for reporting?

Are we pushing spectators and creators through the same entrance?

If any of the three questions stops us, that persona cannot change even one field on the first screen.

So We See This Person as an Immigration Candidate

Redraw the persona this way and the nature of the paper changes. It is no longer a reporting portrait saying, “This is what our customer looks like,” but a migration plan asking, “What must we prepare to bring this person into our world?”

Think of an immigrant. When a person considers moving to a new country, they do not completely forget where they lived and arrive empty-handed. They bring its language, tastes, and habits. They look to see whether the new country can accept them and whether the ways they know will still work there. Bringing someone into our world is the same. To bring an idol fan, we must let them do in our world what they did where they lived: exchange greetings with their favorite, dress them, and build attachment. Otherwise, they turn back at the entrance, feeling that this place is too different from where they lived.

This weight is what the word “immigration” adds to the word “origin.” Immigration is closer to the one-way decision to move one’s household than to sightseeing, so a person who comes must leave something behind. The time they spend in our game does not appear from nowhere; it is taken from the time previously spent on webtoons and short-form content. The country to which they can return—the content where they lived until yesterday—is always open one finger away. If the first experience feels even slightly awkward, they can return at any time. We must prepare not only a welcome, but first understand what they cut off in order to come to us. Chapter 11 calculates the price of that decision again as the fee of experience.

The decision this chapter leaves us is therefore simple. Choose several immigration candidates our game will bring in, then draw each one: where they lived, what they bring, and what will bother them. We need not fill every one of the seven fields in the design note at the end of the chapter—the four basic fields plus three additions. It is enough if imagining the person immediately tells us what to show first and what never to show on the first screen. If nothing comes to mind, the persona is still only a point in a demographic.

The layers of Chapter 6, the unknowns of Chapter 7, and the immigration candidates of Chapter 8: three chapters have progressively narrowed one person until several concrete faces remain on the desk. These faces are material for the chapters that follow. Chapter 9 decides which world will admit the candidates, and Chapter 10 decides whom to place at the front of the first screen to greet them. When candidates’ needs collide—for example, a fan dislikes combat while a candidate from games wants challenge first—do not resolve the collision within one screen. Separate them at the entrances in Chapter 9. One point can be written down in advance. If you choose several candidates, later do not view first-screen completion as one aggregate rate; split it by candidate. A fan cooling off during combat and a webtoon reader getting lost in freedom become indistinguishable when mixed into one number.

Reference Content

Cases in other media where this chapter’s concepts are visible.

Video Games

Solo Leveling: ARISE: Readers of the original, which first built a large following as a web novel and webtoon, brought not combat controls but “the next scene in that story.” The game greets them on entry by reviving famous scenes with comic-panel direction and cutscenes. Refer to its placement of signals familiar to hands from the medium of origin—panel direction and vertical flow—before control training.
Cyberpunk 2077: Even in the same game, choosing a “Corpo,” “Streetkid,” or “Nomad” origin changes the opening’s first scene, first route, and first dialogue in their entirety. Refer to the branching design that separates the very entrance to the first screen by origin.
Umamusume: Pretty Derby: Sending characters into races, caring for them, and raising affinity unlocks character stories. Fans receive what they do not as “collection,” but as raising their favorite and staying at their side. Refer to the design that wraps the same action in the language of attachment rather than calling it “collection” in game language.
LittleBigPlanet: Although anyone can create a level, only some actually create and share one; the majority come to enjoy what others made. Participation temperatures split into “makers” and “spectators” even within the same origin. Use this as grounds for not pushing creation tools at everyone on the first screen.

Apps and UX

Duolingo’s first entry: It immediately asks whether someone is learning for the first time or already knows some of the language. Beginners go to a basic path, while knowledgeable learners skip the basics through a placement test. Refer to onboarding that divides origin—prior experience—with one opening question and separates the entrances.

The remaining reference content is collected in the “Chapter 8 Appendix” at the end of this chapter (collected as Appendix D in the book edition).

Design Note ▶ Try It Yourself

Draw three to five immigration candidates our game will bring in first. Instead of demographics, write only the following four things.

Where has this person been living (medium of origin)? What action became familiar there? What one thing on our first screen will bother them most? What familiar signal will reassure them?

Add three more fields: the first-entry layer chosen in Chapter 6, the experiencer type chosen in Chapter 7, and what this person stops doing to come to us—the existing content that previously received their time. With the first two fields, Chapters 6, 7, and 8 converge from three separate classifications into three fields on one card. The last field is used for the fee calculation in Chapter 11.

Once those four lines come to mind, add one more: one game term or convention that must never appear for this person on the first screen. For an idol fan, it is a classification calling them a “collector type.” For a webtoon reader, it is freedom that suddenly leaves them lost. This line screens out the point that will bother the person in advance. Finally, write one thing to show first on the first screen for that person. When finished, add beside every field one line titled “the first-screen decision changed by this field,” and delete fields left blank. These slimmed-down cards grow into Appendix A’s “Immigration Candidate Cards.”

If the first button comes immediately to hand when you picture one candidate, the persona is working. If not, it remains a point in a demographic and must be redrawn.

One-line summary: Personas drawn from demographics go into drawers. A useful persona includes a person’s pride, attachment, and medium of origin, enabling decisions about what to show first and what to hide on the first screen. See the person as an “immigration candidate into our world,” and the persona becomes a migration plan rather than a portrait. Slots and stats are tools to use after understanding the person, not while first trying to understand them. Next chapter: We have now chosen whom to bring. Next, we choose the “world” they will enter. From where will we move someone into which of our worlds? We turn to the primary worldbuilding, adjacent worldbuilding, and the principle that a world must be able to admit residents.

Chapter 8 Appendix: Reference Content Collection

This collection extends into cases inside and outside games the chapter’s argument that our customers are not one kind, but several people who lived in different neighborhoods and enter through different doors. A persona must be drawn not as a demographic portrait, but as an immigration-candidate card recording where someone lived and what they brought if it is to change first-screen decisions. In addition to the section’s central examples, cases where the same scene is read differently by origin, entrances separated by origin, and origin asked in the first question are grouped by medium from video games to offline life. Each item ends with “what to examine,” explaining what a designer should observe. There is one reading method: identify what different origins take differently from the same entrance, and which design decision that difference changed.

Video Games

The Witcher 3: Wild Hunt: The first two games used Geralt’s amnesia to introduce the world and its people from the beginning for newcomers. The 2015 third game lets go of that device, beginning with Geralt’s memory restored. It places people and events recognizable only to those who know the novels and earlier games in the opening, while letting context flow through dialogue and recollection so that newcomers can follow. Examine how dual signals coexist at one entrance: beginners receive guidance from a blank slate, while existing fans are welcomed by names they know. Does our first screen also contain signals that two origins can each take away?
Detroit: Become Human: This 2018 interactive drama from Quantic Dream accepts the viewing habits brought by people from film and television—the familiar act of following scenes and identifying with characters—and narrows most controls to choices within scenes and brief inputs. At the end of every chapter, a branching flowchart confirms how one’s choices divided the story. Examine the placement that makes the habits of non-game origins—watching and choosing—the first input, and where it puts branching confirmation, the new flavor available only in games.
Kingdom Hearts: As platform-spanning side stories between main entries came to carry central lore, newcomers who had not played the earlier titles and side stories were often said to get lost from the opening of a main title. Examine it as a counterexample of “continuity lock” that excludes people who enter without an origin in prior knowledge. What does someone who missed earlier installments begin without knowing on our first screen?
Street Fighter 6: This 2023 fighting game places Classic controls for series fans accustomed to command inputs beside Modern controls that produce special moves with one button. It separately welcomes fighting-game newcomers through World Tour, a single-player mode where they roam streets and learn to fight. Examine how the same game separates entrances through control systems and modes by origin, and the convergence path that ultimately brings every entrance into the same competitive world.
Pokémon GO: Before the first screen of this 2016 location-based game stood both fans who raised Pokémon in childhood and people who did not play games but installed it as a reason to walk. Fans received a nostalgic signal from the first starter Pokémon, while walkers could begin with the familiar action of walking over a map. The two origins were said to remain in the same app for different reasons. Examine the structure in which two origins gathered at one entrance settle through different rewards, and whether our first action also works through the habits of a person not from games.

Mobile, Apps, and Services

Spotify’s first entry: Immediately after signup, it asks someone to choose at least three favorite artists, then shows a first home screen filled not with emptiness but with recommendations based on those tastes. It asks in its first question which musical neighborhood this person lived in and turns the answer directly into a first-screen decision. Examine the flow where one question about origin actually changes the first screen, and how many fields in our persona change first-screen decisions.

Film and Television

The post-credits scene in The Avengers (2012): After the film, a short scene says that to fight the Avengers is “to court death,” and Thanos turns and smiles. A comics reader knows that Thanos has long adored the entity Death and reads the implication of the smile; a new viewer sees only the appearance of a new villain. Examine how the thickness of the signal taken from the same scene differs by origin, while the treatment still supports the scene as complete for the person who does not know.
The Harry Potter film adaptations: Book readers expect the realization of scenes they imagined, while viewers meeting the story first through film expect a new story. Two different expectations sit side by side in the same theater. Examine how the priority between reproducing the source and guiding newcomers reveals which origin was chosen as the primary immigration candidate.
Dune (2021): Denis Villeneuve’s adaptation handles in one opening both the proper nouns brought by novel readers and the blank slate of new viewers. Terms flow throughout the screen, but the story itself can be followed through the universal structure of conflict among houses: newcomers receive atmosphere, while knowledgeable viewers receive depth. Examine what supports the same proper noun so that it welcomes one origin without becoming memorization for another.
Top Gun: Maverick: The 2022 film forms a complete story for viewers who do not know the 1986 original, while layering separate signals for fans, including Goose’s son Rooster and old photographs. Examine how dual design crosses a gap of decades by giving fans additional rewards without blocking newcomers, and where fan signals are placed so they do not obstruct new viewers’ understanding.
The first episode of HBO’s The Last of Us: The 2023 adaptation lengthens the daily life of Joel’s daughter Sarah, which the game passed over briefly. Viewers who did not play the game first become attached to the character, while game fans revisit a known ending from another angle. Examine how length and viewpoint were rearranged to fit the same opening to two origins, and what our opening lengthens and shortens for each origin.
Game of Thrones: Readers of the novels watch direction and adaptation while knowing what comes next. Television-only viewers experience the developments for the first time. The same episode becomes a subject of comparison for one side and a shock for the other. Examine how, especially in episodes where the two origins’ reactions diverge, what counts as a spoiler and what counts as a reward reverses by origin.

Animation

Demon Slayer: Kimetsu no Yaiba – The Movie: Mugen Train: The 2020 film continues directly after the first television season. Someone who watched the season arrives knowing Nezuko’s situation and the hierarchy of the Demon Slayer Corps, so the emotional line functions immediately. For someone seeing the work for the first time in a theater, the weight of relationships and tension remains blank in the same scenes. Examine the information gap on the same opening scene between origins that saw the previous installment and those that did not, and the judgment of how much guidance to add to bridge it.
Arcane: This 2021 Netflix animation centers the story on two sisters so that viewers who have never played League of Legends can watch without prior knowledge, while giving game fans separate rewards in recognizing familiar champions and place names. Examine dual design that gives additional signals only to knowledgeable viewers without blocking those without an origin, and the line that keeps those rewards from becoming conditions for understanding the main plot.

Music

K/DA’s “POP/STARS”: Riot created this virtual girl-group song with League of Legends champions during the 2018 World Championship. For game fans, it was the event of familiar champions taking the stage; for K-pop and pop listeners who did not know the game, it was simply a good new song. Two origins gathered at the same song through different entrances. Examine the condition under which one piece of content satisfies the differing expectations of two origins: completeness that makes neither side’s prior knowledge mandatory for appreciation.

Publishing and IP Expansion

Film novelizations and game tie-in novels: This is a longstanding practice of moving readers who first encountered a work on screen into the entrance of prose. Readers turn pages already knowing the scenes and actors’ faces. Examine how expectations created in another medium—familiar scenes and voices—work on the first page of the new medium, and what mental image someone moving to our medium will bring.
Drama and game adaptations of original webtoons: Readers accustomed to episodic rhythm and vertical scrolling move into another medium. They arrive knowing what comes next in the story, but must leave their familiar reading action behind. Examine how separating what original readers bring—the expectation of the next scene—from what they leave behind—the habit of reading—produces the fields of the immigration-candidate card directly.
IP expansion from original web novels: Expectations collide when readers accustomed to cathartic developments and fast pacing are brought into video or games. Video slows the rhythm, while games insert control. Examine how changing media creates a new source of friction even in the same story, and whether it can be written in advance in the candidate card’s “what will bother them” field.
Board games based on video games: Fans who first met a work digitally are welcomed by the initial setup of physical components. People must now handle rule processing that the screen performed for them, so the first experience has a different cost structure even within the same IP. Examine how someone who likes the same world gets stuck at a different place in initial setup when the medium changes.

Offline and Everyday Life

A first consultation at a gym: A trainer first asks a new member about exercise history. The opening program differs depending on whether the person has never exercised, is returning after a break, or came from another form of exercise. Examine how one question about origin immediately changes the first program—the first screen—making it the smallest immigration-candidate card, and whether the first question in our onboarding changes decisions as much.