Overall UXE Score
7.96
15 jurors · −0.2 vs season average
- Recognition
- Featured
- Recognized
- 22 Jul 2026
- Season
- Quarter 3, 2026
- Studio
- Casa Nuvola, Milan
- Categories
- Mobile Experience · Consumer · AI & Emerging Experience
- Platforms
- iOS, Android
01/Overview
Language learning built around real conversations with an AI tutor that corrects gently and in context.
Parla gets learners speaking from the first lesson. Short exercises build vocabulary, but the heart of the app is a spoken conversation with Giulia, an AI tutor who adapts to your level.
Corrections arrive inside the conversation — a gentle “small tip” rather than a red cross — so learners keep talking instead of freezing.
Streaks reward speaking minutes, not taps, and pause automatically on holidays you set in advance.
- Client
- In-house product
- Audience
- Adults learning Italian, Spanish or French for travel and relationships
- Industry
- Education
- Platforms
- iOS, Android
- Status
- Live since February 2026
- Markets
- Worldwide
- Languages
- 4 — English, German, Dutch, Polish
02/Honors
Recognized by the editors.
Recognition
Featured
Meets the UXE standard. Earns a public profile, a badge and a place in the quarterly awards.
Conferred by the editors on 22 Jul 2026 · minimum UXE 7.0
Awards · Quarter 3, 2026
Season results- Mobile ExperienceEntered
02 / 12 · Consumer
- AI & Emerging ExperienceEntered
11 / 12
Competing in Q3 2026. Entries close 30 Sept 2026; results are announced 15 Oct 2026.
03/Jury verdict
In the jury’s words.
“Parla gets people talking. Gentle, in-conversation corrections are a genuinely better learning pattern.”
Priya Raman
Head of Product Design, Tessera Cloud · India
On behalf of 15 jurors
04/Score breakdown
How the jury scored it.
- Overall UXE Score−0.2 vs Q3 2026 average
- 7.96
- Strongest criterionLearnability
- 8.6
- Jury agreementaverage spread ±0.42
- Moderate
- Jurorsof Q3 2026’s panel
- 15
- 8.2−0.2
Usability
Can people complete what they came to do — efficiently, confidently and without error?
- 7.7−0.4
Information Architecture
Is content organized, labeled and navigable in the way people actually think?
- 8.1−0.3
UI & Consistency
Is the visual language coherent and purposeful across every screen and state?
- 8.4+0.2
Interaction & Feedback
Does every action respond clearly, promptly and meaningfully?
Jury noteCorrections keep the conversation going instead of stopping it.
- 7.5−0.6
Performance
Does it feel fast — perceived and measured — on real devices and networks?
- 7.2−0.6
Accessibility
Can everyone use it, including people relying on assistive technology?
Jury noteSwitch-control users struggle with word-chip exercises.
- 8.6+0.3
Learnability
How quickly does a first-time user become a confident one?
Jury noteLearners were holding simple conversations within their first week.
Scores are the mean of 15 jurors. Ticks mark the Q3 2026 average across 6 recognized projects.
05/The jury
15 ballotsEvery vote, on the record.
Each juror scored Parla independently, from 1 to 10 on every criterion. Nothing is weighted by seniority and no score is discarded — the published result is simply their mean.
One juror Jury mean
| Juror | Usab | IA | UI | Inter | Perf | A11y | Learn | Overall |
|---|---|---|---|---|---|---|---|---|
| Priya RamanVerdictHead of Product Design, Tessera Cloud · IN | 8.2 | 8.3 | 8.2 | 9.2 | 8.1 | 7.4 | 9.7 | 8.44 |
| Min-jun ChoDesign Lead, Seoul Form · KR | 8.3 | 8.1 | 8.4 | 8.7 | 7.9 | 7.9 | 9.2 | 8.36 |
| Chloé MartinContent Design Lead, Plume & Co · CA | 8.9 | 8.4 | 8.6 | 8.3 | 7.8 | 7.4 | 8.5 | 8.27 |
| Ana CostaInclusive Design Researcher, Maré Digital · PT | 8.9 | 7.7 | 8.3 | 8.5 | 7.6 | 7.2 | 9.3 | 8.21 |
| Ravi MenonWeb Performance Engineer, Quickline · SG | 8.6 | 8.2 | 8.0 | 8.7 | 7.8 | 7.6 | 8.6 | 8.21 |
| Tomasz WójcikMotion Design Lead, Kadr Studio · PL | 8.0 | 7.9 | 8.1 | 8.9 | 8.0 | 7.6 | 8.8 | 8.19 |
| Sofía MarínAccessibility Lead, Clarosol · ES | 8.1 | 7.4 | 8.1 | 8.5 | 7.6 | 7.3 | 8.6 | 7.94 |
| Yusuf DemirProduct Design Manager, Kuzey Labs · TR | 8.7 | 7.8 | 8.5 | 7.8 | 7.3 | 6.4 | 8.4 | 7.84 |
| Freya HolmPrincipal Researcher, Salt Research · DK | 8.1 | 7.3 | 8.2 | 8.9 | 7.4 | 6.9 | 7.9 | 7.81 |
| Liam O’ConnorAccessibility Engineer, Cardinal Web · IE | 8.5 | 7.2 | 7.8 | 8.3 | 7.5 | 7.0 | 8.3 | 7.80 |
| Camille RouxCreative Director, Atelier Nuit · FR | 7.5 | 7.5 | 8.2 | 8.9 | 7.0 | 6.9 | 8.2 | 7.74 |
| Lea FischerDesign Systems Lead, Kanton Group · CH | 7.7 | 8.1 | 8.3 | 7.7 | 7.2 | 7.3 | 7.9 | 7.74 |
| Lukas BrennerPrincipal Product Designer, Feldwerk · DE | 7.7 | 7.0 | 7.9 | 7.6 | 7.5 | 7.3 | 8.7 | 7.67 |
| Olivia GrantVP of Design, Harbor & Pine · US | 8.1 | 7.4 | 7.7 | 8.1 | 7.1 | 6.9 | 8.4 | 7.67 |
| Marco BelliniSenior Product Designer, Studio Vetro · IT | 7.7 | 7.2 | 7.2 | 7.9 | 6.7 | 6.9 | 8.5 | 7.44 |
| Jury mean | 8.2 | 7.7 | 8.1 | 8.4 | 7.5 | 7.2 | 8.6 | 7.96 |
| Spread (σ) | ±0.43 | ±0.44 | ±0.34 | ±0.48 | ±0.38 | ±0.36 | ±0.48 | ±0.29 |
marks each juror’s area of expertise. Deeper tint, higher score.
01
Submitted
1 Jul
Entry received, free of charge
02
Jury scoring
8 Jul – 20 Jul
15 independent ballots
03
Recognized
22 Jul 2026
Featured
04
Awards
Results 15 Oct 2026
Competing in Q3 2026
06/Screens
3 screensThe product, screen by screen.
07/Case study
From problem to proof.
The challenge
Learners completed thousands of taps in gamified apps but froze in their first real conversation.
The approach
Casa Nuvola rebuilt the learning loop around speaking, with corrections that feel like a friend’s tip and streaks that reward talking time.
Research
Learner interviews
44 learners, including 12 who had quit other apps
Speaking assessments
CEFR-aligned tests before and after 8 weeks, n=90
Prototype sessions
Correction styles tested with 30 learners
Key insights
- 01
Fear of mistakes, not lack of vocabulary, stops people speaking.
- 02
Corrections that interrupt the conversation end the conversation.
- 03
Streaks motivate — until they punish a holiday.
Outcomes
+1 level
CEFR speaking gain
median after 8 weeks, n=90
14 min
Daily speaking time
median active learner
52%
Day-30 retention
vs. 23% category benchmark
Outcomes are reported by the submitting team and reviewed by the jury.
08/Accessibility & performance
Built for everyone, on any device.
Accessibility
Partial WCAG 2.1 AA
Audited by In-house audit, January 2026
7.2
Jury
- Adjustable speech speed for the tutor
- Typed answers accepted as an alternative to speaking
- Captions on all audio
Known gaps
- Word-chip exercises are hard to complete with switch control
- Yellow backgrounds reduce contrast for secondary text
Performance
2 of 4 within budget
Pixel 6a and iPhone 11, median of 30 cold launches
7.5
Jury
| Metric | Measured | Budget |
|---|---|---|
| Cold start | 1.1 s (over budget) | ≤ 1.0 s |
| Speech response latency | 900 ms (within budget) | ≤ 1.2 s |
| Crash-free sessions | 99.85% (within budget) | ≥ 99.8% |
| Install size | 142 MB (over budget) | ≤ 100 MB |
Within budget Over budget
09/Design system
Piazza, the system behind Parla.
- System
- Piazza
- Components
- 46
- Design tokens
- 118
Color
- #FFD84D
Sole
- #1A1A1A
Ink
- #FF5A36
Pomodoro
- #58CC6B
Giusto
- #7DB7FF
Cielo
Typography
Bricolage Grotesque
Interface & display
Google Fonts
UX patterns
- Conversational UI
- Gamification
- Streaks
- Voice input
- Onboarding
10/Credits
The people who made it.
Studio
Casa Nuvola
Milan, Italy
Learning products built around conversation, play and a little Italian warmth.
- Founded
- 2019
- Team
- 11–50
- Recognized
- 1
Team
- Giulia FerriDesign Director
- Luca MorettiProduct Designer
- Chiara ColomboLearning Scientist
- Andrea RicciMobile Lead
- Client
- In-house product
- Industry
- —
Built with
- React Native
- Speech recognition
- LLM tutor
- Lottie
- Figma
→/More from the showcase
Whole seasonMore from Q3 2026.
Ostara Care
Helix & Co. · Germany
8.69
UXE
Kairo
Tessel · Canada
8.64
UXE
Verde
Selva Studio · Brazil
7.87
UXE