DEV Community

Cover image for Building a 7 deadly sins test without judging the user
Drew Grant
Drew Grant

Posted on

Building a 7 deadly sins test without judging the user

Someone finishes a personality quiz and gets “Pride: 75.” What should they do with that? Does it mean they are unusually proud, that three-quarters of their personality is pride, or simply that they agreed with a few statements?

Working on a 7 deadly sins test made that ambiguity hard to ignore. Calculating a number is straightforward. Giving someone enough context to read it fairly takes more care.

The project uses the 7 deadly sins as themes for an entertainment quiz. The interesting development problem is what happens between an answer, a score, and the sentence a person remembers afterward.

Start with the question behind the search

“Which deadly sin am I?” invites a single, memorable answer. That makes a character reveal a reasonable part of the experience. But I want that reveal to leave room for a second question: “Which of my answers led here?”

For this project, I treat the reader's task as three connected steps: answer recognizable situations, understand the resulting pattern, and decide whether any of it is worth discussing or sharing.

That framing affects the interface. The character needs supporting context. The scores need an explanation. Sharing needs to be a choice.

There is also a scope decision to make early: these are original characters inspired by everyday habits, not matches from The Seven Deadly Sins anime. A familiar phrase should not leave someone guessing which experience they opened.

Make a 7 deadly sins test score explainable

The homepage Quick Quiz has 21 questions, with three assigned to each theme. It asks people to think about the past four weeks, which gives the answer a time frame rather than asking them to summarize their whole life.

Each usable answer has a value from 0 to 4. Some statements point in the opposite direction, so their values are reversed before averaging. Each theme is then scaled independently to 0–100.

Here is a simplified JavaScript illustration of that calculation:

// Illustrative answers for one theme, not a real user's result.
const answers = [
  { value: 3, reverse: false },
  { value: 2, reverse: false },
  { value: 0, reverse: true },
];

const values = answers.map(({ value, reverse }) =>
  reverse ? 4 - value : value
);

const score =
  (values.reduce((sum, value) => sum + value, 0) / values.length) * 25;

console.log(score); // 75: average of 3, 2, and 4, scaled to 100
Enter fullscreen mode Exit fullscreen mode

The arithmetic is easy to inspect. It does not establish that the questions are a validated measure of personality.

Here, 75 describes responses within one question group. It is not a percentile, a probability, or a measure of moral worth. The seven scores do not need to add up to 100.

The other detail worth handling explicitly is missing information. “This situation doesn't apply to me” should not quietly become zero. In the Quick Quiz, those answers are excluded, and each theme needs at least two usable answers before the user can finish.

If you are building something similar, define that rule before designing the result screen. Otherwise, an unanswered question can accidentally become a confident conclusion.

A concrete example: “Which deadly sin am I?”

Imagine someone who often defends their work immediately when corrected. They also keep thinking about achievements that went unnoticed. In this quiz, those situations contribute to the pride theme.

That is enough to make a reading about recognition and defensiveness relevant. It is not enough to explain their relationships, work history, or reasons for reacting that way.

A useful next step is small: the next time feedback arrives, ask one curious question before explaining your side. The person can try it, disagree with it, or decide the situation needs more context.

There is no need to force a winner, either. If several scores are close, the result should make that visible. The project includes mixed, balanced, and quieter outcomes so the headline can acknowledge the pattern instead of flattening it.

You can inspect that answer-to-result flow in the 7 deadly sins quiz I'm building. The Quick Quiz is free, with no account or result paywall. It pairs an original character with seven scores and the responses behind them, so there is something to examine after the reveal.

Sharing deserves its own design decision

A result that feels fine in a private browser tab may feel different in a group chat.

The project lets people make an illustrated card, choose playful or gentle wording, and decide whether to include scores. A result link excludes individual answers, but anyone holding the link can read the selected information. Shared links cannot be revoked.

Those details belong near the sharing action. A person should understand what will leave the page before copying it.

Friend comparison has a similarly narrow purpose: compare two profiles from the same quiz version and start a conversation. It does not calculate compatibility or establish who has the “better” personality.

For me, the useful design constraint is keeping the character, score, and explanation consistent about what they know. This is entertainment and a prompt for reflection. It cannot diagnose someone, provide a religious judgment, or establish a permanent identity.

A 7 deadly sins test can still be playful within those limits. If you want a concrete example to explore, the Quick Quiz puts the questions, explanation, and optional sharing in one flow. The part I would carry into another project is simple: make the result easy to question, as well as easy to share.

Top comments (0)