Readiness and decay
Your readiness score is a single 0–100 number that summarises how prepared you are for the exam right now. It updates after every session. It can go up. It can also go down — and it does, automatically, if you stop studying.
What the number means
Readiness is a weighted composite. Five inputs feed it:
| Input | Weight | What it measures |
|---|---|---|
| Domain average | 35% | Mean score across all certification domains |
| Session frequency | 15% | Completed sessions in the last 14 days, capped at 10 |
| Error trend | 15% | Whether unresolved errors are decreasing or growing |
| Mock test average | 25% | Mean score across your mock exam attempts |
| Retention factor | 10% | How recently you last studied |
The result is rounded and clamped to 0–100. 
ARIA recalculates it whenever a session completes or a mock test posts.
The decay rule
The retention factor is the part that punishes inactivity. It works like this:
100 if you studied today, minus 3 points for every additional day since your last completed session, floor 0.
So:
- Day 0 (studied today): retention = 100.
- Day 1 gap: retention = 97.
- Day 7 gap: retention = 79.
- Day 30 gap: retention = 10.
- Day 34+ gap: retention = 0 (and that is the floor).
Retention is 10% of the readiness score, so a one-day gap costs you ~0.3 readiness points. A two-week gap costs ~4 points. A full month off costs ~9 points just from this input — and your session-frequency weight will have collapsed too, so the real loss is larger.
This is not a punishment. It is a model of what is actually happening in your head — spaced-repetition retention drops on a roughly exponential curve, and I would rather show it to you than let you believe a stale score.
Why decay matters more than it looks
A 3-point retention drop sounds small. The damage compounds because:
- Session frequency decays with it. Fewer sessions in the last 14 days → that 15% weight craters too.
- Error trend can flip. If you study after a long gap, your "recent unresolved errors" spike against the older count → trend turns negative → another 15% weight drops.
- Schedule status worsens. Once you fall behind your phase pace, the schedule pill flips from On track to Slightly behind → At risk → Off track.
The first time you see a meaningful drop after a gap, the temptation is to write off the score. Do not. The number is honest about what your future test self will look like if you keep going at this rate. Treat it as the warning it is.
Recovery
Recovery is doable. Run roadmap sessions on consecutive days and:
- Retention climbs back to 100 the day you study.
- Session frequency rebuilds over 14 days as you fill the rolling window.
- Error trend flips positive once you resolve errors faster than you accumulate them.
In practice: a one-week gap recovers in 3–5 days of consistent study. A one-month gap recovers in 2–3 weeks. There is no magic — the same retention curve that punished you also rewards consistent input.
Pass probability and predicted-ready-date
On the dashboard, you may see two derived numbers next to your readiness:
- Pass probability — my estimate that you would pass the certification if you sat the exam today, computed from your readiness, mock-test history, and the certification's passing score.
- Predicted-ready-date — the date by which my model expects your readiness to cross the threshold needed for the pass guarantee, given your current pace.
Both move when readiness moves. Both are informational — they do not gate anything. The actual exam-ready signal is the pass guarantee eligibility flag, which has its own five hard conditions.