Air Force Score Calculator
Calculates application scores by condition and compares them with past cutoffs. Over 10,000 monthly active users (MAU) use it. The mobile input screens were improved using usage records and interviews.
Role
A 1-person service, handling planning, development and operation
Problem
Applicants had to calculate their scores by hand from the recruitment guidelines and compare them with past cutoffs
Solution
- Separate configurationSeparated recruitment stage, major, certificate and bonus point criteria into JSON configuration so changes to the guidelines are applied without code changes, and ran the calculation on the client only
- Range guidanceProvided safe, fair and risky ranges from the cutoffs and trends of the last 2 years
- UsabilityFound and fixed buttons that were hard to reach with a thumb and dialogs that were hard to close, using Clarity heatmaps, session replays and interviews with 2 users
Result
10,000+ MAU and about 28,000 page views
Overview
Air Force applicants had to find their own conditions in the recruitment notice, calculate a score, and then compare it with past cutoffs. A score calculation service was built so that this whole process can be done in one place. The flow is: check the recruitment notice, enter conditions, calculate the score, and compare with past cutoffs.
This was a personal project covering planning, development, deployment and operation. Reflected changes to the recruitment criteria, watched how people actually used it, and improved the mobile input screens.
Key features
- Entering recruitment conditions: Select the recruitment stage, major, certificates and bonus items
- Score calculation: Calculates the score for the entered conditions in the browser
- Comparison with past criteria: Gives guidance as a safe, fair or risky range, using the cutoffs by specialty over the last 2 years, the results for the same month last year and the trend over the last 3 months
Calculation and prediction run in the browser, and the entered conditions are not sent to a separate server for calculation.
Implementation and improvements
-
Separated the recruitment criteria from the screen.
Defined the recruitment stage, major, certificates and bonus items in JSON, and set up the React screen to draw the input fields from that configuration. The criteria are not tied directly to screen code, so when they change, the input screen can be managed from the matching configuration. Score and reference-criteria calculation run in the browser, and the service is deployed as static pages.
-
Used past results as the reference for application range guidance.
Gave guidance as a safe, fair or risky range by combining the cutoffs by specialty over the last 2 years with the results for the same month last year and the trend over the last 3 months. Instead of comparing only a single cutoff with the score, this lets users look at changes by period and specialty together. The range is shown as reference information for applying, based on past data.
-
Adjusted the mobile screens to the usage difficulties observed.
Looked at Microsoft Clarity heatmaps and session replays and interviewed 2 users. This turned up buttons that are hard to reach with a thumb and dialogs that are hard to close, and these were fixed. Adjusted the position and visibility of the main buttons, and showed selectable items and selected items in different states.
Tech stack
Next.js · TypeScript · Tailwind CSS · Zustand
Microsoft Clarity · GA4
Results and current status
In operation, it recorded 10,000+ MAU and about 28,000 page views.
10,000+ MAU
Recorded 11,280 visitors and 27,977 page views in the last 30 days.
Separated the recruitment conditions from the screen as configuration data so that criteria changes can be handled, and built the flow from checking the score to comparing with past criteria to run in the browser. On mobile, adjusted the buttons and dialogs to the operation problems found through observation and interviews.
Operated it from October 2025 to March 2026, responding to changes in the recruitment criteria.