State of Learning at PrepEx

This report summarizes real product usage and scoring trends across reading, listening, speaking, and writing on PrepEx.

Published Data snapshot Updated

Quick findings

  • 81.3% of repeat practice test takers improved total score.
  • 62.1% of Academic Text learners gained 10+ points out of 100.
  • 79.4% of Listen and Repeat learners improved on latest attempt.
  • 78.6% of writers improved best score versus first attempt.
81.3%
Practice test takers improved total score
62.1%
Academic Text learners gained 10+ points out of 100
79.4%
Listen and Repeat learners improved on latest attempt
78.6%
Writers improved their best score versus first

Methodology

Data snapshot: Feb 13, 2026 (UTC)
  1. Data from anonymized event and submission logs in PrepEx Database; personal identifiers removed before analysis.
  2. Repeat learners only (2+ attempts); outcomes compare each learner's first score to their latest or best score.
  3. Paired before-after comparison per learner, aggregated across sessions.
  4. Practice counts include reading, listening, speaking, pronunciation, and writing submissions.
  5. Observational data; no controlled experiment or causal claims.

1. Outcomes

Academic Text Reading

+48.7% relative lift to best score
62.1% gained 10+ points out of 100
90+ score rate rose from 34.8% to 87.4%

Complete the Words

+31.2% relative lift to best score
64.8% gained 10+ points out of 100
90+ score rate rose from 46.2% to 88.1%

Listen and Repeat

79.4% improved on latest attempt
87.2% improved versus first at best score
2.5+ proficiency threshold rose from 32.9% to 72.6%

Full Practice Tests

81.3% improved total score
54.2% gained 20%
Listening improved for 56.7% of repeat test takers

2. Consistency and Skill Breadth

1-4 attempts

19.2%
Used 2+ skills

5-9 attempts

72.4%
Used 2+ skills

10+ attempts

93.1%
Used 2+ skills

3+ skill usage

1.9% to 79.2%
From 1-4 attempts to 10+ attempts

All 4 skills usage

0.2% to 51.8%
From 1-4 attempts to 10+ attempts

Lift in all-4-skill usage

+51.6 pts
All-4-skill usage moved from 0.2% to 51.8% between low and high practice intensity.

3. AI Coaching Outcomes

Petra Focus

Petra retention signal

43.2%

Petra users who returned for at least 2 coaching sessions.

Interview confidence growth

69.8% / 76.1%

Share improving delivery confidence / language-use confidence at best score.

Writing score threshold shift

17.9% to 52.4%

Share of repeat writers reaching 4+ moved up substantially from first to best score.

Interview score threshold shift

21.3% to 47.2%

Share of repeat interview speakers reaching 20+ grew from first to best score.

Frequently asked questions

What does this report measure?
It measures before-after performance trends for repeat learners across reading, listening, speaking, and writing, plus usage breadth and Petra coaching signals.
Is this a controlled experiment?
No. This is observational data from anonymized product usage logs and submissions. It reports correlations and trends, not causal proof.
Who is included in the data?
Free and Pro users are both included. The outcome metrics are based on repeat learners with at least two attempts.
How often is this report updated?
This release is based on a Feb 13, 2026 UTC snapshot. Future releases will publish new snapshot dates and revised trend metrics.

See how this maps to your plan

Free and Pro users both appear in this dataset. Pro unlocks higher practice volume, more Petra time, and unlimited AI feedback loops.

State of Learning summary data

Practice test takers improved total score: 81.3%.

Academic Text learners gained 10+ points out of 100: 62.1%.

Listen and Repeat learners improved on latest attempt: 79.4%.

Writers improved best score versus first attempt: 78.6%.

Method: anonymized repeat-learner before-after analysis, data snapshot Feb 13, 2026 UTC.