What is wrong with Comic Sans? The answer, with research

Comic Sans was drawn in 1994 for cartoon speech balloons, then shipped to every PC on earth. Designers object to its spacing and its misuse; the research on reading and dyslexia says something more interesting.

An orange wall-mounted defibrillator cabinet with the word Defibrillateur printed twice in Comic Sans above and below the device
Photo: Snowdog, Public domain, via Wikimedia Commons

Nothing is wrong with Comic Sans as a tool; a lot is wrong with how it gets used. Vincent Connare drew it at Microsoft in 1994 for cartoon speech balloons, so its loose spacing, wobbly strokes and childlike tone suit a talking dog, not a memorial plaque or a resignation letter. The research adds a twist: its simple letter shapes help some beginning and dyslexic readers, but no study shows it beats plain sans serifs like Arial or Verdana, and the "ugly fonts help memory" result has not held up well.

Key takeaways
  • 1994: Vincent Connare drew Comic Sans for a cartoon dog's speech balloons, not for paragraphs.
  • Bundling, not choice, spread it: Windows 95, Publisher, Internet Explorer and Comic Chat.
  • The real fault is register: the most casual tone on any office PC, used on memorials and notices.
  • 882 people in four experiments found no memory gain from a hard-to-read font.
  • For dyslexic readers, spacing (letter spacing near 35% of letter width) matters more than the font.

The Graphic Design Stack Exchange question "What is wrong with Comic Sans?" has two top answers that between them pull in more than 350 votes. The first argues there is nothing wrong with the font itself and blames over-exposure and wrong context. The second lists technical faults: no italic for years, poor kerning, a style that copies neither real hand lettering nor printed comic lettering.

Both are right, and the evidence lets us say how right. This post takes the font apart in four layers:

  1. where it came from;
  2. what is wrong with its construction;
  3. what reading research says;
  4. where it is still the correct choice.
1994year Vincent Connare designed Comic Sans at Microsoft
44%of teachers in a 2020 TES poll said they used it
882people across four experiments that found no memory gain from a disfluent font (Taylor et al., 2020)
35%of average letter width: the letter spacing the British Dyslexia Association recommends

Where did Comic Sans come from?

1994Vincent Connare drew it at Microsoft
1996Comic Chat shipped with Internet Explorer 3.0
1 dogRover, the speech-balloon speaker it was drawn for

Comic Sans was designed by Vincent Connare at Microsoft in 1994, after he saw a beta of Microsoft Bob, a cartoon-style home interface for new computer users, putting a dog's speech balloons in Times New Roman.

The story

In 1994 a Microsoft program manager, Tom Stephens, handed Connare a CD of a beginners' program code-named Utopia, later sold as Microsoft Bob, and asked what he thought of its type. A cartoon dog called Rover walked on screen and spoke in balloons set in Times New Roman, the system font. Connare thought that was lazy: a talking dog should not sound like a newspaper. He drew a looser alphabet with his mouse, modelled on the lettering in Watchmen (Dave Gibbons) and The Dark Knight Returns (John Costanza), which he kept in his office, and mocked up Rover speaking in it. He has said he never set out to make a font. The team liked it and asked him to finish it, but Bob was in final beta and could not change its default. Comic Sans shipped first in 3D Movie Maker instead.

Dezeen: interview with Vincent Connare (2014)

The font missed Microsoft Bob's ship date, so its real spread came from bundling:

  • the Windows 95 Plus! pack, then a system font on OEM copies of Windows 95;
  • a default in Microsoft Publisher and Internet Explorer;
  • Microsoft Comic Chat (1996, with Internet Explorer 3.0), which turned IRC conversations into comic strips in front of millions of new internet users.
It was never chosen by the world; it was pre-installed on it.
1994Drawn forMicrosoft Bob1995Windows 95Plus! pack1996Comic Chatwith IE 3.01999Ban Comic Sanscampaign starts2011Comic Sans Pro(Monotype)2012CERN Higgsslides2014Comic Neue2026Comic Chatsource released
Thirty-two years of Comic Sans. Its spread came from software bundling; the backlash followed within five years.

How the backlash grew

The backlash was organised early. Dave and Holly Combs, two designers in Indianapolis, conceived the "Ban Comic Sans" campaign in 1999 and ran it for two decades. In July 2012 the font became a global joke when the ATLAS team's slides announcing evidence for the Higgs boson at CERN were set in it.

Monotype's Terrance Weinzierl expanded the family as Comic Sans Pro, released on 1 April 2011, adding the italics, small capitals and swashes the original lacked. The release date made many people assume it was a prank.

What is technically wrong with Comic Sans?

Judged as a typeface for continuous reading, Comic Sans has construction faults: inconsistent letter widths and slants, a heavy and nearly uniform stroke, loose and uneven spacing, and, until 2011, no true italic or bold italic. These are the same tests the second Stack Exchange question uses to separate a well made font from a poor one.

The mechanics type designers check

  • Even spacing, so no word looks blotchy;
  • consistent stroke weight across glyphs;
  • careful kerning on pairs like "AV" and "To";
  • correct overshoot, so round letters like O and C look the same height as flat ones;
  • a coherent x-height.

Comic Sans was drawn quickly, by its designer's own account, for short on-screen captions, not tuned over years for paragraphs. At 11 point in a long document, its rhythm breaks: some letters lean, some stand upright, and the gaps between them vary enough that words lose their shape.

Quality testWhat a text face doesWhat Comic Sans doesDoes it matter for a sign?
Spacing rhythmEven grey texture across a paragraphLoose, uneven gaps; blotchy in long textLittle; signs are a few words
Stroke contrastPlanned thick-thin or deliberate monolineHeavy monoline that imitates a felt-tip penNo; helps at a distance
Slant and widthConsistent across glyphsSlight, uneven lean on some lettersAdds informality
FamilyRegular, italic, bold, bold italic at minimumRegular and bold until Comic Sans Pro (2011)Rarely
RegisterNeutral, or chosen to match the messageStrongly childlike and playfulYes: this is the real problem

The last row is what people mean when they say Comic Sans is "wrong". Every typeface carries a tone. Comic Sans carries the strongest casual tone of any font installed on most office computers, so when it lands on a cancer clinic notice, a war memorial or a legal letter, the mismatch is obvious to anyone.

Wikipedia's list of famous misuses includes a 2012 Dutch Second World War memorial in Geffen with names set in Comic Sans, and the 2010 letter from the Cleveland Cavaliers' owner after LeBron James left.

A red panel on a white delivery van with a garden design company's name in bouncy yellow letters that jump up and down along the line
A garden design firm's van in the Netherlands, photographed by a type designer who joked that Comic Sans would have been too formal. Novelty lettering is fine for a van; the trouble starts in paragraphs.Photo: Nina Stössinger from Den Haag, Netherlands, CC BY-SA 2.0, via Wikimedia Commons

Why do designers hate Comic Sans so much?

Designers object for three reasons that are easy to separate:

ReasonWhat kind of problemIs it the font's fault?
Over-exposureSocialNo: bundling made it everywhere
Poor fit to serious contextsMeaningPartly: its tone is fixed, the choice is the user's
Craft faults (spacing, slant, family)ConstructionYes

Over-exposure

Over-exposure is the main driver. A neutral sans serif like Arial or Helvetica is invisible to most readers, so nobody notices it in a thousand places. Comic Sans is distinctive, so every misuse is noticed and remembered.

Dave Gibbons, whose lettering partly inspired it, called it ugly in a 2009 interview. Its own designer has repeatedly said he has hardly ever used it, and called the global contempt mildly amusing. There is also a class edge to the dislike: in 2019 the writer Lauren Hudgins argued that mocking Comic Sans users is ableist, because many people with dyslexia choose it on purpose.

The credibility cost

The fit problem has data behind it. In 2012 the film maker Errol Morris ran an online experiment for The New York Times: readers saw the same passage in one of six typefaces and were asked whether they believed a claim in it.

Per the Wikipedia summary of his write-up, readers who saw it in Comic Sans were slightly less likely to agree than those who saw it in the other five. A font that makes a true statement look less credible is a real cost for anything serious.

Does Comic Sans help people with dyslexia?

35%letter spacing as a share of letter width
1.5line height to try before swapping fonts

It can, mainly because of its simple letter shapes and generous spacing, but controlled studies show no special benefit over ordinary sans serifs, and fonts built specifically for dyslexia also fail to beat Arial. The most useful lever turns out to be spacing, not letterform.

The features that help

  • A single-storey "a" and "g", the way children learn to write them.
  • Serifs on capital I, so "Il1" is less ambiguous.
  • Fewer mirror-image letters than geometric fonts, where "b", "d", "p" and "q" are the same shape flipped.
  • Wide spacing, which reduces crowding.

That is why it is popular in schools: in a 2020 Twitter poll by the TES, 44% of teachers said they used it. The British Dyslexia Association's style guide names "Arial and Comic Sans" as examples of readable sans serifs, alongside Verdana, Tahoma, Century Gothic, Trebuchet, Calibri and Open Sans.

double-storey asingle-storey adouble-storey gsingle-storey gSimpler shapes match handwriting taught in school
Letterform close-up. Comic Sans uses the single-storey forms on the right of each pair, which beginning readers recognise from handwriting. Drawn from the descriptions in Wikipedia and the BDA guide; not traced from the font.

A myth about the Rello study

What the evidence does not show is that Comic Sans beats the alternatives. A widely shared claim says the 2013 eye-tracking study by Luz Rello and Ricardo Baeza-Yates found Comic Sans best for dyslexia.

Summaries of that study (48 readers with dyslexia, 12 font conditions) list the families tested as Arial, Times, Courier, OpenDyslexic, Verdana, Garamond, Helvetica, Myriad and Computer Modern, plus some italic versions; Comic Sans was not among them. Their recommendation was Helvetica, Courier, Arial, Verdana and Computer Modern, and italics did worst.

The open studies on purpose-built dyslexia fonts are clear. If a font designed only for dyslexia does not help, a cartoon font is unlikely to do better on letter shape alone.

Dyslexie font does not benefit reading in children with or without dyslexiaKuster, van Weerdenburg, Gompel and Bosman · Annals of Dyslexia · 2018

Two experiments in the Netherlands. In the first, 170 children with dyslexia (about 7 to 12 years old) read text in Dyslexie and Arial; reading speed did not differ (F < 1) and accuracy differences were not significant. In the second, 102 children with dyslexia and 45 without read word lists in Dyslexie, Arial and Times New Roman; font had no significant effect on words read per minute (p = .30). Only 11.6% preferred Dyslexie; 38.1% preferred Arial. Limit: horizontal letter spacing was not matched across fonts, and the materials were Dutch.

Read the paper (free)
The effect of a specialized dyslexia font, OpenDyslexic, on reading rate and accuracyWery and Diliberto · Annals of Dyslexia · 2017

Twelve students with dyslexia in grades 3 to 6 at a US school read letters, real words and nonsense words in OpenDyslexic, Arial and Times New Roman, in a single-subject alternating design. OpenDyslexic did worse on every measure: for word-reading fluency its improvement rate difference was about -89% against Arial. None of the students preferred it. Limit: a very small sample from one school, and comprehension was not tested.

Read the paper (free)

The Kuster paper also cites an English study by Marinus and colleagues (2016) in which Dyslexie's small advantage, about 7% more words per minute, disappeared once letter and word spacing were matched. The lesson for Comic Sans is the same: much of what readers feel as "easier" is spacing. That is why the BDA puts numbers on spacing rather than on letter shape.

Use this to check your own text settings

The British Dyslexia Association's style guide recommends:

Enter your body size and the average letter width of your font as a fraction of the em (around 0.5 is typical for sans serifs; measure your font if you can) to get the spacing values.

Dyslexia-friendly spacing from the BDA style guide

At 18 px with a 0.5 em average letter width, the guide's target works out to about 3.2 px of tracking (0.175 em) and word gaps of at least 11 px. That is far looser than most body text on the web, and it is roughly the open texture Comic Sans gives by default.

If you want a dyslexia-friendly page without the cartoon tone, set Verdana or Arial with these values. The mechanics of tracking and leading are in kerning, tracking and leading explained.

Do harder-to-read fonts like Comic Sans help memory?

28adults in the original lab study
86.5%recall with hard fonts (vs 72.8% with easy)
222high school students in the field study
882people across four larger experiments: no gain

One famous study said yes, and later, larger studies mostly say no. The original result is real but fragile; it should not be used to justify setting study material in an awkward font.

Fortune favors the bold (and the italicized): Effects of disfluency on educational outcomesDiemand-Yauman, Oppenheimer and Vaughan · Cognition · 2011

In a lab study, 28 adults aged 18 to 40 spent 90 seconds learning facts about fictional aliens, were distracted for 15 minutes, then tested. Those who read the facts in an easy font answered 72.8% correctly; those who read them in harder fonts answered 86.5%. A field study with 222 high school students in Chesterland, Ohio, set class materials in Haettenschweiler, Monotype Corsiva or Comic Sans Italic, and those students did better on normal class tests. Limit, stated by the authors: a font that is too hard would hurt, and weaker or easily discouraged students might give up.

Read the authors' summary (free)
Recall after 15 minutes, easy vs hard-to-read fonts (lab study)
See the data as a table
Correct answers (%)
Easy-to-read font72.8
Hard-to-read fonts86.5

Source: Diemand-Yauman, Oppenheimer and Vaughan (2011), Princeton summary · 28 adult volunteers. Later, larger replications with other disfluent fonts did not find a memory benefit.

Disfluent difficulties are not desirable difficulties: the (lack of) effect of Sans Forgetica on memoryTaylor, Sanson, Burnell, Wade and Garry · Memory · 2020

Four experiments with 882 people tested Sans Forgetica, a font built to be hard to read so that readers remember more. Readers did rate it harder to read than Arial (3.04 vs 2.32 on the difficulty scale), but memory was equivalent and, for some word-pair tests, worse. The authors conclude that feeling disfluent is not the same as learning more. Limit: online samples and one specific font; a 2021 study found a small benefit only when people did not expect a test.

Read the paper (free accepted manuscript)

Put the two together and the disfluency case for Comic Sans collapses. The 2011 study used small samples and several odd fonts at once; the 2020 work, with about 30 times as many people, finds that making text harder to read mostly just makes it harder to read.

For dyslexic readers, who already spend effort decoding, adding difficulty on purpose is the opposite of what they need.

Hand-painted signs in looping brush letters for a food mart in New Orleans, advertising fish, chicken and hamburgers above a plastic table and chairs
Real hand lettering on a New Orleans food mart, 2023. Each sign has a living brush rhythm that Comic Sans, a fixed digital imitation, cannot reproduce.Photo: Infrogmation, CC BY-SA 4.0, via Wikimedia Commons

When is it fine to use Comic Sans?

Use it where its tone and its simple letter shapes are the point: material for young children learning to read, informal notes, comic speech balloons, and documents for a reader who asks for it. Avoid it where trust, grief, authority or money are involved.

ContextVerdictWhyBetter alternative if not
Early-reader worksheetsFineSingle-storey a and g match taught handwritingAndika, Sassoon-style school fonts
Reader with dyslexia who prefers itFinePreference and comfort matter; BDA lists itVerdana or Arial with BDA spacing
Comic speech balloonsWeakIt was built for this, but real comic fonts letter betterA dedicated comic lettering font
Medical or safety noticesAvoidCasual tone undercuts urgencyA neutral sans serif
Memorials, condolence, legal or financial lettersAvoidTone mismatch reads as careless; lower credibility in the Morris testA serif or a quiet sans
Brand identityAvoidInstalled everywhere, so it cannot be distinctiveA licensed or custom face

The cover photo of this post, a defibrillator cabinet in Monaco labelled in Comic Sans, is the clearest case of the last rows. The word still reads, but nobody designing for a cardiac emergency would choose a cartoon voice.

If you want the friendliness without the baggage, Craig Rozynski's Comic Neue (2014) redraws the idea with more even spacing and construction. For Indian-language work, where many scripts have few good free options at all, see Indian script fonts on Google Fonts, ranked and how Devanagari fonts work.

Is Comic Sans a bad font, or just a misused one?

It is a competent novelty face that was given the job of a text face. By the craft standards that separate well constructed fonts from poor ones, it is below average for paragraphs. By the standard of what it was asked to do, caption a cartoon dog in 1994, it did its job, and its wide spacing and plain shapes accidentally made it useful for children and many dyslexic readers.

That is the full answer to the Stack Exchange question. The font is not evil; it is loud. Loud is fine on a lunch-box label and wrong on a grave. When a designer says "never use Comic Sans", read it as "never use a font whose tone fights the message", which is advice that applies to every typeface. The same reasoning explains why some logos age badly: see why logos look dated. For the wider world of Kala, the studio, start at Kala.

Questions people ask

Who designed Comic Sans and why?

Vincent Connare, a typographic engineer at Microsoft, in 1994. He thought Times New Roman looked too formal in the speech balloons of Microsoft Bob, a cartoon interface for beginners, and drew a casual font based on comic book lettering from Watchmen and The Dark Knight Returns.

Is Comic Sans the best font for dyslexia?

No study shows that. It is acceptable, and the BDA lists it among readable sans serifs, but controlled studies of dyslexia fonts find no advantage over Arial or Verdana, and spacing matters more than letter shape. Let individual readers choose what they find comfortable.

What is the difference between a well constructed and a poorly constructed font?

A well made font has even spacing and kerning, consistent stroke weights, correct overshoots on round letters, a sensible x-height, a full family of weights and styles, and good coverage of punctuation and accents. A poor one looks blotchy in paragraphs even if single letters look fine.

Why does Comic Sans look unprofessional?

Because its tone is childlike and because it is installed everywhere, so readers associate it with amateur notices. In Errol Morris's 2012 online experiment, readers were slightly less likely to believe a statement set in Comic Sans than in five other typefaces.

Sources

  1. Wikipedia: Comic Sans (origin, release history, misuses, TES poll, Morris experiment summary, Comic Neue)
  2. Dezeen: Vincent Connare interview, "People who don't like Comic Sans don't know anything about design" (2014) (source for the story: Tom Stephens, Utopia, Rover, 3D Movie Maker)
  3. Graphic Design Stack Exchange: What is wrong with Comic Sans?
  4. Graphic Design Stack Exchange: What is the difference between a well constructed and a poorly constructed font?
  5. Kuster et al.: Dyslexie font does not benefit reading in children with or without dyslexia, Annals of Dyslexia (2018)
  6. Wery and Diliberto: The effect of a specialized dyslexia font, OpenDyslexic, on reading rate and accuracy, Annals of Dyslexia (2017)
  7. Princeton: summary of Diemand-Yauman, Oppenheimer and Vaughan, Fortune favors the bold (and the italicized), Cognition 118 (2011)
  8. Taylor et al.: Disfluent difficulties are not desirable difficulties, Memory 28(7) (2020)
  9. British Dyslexia Association: Dyslexia friendly style guide (2023 edition; spacing and font recommendations)
  10. BYU Editing Research: Which fonts are best for dyslexia? (summary of Rello and Baeza-Yates, 2013)
  11. Dyslexia.com: Good fonts for dyslexia, an experimental study (fonts tested in Rello and Baeza-Yates, 2013)
  12. Monotype: How a dog talks, the surprising redemption of Comic Sans (2026)
  13. Microsoft: Comic Chat source code repository (2026)
  14. Creators Syndicate: Marco Rubio, have you considered Comic Sans? (2025)
comic sanstypographyVincent Connaredyslexia fontsdisfluencyreadabilityletter spacingBritish Dyslexia Association
Sapta

Sapta (Saptarshi Sarkar) is a mechanical engineer turned marketer in Bengaluru and a solo touring rider. He writes the seven Saptaverse blogs. Work with him or read more of Artblog.

Keep reading in Artblog

From the other blogs