Brands over time

RAC
over time

Every year keeps its colour across every chart.│ ChatGPT, 30 Nov 2022

193 of RAC's own blog posts, press releases and announcements, 2015 to 2026, measured sentence by sentence, 55 of them recovered from the Internet Archive. How the writing has moved over the years, how it compares with 6 other motoring brands, and what it sounds like in its own words.

Pages measured
193
Years covered
2015–26
Archived before ChatGPT
55
Blog · press
94 · 99

01 · Homepage

One capture a year

RAC's homepage from 2000 to 2025, closest to 1 July, from the Internet Archive. Pick a year or press play. Each headline is quoted from its capture and links to the archived copy.

2025rac.co.uk2 Jul 2025
RAC homepage, 2 Jul 2025

Screenshot of RAC’s homepage (rac.co.uk) as archived by the Internet Archive on 2 Jul 2025. © RAC or its licensors. Shown for comparison and review; not endorsed by RAC.

Homepage

2025

“Complete peace of mindTM”

Page title

RAC Breakdown Cover & Car Insurance | Route Planner | RAC

Open on the Internet Archive
  1. 2008Click to unmute Welcome to RAC
  2. 2010Click to unmute Breakdown cover Special offer
  3. 2014News, Advice and Social
  4. 2017Save up to £30 in our summer sale*
  5. 2021Plan your trip
  6. 2025Complete peace of mindTM

25 captures. Screenshots are rendered from the archive and can pull in images archived later than the page itself; dates come from the capture. Years without a capture are greyed out in the strip.

02 · Measures by year

How the writing moved

Every year’s pages, averaged. Pick a measure. Hatched bars rest on fewer than 10 pages: context, not a finding. The page count sits under each year.

Sentence lengthAverage words per sentence.
6
12
18
Motoring avg 14.3
15.8
12.8
15.8
14.4
17.5
14.4
15.5
13.6
17.8
15.7
14.3
12.6
ChatGPT · Nov 2022
201515 pp
201611 pp
201716 pp
201818 pp
201912 pp
202016 pp
202115 pp
20229 pp
20239 pp
202410 pp
202511 pp
202631 pp
10 or more pagesFewer than 10 pages: context, not a findingSector averageChatGPT released
Show as a table
YearSentence lengthPages
201515.815
201612.811
201715.816
201814.418
201917.512
202014.416
202115.515
202213.69
202317.89
202415.710
202514.311
202612.631

Years with pages: 2015 (15), 2016 (11), 2017 (16), 2018 (18), 2019 (12), 2020 (16), 2021 (15), 2022 (9), 2023 (9), 2024 (10), 2025 (11), 2026 (31). Dates come from page metadata, structured data, the URL, or the month a page was first archived (a latest possible date).

03 · Before and after

2019 to Nov 2022 against 2024 to 2026

The same measures on 52 pages from before ChatGPT launched and 52 pages from 2024 on. A change here is a change in style. It does not show how any page was written: people used every one of these words and shapes long before 2022, and SGR-011 found most of the shift across brands was a trend already under way.

2019 to Nov 2022 · 52 pp2024 to 2026 · 52 pp

Sentence length down 11%

words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Sentence length down 11%. words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Sentence length15.313.6−1.7

Sentence rhythm down 3%

How much sentence length varies (standard deviation over mean). Higher reads less even.

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Sentence rhythm down 3%. How much sentence length varies (standard deviation over mean). Higher reads less even. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Sentence rhythm0.780.76−0.02

Contractions up 43%

per 1,000 words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Contractions up 43%. per 1,000 words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Contractions10.815.4+4.6

"You" up 1 pt

share of I/we/you

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

"You" up 1 pt. share of I/we/you. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
"You"56%56%+1%

"We" up 1 pt

share of I/we/you

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

"We" up 1 pt. share of I/we/you. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
"We"34%35%+1%

Questions up 133%

per 1,000 words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Questions up 133%. per 1,000 words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Questions2.76.2+3.6

Exclamations down 65%

per 1,000 words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Exclamations down 65%. per 1,000 words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Exclamations0.340.12−0.22

Long words down 2 pts

share of words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Long words down 2 pts. share of words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Long words26%24%−2%

Em dashes up 17%

per 1,000 words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Em dashes up 17%. per 1,000 words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Em dashes3.714.33+0.62

AI-associated words down 36%

per 1,000 words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

AI-associated words down 36%. per 1,000 words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
AI-associated words0.400.26−0.15

AI-era marketing words down 50%

per 1,000 words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

AI-era marketing words down 50%. per 1,000 words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
AI-era marketing words0.360.18−0.18

Quieter LLM-favoured words down 22%

per 1,000 words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Quieter LLM-favoured words down 22%. per 1,000 words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Quieter LLM-favoured words1.200.93−0.26

AI-associated sentence shapes down 38%

per 1,000 words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

AI-associated sentence shapes down 38%. per 1,000 words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
AI-associated sentence shapes0.200.13−0.08

Page length up 39%

words

  • 2019 to Nov 2022 · 52 pp
  • 2024 to 2026 · 52 pp

Source: SGR-011

Page length up 39%. words. Source: SGR-011
m2019 to Nov 2022 · 52 pp2024 to 2026 · 52 ppChange
Page length1,3481,870+522

Every chart is centred on its before value and shares one relative scale, so a longer bar means a bigger percentage change; the values themselves are in different units and do not compare. Before = 2019 to Nov 2022 (52 pages); after = 2024 to 2026 (52 pages). Method: SGR-011.

04 · Against its sector

RAC among motoring brands

Each dot is one brand’s mean across all its measured pages; the tick is the sector average. Only brands with 10 or more measured pages are shown or counted. Select a dot to open that brand.

  1. Contractions

    25% less than its sector

    12.2 against a sector average of 16.2; 2nd lowest of 7.

  2. Questions

    24% less than its sector

    3.5 against a sector average of 4.6; mid-table, 4th of 7.

  3. Em dashes

    19% more than its sector

    4.13 against a sector average of 3.48; mid-table, 4th of 7.

Sentence length2nd highest of 7
RAC 14.8
12.916.6
Contractions2nd lowest of 7
RAC 12.2
11.922.6
"You"Lowest of 7
RAC 56%
56%73%
"We"Highest of 7
RAC 34%
22%34%
QuestionsMid-table, 4th of 7
RAC 3.5
2.98.0
Long wordsHighest of 7
RAC 26%
18%26%
Em dashesMid-table, 4th of 7
RAC 4.13
1.745.19
AI-associated wordsLowest of 7
RAC 0.38
0.381.81
ExclamationsLowest of 7
RAC 0.21
0.211.61
Page length2nd highest of 7
RAC 1,384
7535,102

Pages measured: Halfords 22 · Kwik Fit 170 · RAC 193 · cinch 311 · Arnold Clark 141 · carwow 320 · Motorway 194. Measures and method: SGR-010; this sector and its brands were added by Brands over time, measured exactly as the studies measured theirs. Sector averages and ranks cover every brand in the sector on this site.

05 · Vocabulary

Words RAC uses more than its peers

The words this brand reaches for more often than the rest of its sector. This is subject matter more than voice.

insurance60.9 vs 9.0
advice51.9 vs 7.6
drivers73.5 vs 19.6
cover36.3 vs 4.3
breakdown32.3 vs 1.1
drive73.1 vs 26.4
parking27.4 vs 5.1
roads28.6 vs 6.8
warning19.4 vs 2.9
driving70.7 vs 32.0
maintenance26.1 vs 6.6
breakdowns14.8 vs 0.5
motorists17.1 vs 2.9
licence14.0 vs 1.6
RACSector peers

Uses per 10,000 words, ranked by weighted log-odds (Monroe et al. 2008) across 193 pages. Checked for product names and page furniture.

06 · In its own words

What it sounded like

Openings of real pages, quoted word for word, oldest first. Each links to the page, or to the Internet Archive’s copy when it has gone.

202422 Apr · Press
“But the RAC believes drivers may have actually ‘dodged the pothole bullet’ in what is normally the worst three months of the year for them. Milder weather led to patrol call-out rates dropping by 22% from 10,076 last year to 7,904 in 2024.”
RAC pothole-related breakdowns ‘spring’ forward by 50% in 2024

07 · Every page

All 197 pages, by year

The full list behind every number above, with links to the live page or the archived copy. 193 count towards the measures; the rest were not measured (2 under 150 words of text, 2 mostly short lines (lists, tables, link pages) rather than prose). Pages with no reliable date count in the overall numbers but in no year.

202631 pages
202511 pages
202410 pages
20239 pages
20229 pages
202115 pages
202016 pages
201912 pages
201818 pages
201716 pages
201611 pages
201515 pages
In the overall numbers, no year20 pages
Not measured4 pages

08 · Method

How this was measured

The pages

RAC's own blog posts, press releases and announcements, fetched from the live site in 2026 and from the Internet Archive for 2015 to November 2022. Homepages are shown as captures; product and help pages are not measured.

The measures

Code counts from ScriptGrain’s writing instruments: sentence length and rhythm, contractions, pronouns, punctuation, long words, and three published lists of AI-associated words. No AI model reads or judges any page.

What it cannot say

These measures describe style. They cannot tell who or what wrote a page, and nothing here says or implies that any brand used AI to write.

Sample sizes

Every number carries its page count. Years under 10 pages, and before-and-after comparisons with under 10 pages on a side, are shown as context only.

Sentence length
Average words per sentence.
Sentence rhythm
How much sentence length varies (standard deviation over mean). Higher reads less even.
Contractions
Contractions (we're, you'll, it's) per 1,000 words. A rough gauge of how conversational the writing is.
"You"
Share of I/we/you pronouns that are "you": how much the writing addresses the reader.
"We"
Share of I/we/you pronouns that are "we": how much the writing talks about the organisation.
Questions
Question marks per 1,000 words.
Exclamations
Exclamation marks per 1,000 words.
Long words
Share of words with seven or more letters.
Em dashes
Em dashes per 1,000 words.
AI-associated words
Words from SGR-007's three AI-associated lists, per 1,000 words. These are style markers that people used long before ChatGPT; a higher rate says nothing about who or what wrote the page.
AI-era marketing words
Marketing words that became more common in AI-era copy (seamless, elevate, unlock...), per 1,000 words. A style measure only.
Quieter LLM-favoured words
Everyday words language models are known to over-use, per 1,000 words. A style measure only.
AI-associated sentence shapes
Sentence patterns such as "not just X, but Y", found by ScriptGrain's cliche checker, per 1,000 words. A style measure only.
Page length
Average words per page.

Studies: SGR-009: Do brands in the same sector sound the same? · SGR-010: How UK brands write · SGR-011: Did brand writing change after ChatGPT?.