Prep Signal

See the problem. Fix the problem.

Comment

Inspiration

Standardized test prep has a huge problem: students can do hundreds of practice questions and still not understand why their score is not improving.

Most platforms tell students what they got right or wrong. But they often do not answer the more important questions:

  • Am I repeatedly missing the same skill?
  • Am I rushing?
  • Am I spending too much time on questions I eventually get wrong?
  • Do I perform worse under timed conditions?
  • Am I guessing near the end of a section?
  • Are my mistakes conceptual, strategic, or caused by timing?

I wanted to build something that treats practice data as more than just a score.

That idea became Prep Signal, a free ACT preparation platform designed to turn student performance into actionable feedback.

Instead of simply saying, "You scored 27," Prep Signal tries to answer:

What exactly is holding you back, and what should you work on next?

What it does

Prep Signal combines ACT practice with a personalized analysis system that tracks how students perform over time.

Students can complete practice questions and tests across ACT subjects while the platform analyzes patterns such as:

  • repeated weaknesses in specific skills
  • unusually fast incorrect answers
  • unusually slow incorrect answers
  • high-confidence mistakes
  • performance differences between timed and untimed practice
  • answer-changing patterns
  • possible guessing behavior
  • improvement or decline across topics

The goal is to transform a normal practice session into a detailed learning signal.

Rather than making students manually search through dozens of mistakes, Prep Signal organizes those patterns into reports that help students decide what to study next.

Mistake Intelligence

The main feature I focused on is what I call Mistake Intelligence.

Individual mistakes are not always very meaningful. Patterns are.

For example, missing one geometry question may simply be random. But repeatedly missing geometry questions, answering them much faster than expected, and performing significantly better when untimed tells a much more useful story.

Prep Signal looks across multiple attempts to identify patterns like these.

The system can then help distinguish between problems such as:

Knowledge weakness The student consistently struggles with a specific concept.

Timing weakness The student understands the material but performs substantially worse under time pressure.

Rushing Incorrect answers repeatedly occur much faster than expected.

Overthinking The student spends excessive time on questions and still answers incorrectly.

Confidence mismatch The student feels confident about answers that repeatedly turn out to be incorrect.

This gives students something more useful than another percentage score.

Interactive learning

Prep Signal is also designed to connect diagnosis with learning.

If the system detects that a student is struggling with a particular topic, the student can move directly into targeted practice and learning material instead of searching through an entire ACT book.

For math topics, I experimented with interactive explanations where students can manipulate parts of a problem rather than only reading static text.

The long-term goal is to create a loop:

Practice → Analyze → Diagnose → Learn → Practice again

Every new attempt gives the system more information about the student.

Progress tracking

Prep Signal keeps track of performance across topics and over time.

Students can use the platform to see:

  • topic accuracy
  • practice history
  • timing trends
  • mastery progress
  • recurring mistakes
  • recent improvement
  • areas that need additional practice

I also designed weekly-style reports so students can understand their progress without having to interpret every individual practice session.

This is especially useful for students who know they need to improve their ACT score but do not know how to structure their studying.

Community and accountability

Test preparation can become repetitive and isolating, so I also explored ways to make studying more collaborative.

Prep Signal includes ideas such as student profiles, rankings, study spaces, progress sharing, and other community features that make preparation feel less like completing an endless worksheet.

The objective is not simply competition. It is to give students visible progress and another reason to consistently return and practice.

Prep Signal has also been reviewed and recommended by students at Adlai E. Stevenson High School and Wheeling High School. So far, more than 200 students have taken the diagnostic test, giving me real feedback on how students use the platform and what actually helps them. That feedback has been important in improving the experience, especially the diagnostic system, progress tracking, and the way results are explained. My goal is to keep building Prep Signal around what students actually need, not just what looks good on paper.

How I built it

I built Prep Signal as a web application with a focus on keeping the interface fast, clean, and usable on normal student devices.

The application uses Next.js for the frontend and application logic, with a database-backed system for storing student progress, question attempts, timing data, and performance history.

A major part of the development process involved designing the underlying data model.

For every attempt, the platform needs more than:

correct / incorrect

It can also consider information such as:

  • topic
  • question type
  • difficulty
  • response time
  • whether the session was timed
  • confidence
  • previous attempts
  • historical performance

That information can then be aggregated across many attempts to detect patterns.

I intentionally designed the interface to remain relatively minimal. Educational dashboards can quickly become overwhelming, especially when they contain many statistics. I wanted students to be able to look at a report and immediately understand what mattered.

Challenges

Turning data into useful advice

One of the hardest problems was deciding when a pattern is actually meaningful.

For example, if a student answers one question quickly and gets it wrong, that does not necessarily mean they are rushing.

The platform therefore needs enough evidence before identifying a behavior.

I designed rules that consider minimum attempt counts, repeated occurrences, timing ratios, accuracy differences, and other thresholds before generating certain insights.

That helped prevent the platform from making conclusions based on a single unusual question.

Designing for different students

Two students with the same score may have completely different weaknesses.

One student may lack content knowledge while another understands the material but struggles with pacing.

Building the platform made me realize that personalization is not just recommending different questions. It also means understanding why performance differs.

Building a large full-stack application

Prep Signal became much larger than my original idea.

I had to work through database architecture, authentication, question organization, responsive interfaces, analytics, deployment, performance, and the interactions between many different features.

Each new feature also had to work with the existing student data rather than functioning as an isolated page.

Keeping complex analytics understandable

Another major challenge was presentation.

A system can calculate dozens of statistics, but students should not need to be data scientists to understand them.

I spent a lot of time thinking about how to turn raw information into simple conclusions such as:

You know this topic, but your accuracy drops significantly when timed.

That translation from data to understandable feedback became one of the most important parts of the project.

What I learned

Building Prep Signal taught me that creating an educational product is very different from simply building software.

A technically impressive feature is not valuable if students cannot understand how to use it.

I learned to think more carefully about:

  • product design
  • user behavior
  • educational psychology
  • data interpretation
  • database architecture
  • scalable web applications
  • UI/UX
  • performance
  • iterative development

I also learned how important feedback is.

Features that seem obvious while developing a product may be confusing to somebody seeing it for the first time.

That pushed me to simplify interfaces and focus more heavily on the student's actual workflow.

What's next

I want Prep Signal to eventually function as a complete feedback system for standardized-test preparation.

Some of the areas I want to continue developing include:

  • deeper diagnostic reports
  • more interactive lessons
  • larger practice libraries
  • better long-term progress analysis
  • teacher and tutor dashboards
  • school-level integrations
  • collaborative study tools
  • more personalized study recommendations

I am also working on getting Prep Signal into the hands of more students and schools so that real student feedback can guide the next stage of development.

Ultimately, I want Prep Signal to make high-quality ACT preparation accessible without requiring students to purchase expensive tutoring or preparation programs.

The goal is simple:

Don't just give students more practice. Help them understand what their practice is telling them.

Built With

Share this project:

Updates

Submission history