Public Language Index

Measuring how European institutions write, and how that is changing.

PRISM™ scores 10 linguistic dimensions across thousands of publicly accessible web pages. The index tracks these scores over time for organisations across sectors.

organisations documents sectors2015–2025

Where organisations stand, and how far they moved

Each dot is an organisation. The horizontal axis shows the PLI score for the selected year. The vertical axis shows change from baseline (2015–2019 average). The dashed line marks the corpus mean. Filter by sector or drag the year to animate.

Sectors
2025
20152025
Loading chart data…
Corpus mean
Zero line (no change from baseline)
Corpus mean
Spread (IQR)
Std deviation
Mean shift
Visible

Sector PLI scores are converging over time

Average PLI by sector, 2015–2025. Lines that were far apart are drawing closer together. The grey band marks the corpus mean ± one standard deviation. The narrowing band is the signal.

Loading chart data…
Corpus mean
± 1 std deviation
EtroAccenture Europe
100%
Linguistic similarity. Was 97% in 2015–2019 (+3 points)
BentleyAccenture Europe
100%
Linguistic similarity. Was 95% in 2015–2019 (+5 points)
EtroBentley
100%
Linguistic similarity. Was 95% in 2015–2019 (+5 points)
EDHECErste Group
100%
Linguistic similarity. Was 87% in 2015–2019 (+13 points)
ARM HoldingsEDHEC
100%
Linguistic similarity. Was 88% in 2015–2019 (+12 points)
PradaChanel
98%
Linguistic similarity. Was 34% in 2015–2019 (+64 points)

Ten sectors, ten trajectories

Each sector responds differently to the same pressures. Some voices are eroding. Others hold steady. Select a sector for the full trajectory, organisation breakdown, and L1 metric detail.

Loading sector data…

What the index shows

Cross-cutting patterns from the corpus. Each card contains its own evidence and visualisation.

Convergence
Two unrelated brands now produce linguistically identical communications

A Milan fashion house and a European management consultancy — organisations with distinct histories, audiences, and product categories — registered 99.8% pairwise similarity on the PRISM™ index in 2023–2025. Their baseline similarity in 2015–2019 was 97.4%. The residual distance between two unrelated voices has effectively closed.

201520202025100%50%0%97.4%99.8%
Pairwise linguistic similarity between Etro and Accenture Europe, 2015–2025.
Temporal
Score compression accelerated sharply after 2022

The interquartile range of PLI scores across the corpus narrowed at a consistent pace from 2015 to 2021. Between 2022 and 2023, the rate of compression doubled. The timing aligns with widespread adoption of generative AI tools in institutional content production.

20152018202120232025Nov 2022
Interquartile range of PLI scores, 2015–2025. Dashed line marks ChatGPT release.
Language
English-language communications show the strongest convergence signal

When the same organisation publishes in multiple languages, the English-language output consistently scores lower on Cultural Markers and Idiom Density than its counterparts in French, Italian, or German. English, as the dominant language of LLM training data, may be more susceptible to algorithmic flattening.

Cultural MarkersENFRITDEIdiom DensityENFRITDE
Three L1 metrics across languages. n = 127 multilingual organisations.
Government
Institutional editorial controls appear to stabilise voice against algorithmic drift

Government bodies shifted an average of 3.1 PLI points from baseline, the smallest movement of any sector. Readability and sentence structure remained consistent across the period, suggesting that formal approval processes, editorial sign-off chains, and established style guides acted as a buffer.

201520202025Corpus meanΔ 3.1 pts over 10 yearsBand = ±1 SD
Government sector PLI with ±1 standard deviation band, 2015–2025.
Methodology
PRISM™ measures 10 linguistic dimensions using statistical analysis of publicly accessible web content. Scores are normalised 0–100.
Read the full methodology →
Data access
Top-level PLI scores by organisation are available for academic and journalistic use under open licence.
Download dataset →
How to cite
Walsh, M. (2026). Public Language Index. CopyMama S.R.L., Milan.
Citation formats →
Analyse your own
Run PRISM™ analysis on your organisation's public communications and benchmark against the index.
Try PRISM™ →