16 Pixels
For Body Copy. Anything Less Is
A Costly Mistake
I know what you’re thinking. “Did he just say 16 pixels? For body copy? Obnoxiously big! 12 pixels is ideal for most websites.”
I’d like to persuade you otherwise.
As usability expert Oliver Reichenstein says in “The 100% Easy-2-Read Standard”:
[16 pixels] is not big. It’s the text size browsers display by default. It’s the text size browsers were intended to display… It looks big at first, but once you use it you quickly realize why all browser makers chose this as the default text size.
In this article, I’ll explain why 16 pixels should generally be the minimum size for body copy in modern Web design. If I don’t change your mind, perhaps you could chip in at the end and let me know why.
You see, in most cases, if you’re building websites with the font size set between 10 and 15 pixels, you are costing your clients money. And I aim to prove it.
Readership = Revenue
If you’re building a website for someone — even yourself — chances are its purpose is to make money.
Perhaps it’s to sell a product directly, or to offer a service, or just to generate leads. Whatever the case, it’s a business asset, and ultimately it has to generate a return on investment. It has to fulfill a revenue goal.
So, every element should be designed to achieve that goal. Including the copy. Especially the copy — because the copy is what convinces visitors to do whatever it is you want them to do on the website.
Think about it. If you don’t explain what people should do, or why they should do it, then they certainly won’t. And the only way to tell them is with text. And text means reading.
Important Facts About Reading
There are some particular findings that are pivotal to issues such as readership and readability and comprehension, which is really what body copy is all about. If people won’t read it, or if they can’t read it or understand it, then what’s the point of having it?
- At age 40, only half the light gets through to the retina as it did at age 20. For 60-year-olds, it’s just 20%.
- Nearly 9% of Americans are visually impaired, meaning their vision cannot be completely corrected with lenses.
- The distance at which we can read letters is a common measure of both legibility and reading speed. The greater the distance, the higher the overall legibility and comprehension are considered to be. The biggest factor that determines how far this distance can increase is font size. Seen any billboards lately?
- Most people, when sitting comfortably, are about 20 to 23 inches from their computer screens. In fact, 28 inches is the recommended distance, because this is where vergence is sufficiently low to avoid eye strain. This is much further than the distance at which we read printed text — most people do not hold magazines at arm’s length!
- 16-pixel text on a screen is about the same size as text printed in a book or magazine; this is accounting for reading distance. Because we read books pretty close — often only a few inches away — they are typically set at about 10 points. If you were to read them at arm’s length, you’d want at least 12 points, which is about the same size as 16 pixels on most screens:

- In a 2005 survey of Web design problems, bad fonts got nearly twice as many votes as the next contender, with two-thirds of voters complaining about small font sizes. If you think the situation has improved since then, think again. I randomly sampled some new blog designs on SiteInspire and found that the average font size for body copy was a measly 12 pixels. Some designs even used a minuscule 10 pixels. And none were over a weaksauce 14 pixels. Similarly, if you randomly sampled offerings on the popular Elegant Themes and ThemeForest, you’d find that every single theme sets the main content at just 12 or 13 pixels.
Fact: Most Web Users Hate The “Normal” Font Size
With this research in mind, let me ask: how many of your client’s readers are around 40? Because they have to work twice as hard to read as 20-year-olds. If they’re closer to 60, they have to work four times as hard.
Almost 1 in 10 of your readers also has trouble with their eyes. Of the rest who don’t, most will still have to strain to read text smaller than 16 pixels, even if they don’t notice that they’re doing it. (How often do you find yourself hunching over the screen?) And that’s if they’re leaning close, which they would likely find awkward and unergonomic. Their natural sitting position will be at least an arm’s length from the screen!
In short, for the average Web user, reading most websites is not unlike taking an eye exam.
The harder your text is to read, the less of it will get read — and the less of what is read will be understood. 10 pixels is essentially pointless. 12 pixels is still much too small for most readers. Even 15 pixels will turn off visitors who might have otherwise converted.
Thus, we can conclude that if you want the maximum number of people to read, understand and act on your text, then you need to set it at a minimum size of 16 pixels.
“But Users Can Zoom”
If you code it right, people with vision issues can always use the Option and + keys to enlarge the text.
Thus spake one Web designer in a discussion I had on this very issue.
Not so. The users who will most need to adjust their settings usually don’t know how. And the users who do… well, they’ll probably just take the easier path by hitting the “Back” button. It goes without saying that we shouldn’t take our clients’ money and then design websites for them that will make their visitors uncomfortable. Our personal tastes are not more important than best practices in usability.
Web design is not about what designers like. It is about what users want and what will best achieve our client’s goals.
If the objective of a client’s website is to achieve some revenue goal, then our role as designers is to come up with something to achieve that goal as effectively as possible. Picking a font size that we know most users will struggle to read, a font size that will reduce readership, a font size that will ultimately cost conversions, is not good choice.
In print, type as small as 8 points is an ideal compromise between readability and cost, because you have to pay for every millimeter you use. On the Web, you pay nothing for using more space — provided that readers find your copy compelling, of course. What you do pay for is readers finding your compelling copy so hard to read that they click away instead of converting.
So, the question is, are you prepared to cost your clients money for the sake of what looks good to you?
16 Pixels Is Not Big
Our tastes and aesthetic preferences as designers are a lot more malleable than we think. What “looks good” to us is largely the result of what we’ve seen other designers doing and what we’ve come to expect.
Unfortunately, most of the websites we’ve seen are packed with tiny copy, because once upon a time screens were tiny, and so designers matched them with tiny text — but no one got out of the habit.
This article is set in TeX Gyre Schola at 19 pixels. I picked this size because 16 to 18 pixels looked too small to me: as I sat back comfortably in my chair, 28 inches from the screen, I found myself squinting to see it. If I’d used Georgia or Verdana, 16 pixels probably would have been fine: they were designed with a larger x-height and so display better on screen.
Now, check out the footer just below and see whether you don’t hunch forward automatically, screw up your eyes and furrow your brow. That’s 12 pixels for ya. And if you still disagree after that, tell me why in the comments.
Tweet

ptitgraig
October 7th, 2011 2:15 amYou make a good point. I’ll try that on the new website I’m working on.
Richard Williams
October 7th, 2011 6:22 amIt is a good point, but I still don’t think a large font is suited to long pieces of text.
When I got to the small text at the bottom I found it a relief to not have to move my eyes so far/quickly. I’m slower reading the large font.
Karin
October 7th, 2011 7:07 amI think you point at something interesting. The large font make us read it slower and the smaller to just go trough it and not really read it. To enlarge the font was a tip I got years ago when it comes to proofreading as it is to easy to just “skim” it when the text is small.
DJ
October 7th, 2011 9:13 amJust the opposite – the larger font IS more suited to long text. How old are you? (as the article asks) – I actually found the body text of the article easier to scan/read than the smaller text below. AND, his points are valid – I’m so tired of needing to zoom in order to read simple text on web sites (especially some web development sites); AND, the web is becoming so full with alternate sources of information that you CAN read – I’ve just given up on making the attempt. I have found myself completely bailing on 4 or 5 sites a week when I’m confronted with small text. Much more cost effective of my time to scroll down the Google list to a new site.
Julian Gaviria
October 7th, 2011 11:08 amAnd what size does Google use? Definitely not 16.
Greg
October 8th, 2011 5:51 amOutside of the initial search page, Google is hardly a paragon of usability. Sure their interfaces look cool and have some pretty cool features, but many of them are hard to use or break web ui conventions.
Ryan
October 8th, 2011 7:06 amWhy does everyone look to Google for pointers? Truth be told, their user experience people are effing useless. Whatever you do, don’t look to Google for UI/UX pointers – they just don’t have much talent in that department.
Luftmensch
October 9th, 2011 7:17 pm95 pixels.
Zendy
October 10th, 2011 10:23 pmYes, if you are building a search engine, you can copy Google. But this article is talking about full content website, not just search engine.
Kyle
October 7th, 2011 9:20 amThe first thing I did to read this was reduced the font size down to something around 12 point, then when I got to the comments I bumped the size back up sot that the comments are readable. I think 16 is a bit much, but a 14 or 15 might be an acceptable compromise. As pointed out in the article those that like a smaller font can probably more easily figure out how to reduce display size than those that prefer a larger font. I can’t help but wonder though that for those who override website settings to display fonts much larger than normal if when viewing a page like this they don’t get slapped in the face with 30 point fonts or something similar.
Rob O'Dwyer
October 7th, 2011 9:27 amI wondered the same thing myself about whether the text zoom settings would crank the font size up to unacceptable levels.
I just tested in Chrome, Firefox and Safari and it looks like the zoom is per-tab, or maybe per-domain.
So unless there’s another setting in some browsers that universally changes the zoom, it shouldn’t be a problem.
Tan
October 11th, 2011 10:01 am–moved to appropiate reply spot
Philip Tellis
October 7th, 2011 7:48 pm16px != 16pt. 16px == 12pt.
Bjørn
October 11th, 2011 5:05 amThat really depends on your platform and settings. In Windows by default and ALL browsers 16px==16pt.
Chris Howard
October 8th, 2011 4:02 amHey, Kyle. Problem is, this post is actually set in 19px. A lot of people missed that in the article – which proves they were skimming, and prob coz it was hard reading.
So, the font was too big. And that’s why it was hard work to read. At 16px it’s much more readable.
I don’t understand how or why on earth he’d try to argue for 16px then not even use it! Totally undermined the credibility of his arguments. (Especially when he says “See how small 12px looks now. Well of effing course it looks small against **19px**!!)
Eliseu Monar
October 10th, 2011 5:44 pmIt also depends on your screen resolution. Mine is set to Full HD, so the letters don’t look that big, they look pretty comfortable to read. He set the size to 19px because the font family was made with different height, so the author tried to match normal sizes like those from families like Verdana. Read this reply from José Luis, he talks about pixel density and screen size, that’s the point.
It would be nice to see a CSS measure that adjusts to screen resolution =D
Adrian
October 11th, 2011 9:22 pmI totally agree with you, at 16 px the above text is way better for easy reading. Meanwhile 19 px it is quite shocking and requires a careful text formatting, and in most cases clients don’t care about that. They just throw in the text and expect to look good.
Also as a web designer I often find that clients want a lot of information on the visible part of the homepage(the one you see without scrolling). For those cases you cannot use all text at minimum of 16px, because the design falls apart. you may use some text at at even higher number but the rest cannot be at 16 px min.
Maybe when everyone will have a screen resolution of a minimum 1440 px in width then it will be suitable, but as long as we still have to take into account the 1024 px screens I do not think it is suitable as much.
PS: you really ruined your article with your 19 px font size instead of 16. Maybe it is not to late to change it.
Will
October 13th, 2011 8:59 amYou obviously didn’t read the article very clearly. Repeated in many places, including the *opening line*, he says 16px should be the *MINIMUM*. Nowhere did he say it was necessarily bad to go higher. He didn’t undermine anything he said.
Ryan
October 8th, 2011 7:09 amI think your very much confusing familiarity with usability. Trying thinking and analyising instead of just throwing out reflex opinions.
Kyle
October 8th, 2011 8:53 amTo those that pointed out the confusion between point and pixel, you are correct, I had missed that, but it doesn’t change the point of my comment which is that the first thing I did when trying to read this article was bump the font size down about 6 units, in whatever system you want to measure those units in. To be fair, I like really small text, so I’m probably one of the outliers (I’ve got a nook and I always read books on it using one of the smallest font sizes it provides).
“I think your very much confusing familiarity with usability. Trying thinking and analyising instead of just throwing out reflex opinions.”
I can’t tell if you’re trolling or actually serious. Assuming you are serious, there are a couple points, first of which is that usability and familiarity are interrelated concepts. It’s practically a truism that what we are most familiar with we find the most usable. That’s the whole point of things like look and feel standards. Now, that doesn’t mean that you can’t ever change things and deviate from what people are familiar with without destroying usability, but such deviations should be gradual, consistent, and well reasoned about. This article makes some excellent points, and I do think in general we probably favor smaller size fonts than perhaps the majority of the population is comfortable with. I personally feel the article over-reacts and pushes the fonts a little too big (and as pointed out by others the article itself actually uses a font size larger than what the article suggests you use), but as I said above I like really small fonts so I probably constitute an outlier.
Jordan
October 8th, 2011 1:17 pmI did exactly the same thing… when I got onto the main article from the smashingmagazine homepage I read the first paragraph and dropped the size down.
I found reading body font at that size on a screen was more difficult than reading the usual sized text.
It is true that some people have difficulty reading small text because of eyesight or age but should the default change to cater for that? Some people have very limited vision and require screen readers so why not just have a blank page with a voice over?
It is correct that Web designers should be be designing for themselves but for what the users want and what is best at converting but as a designer who has worked on everything from major global cosmetics eCommerce websites to some of the fastest growing web hosting companies in the world, I would have been laughed out of the office if I had tried to use 16px as a body font.
I think line height and paragraph stylings (width, orientation, alignment) would have more of a impact on improving readability.
I also believe that the issue regarding people in their 40′s and 60′s struggling to read the screen is a bit of a weak point. If you struggle to read 12px font then you obviously require reading glasses or contact lenses and to read without corrective measures would only be damaging your eyes further!
D Bnonn Tennant, Information Highwayman
October 7th, 2011 12:34 pmHey Richard, a little factoid I didn’t include in the article is that studies show small text encourages scanning, while larger text encourages close reading.
So large text is absolutely appropriate for text that you want people to read, rather than skim—such as this article :)
D Bnonn Tennant, Information Highwayman
October 9th, 2011 2:31 pmEr, actually I got this the wrong way around. Small text promotes close reading. Large text promotes scanning.
Rizqi Djamaluddin
October 8th, 2011 1:53 amThe drawback I find with large text is that it increases scrolling. Less eye strain figuring out letters, yes, but as mentioned, it means the eyes move around faster and fingers have to scroll more, and it’s even more straining for the eye to have to follow around text as it scrolls around. I found this to be especially true for mobile devices – you have to hit that sweet spot between readable text, and not too much scrolling to get what you want.
whatsat
October 8th, 2011 11:06 amThe length of text should not matter. It is Irrelevant.
The ability to READ it , is the issue.
Agree 100% with Bnonn , 16 is as small as you should go.
There is NO justifiable reason to disagree.
Font’s are an issue as pointed out in the article ,
HOWEVER the times font used in the article can hardly be described as easy to read. ( practice what you preach ? Bnonn )
Every letter in that font has lots of useless accents that blur the true shape of intention.
eg..( have a really close look at the letter i )
**** I know Times roman is a tradition in print.
( therefore it must be good ? ) no
Think about the ease of text transmission.
For the most efficient transport from page to eyes. The font needs to be the most basic font available. Not cluttered with artistic shapes.
Anyway , great article to make those webdevs who think form over function is priority.
When the reality is…..
COMMUNICATING content , is the ONLY reason for text to exist.
Danny
October 7th, 2011 2:17 amShouldn’t the “12-point text printed on paper” be the reference and stay the same size in the 2 examples? (also the second one says 12-inch text)
Other than that, definitely something to ponder…
Mario
October 7th, 2011 3:31 amHaha… I noticed exact the same!
Devan
October 7th, 2011 4:52 amMaybe a bit of an oversight but it doesn’t actually weaken the claim being made, right? For the purposes of the argument, it doesn’t matter if the print example is the same size across the two examples—only that in each separate example, they’re shown to scale with the on-screen type.
LXicon
October 7th, 2011 7:11 amwhen i saw those two example screen shots which didn’t match each other (and were both WAY smaller than the text of the article) i felt like the author didn’t know what he was talking about. i stopped reading after that.
Jean
October 7th, 2011 8:28 amAre you retarded? It’s the scale that matters, you could zoom out until the printed text is equal and the screen text is smaller, exactly the same.
John
October 7th, 2011 2:24 amInteresting points, it was definitely very easy to read this article.
Scott
October 7th, 2011 2:25 amI’m not sure if it’s on purpose or not, maybe I’m missing something…
BUT
I find it hard to take the article seroiusly (even though I agree with the point being made) due to the styling used throughout;
The Red Impact headers and the drop shadow & spacing on the first letter of the new paragraph. It’s horrible.
I was also under the impression that Sans-serif font’s are easier to read on screen than serif fonts, so on an article about legibility, why choose a serif font (no matter the size?).
Maybe it’s ironic? Maybe I’m dumb? Maybe I just missed an inside joke?
Sequoia
October 7th, 2011 4:28 am>I find it hard to take the article seroiusly (even though I agree with the point being made) due to the styling used throughout;
Are you joking? He presents usability studies and examples, and you dismiss it because you don’t like the fonts he chose? *sigh* You’re not doing much to combat the combat the stereotype of the myopic, superficial designer. This is exactly the attitude an article like this is meant to combat; there’s more to usability than your personal opinion!
Shelby
October 7th, 2011 7:14 amI was just about to say the same thing!
Scott
October 7th, 2011 7:24 amIt’s not that I don’t like the chosen fonts, it’s that I think they adversely affect the legibility of the article.
I’m using Firefox 7.1 (I’ve noticed in Chrome the font’s pulled in by the browser are different)
Take for example the “But Users can Zoom” section.
The styled “T” at the start of “Thus spake one Web desiginer…” the large red lettter and large grey drop shadow touches the “h” from thus. I find that difficult to read.
The article is about making text easier to read… The next paragraph starting ‘Not so.” Is unstyled. I find it much easier to read.
That was only my point. Not that it’s ugly – just it’s harder to read in that style.
There are further comments below stating my thoughts on serif vs sans-serif for screen use are incorrect. I accept this and as mentioned would appreciate any links to research on these to read up on.
foljs
October 7th, 2011 4:31 amThe Red Impact headers and the drop shadow & spacing on the first letter of the new paragraph. It’s horrible.
Yes, let’s confuse aesthetic matters with readability, which is what the post is about.
I was also under the impression that Sans-serif font’s are easier to read on screen than serif fonts, so on an article about legibility, why choose a serif font (no matter the size?).
You were under the WRONG impression based on old-wives tales. Go read some current studies with actual stats from users. Oh, and dpi’s are higher now, and font rendering smoother than when that old wives tale started.
Scott
October 7th, 2011 4:45 amMy second to last point stands then :-)
Any links to the current studies are welcomed, looks like I need to do some reading.
sean
October 7th, 2011 6:29 amHey Scott, I know what you mean about the headers. The drop shadow is encroaching into the text and that does affect readability. I also understand where the others are coming from. I’m just writing to say I thought you handled all this well and the fact that you wrote back and requested documents to read further shows you love what you do and you have respect for the site in general.
Cheers,
Scott
October 7th, 2011 7:00 amYeh I was more looking to add my thoughts on it all, not start any conflicts.
Quite happy and understanding of the feedback / comments I’ve received.
Re-reading my response, the point “second to last” I was referring to that I was correct was ‘I’m dumb…’ (semi-serious), not any of the criticism of the article.
I’ve also now noticed that some of the encroachment of the drop shadow I’m seeing (in firefox 7.1) isn’t present in Chrome. My original comments where aimed at the Firefox version I was seeing.
Sean
October 7th, 2011 9:43 amDon’t you mean ppi?
AJ
October 12th, 2011 9:47 amWas about to point this out myself. DPI does not exist on screen.
William Key
October 7th, 2011 4:32 amThe latest research indicates that people no longer seem to have any issue with reading serif vs. sans-serif fonts onscreen vs. in print. One factor is that monitor screen resolutions are generally high now so that we don’t encounter the same problems with the terribly pixelated serif strokes that occurred in older monitor displays. Another possible factor is that we’ve become so used to reading sans-serif onscreen that we don’t need the serif fonts to improve legibility in print. Obviously there may be generational differences between people who consumed printed text for most of their lives vs. those for whom digital text has been equally or more significant than reading print.
Eric Barker
October 7th, 2011 5:07 amI’d like to see the studies on sans vs serif people are referencing, but not referencing. Print or post it on the highest resolution available and apply all the stats from this post and you will struggle to read the serif more than the sans. You don’t need to spend any $ researching this. Squint your eyes people, serif’s are harder to read.
Shaun Kelly
October 7th, 2011 5:47 amGood piece here goes through a ton of peer-reviewed journal articles on this:
http://alexpoole.info/which-are-more-legible-serif-or-sans-serif-typefaces
Jacques
October 7th, 2011 6:00 amWell yeah, if you’re squinting, of course it’s going to be harder to read.
Nick
October 7th, 2011 9:27 pmlol, you didn’t get the point. If you squint and can read sans-serif but not serif, then you’ll know which one would be easier to read in general. I wont go into details, i don’t care that much.
CA
October 10th, 2011 7:15 amLegibility and readability are not the same thing.
D Bnonn Tennant, Information Highwayman
October 7th, 2011 12:43 pmFor the record, I didn’t choose Impact; I chose Limelight, with the article subhead set in Ostrich Sans. And the drop shadows looked sweet with those :P
The typeface changes were entirely the decision of Smashing Magazine.
D Bnonn Tennant, Information Highwayman
October 9th, 2011 2:36 pmHey Scott, I’m not sure what’s up with the styling in Firefox. I actually created the CSS file in Firefox, but since submitting it for publication, summat nasty has happened to it.
Check the article in Chrome for how I originally wanted it to look.
Re sans serif fonts being easier to read, there was a time when this was true on screen, because of low pixel density making the serifs really jagged. However, these days it’s not a problem at all, provided you’re using a good font size to begin with :)
See http://alexpoole.info/which-are-more-legible-serif-or-sans-serif-typefaces
bora yalçın
October 7th, 2011 2:27 amI’m actually a web developer not designer but as a “i will try and learn design” project i designed my personal site ( http://borayalcin.me ). At first it was 12px as I’m used to see but even though I’m 25 its hard and tiring for me to read my own stuff.So, I’ve increased it to 16px and everything is fine
now after reading the article now can “scientifically” defend 16px+
Joe
October 7th, 2011 2:27 amWith the advent of Typekit & Google Web Fonts, it’s harder and harder to have a go-to body font size. What works for one font at 16px might look tiny for another.
I suppose that’s where Less comes in handy.
RustyH
October 7th, 2011 2:30 amMy tests indicated younger viewers (< ~30) are less likely to read larger web copy. My guess to the reason was everything blends together and more scrolling is required.
Personally, I've always found large text to look like an old school ebook sales page.
Ilia
October 7th, 2011 5:50 amThat seems more like a styling issue. With proper line spacing, paragraphs and well thought out layout the text will flow naturally and comfortably instead of looking like a wall of text.
Jan
October 7th, 2011 6:01 amThe answer then, of course, is for web sites to respect the text size users configure in their browsers. Use relative sizes, not absolute ones.
Carson Weber
October 7th, 2011 2:31 amI’m having difficulty reading these comments, which are appearing at 14.9667px. :-)
Viktor
October 7th, 2011 2:34 amso if this is true, why smashingmag mainly only uses .75em?
Devan
October 7th, 2011 5:00 amSpecifying font size in .em calculates it for the given element relative to the base font size specified elsewhere. The body copy of this article renders in my browser (with default font sizes) at 19px, which is, of course, well in line with the article’s claims.
Patrick Hazard
October 7th, 2011 2:34 amReally interesting, I think I also subconsciously skim articles more if they are in a small font size and read larger font articles more fully. It’s easier on the eyes even for a twenty-something year old with perfect vision.
You also missed the point that it’s much easier to read on a phone or tablet but they that obviously brings up the whole ‘responsive website’ can of worms!
Web Strategy
October 7th, 2011 8:25 amYa, the readability on mobile was my thought also … that’s where I really see the attrition even more so than on desktop as more and more people are connecting via smartphones.
Chris
October 7th, 2011 2:35 amAs you’ve mention it yourself, “This article is set in TeX Gyre Schola at 19 pixels. I picked this size because 16 to 18 pixels looked too small to me …”, gave a lil’ bit contradiction to your article
I believe textface size is “hardcoded” to their initial design DNA.
Since typical “websafe” fonts are most likely sized well in 16px on screen (hence 1em in browser’s standard size) then using 16px is pretty much logical.
But setting an 16px to EVERY FONTS for textface as a must is not very ideal as well..
I believe typographical knowledge still reign in this area of expertise.. including textsizes, kerns, spacings, etc..
unless for a generic font-style: serif/sans-serif :)
Great article though! I strongly agreed that most of nowadays textfaces are treated too small to be comfortable.. :D
Chris Armstrong
October 7th, 2011 2:37 amGreat article, however I’d go one step further and suggest that anything less than 1em is a mistake :)
The browser defaults have been set for readability across all devices, so keeping 1em for body copy (and basing your layout in ems) helps with great readability on any device.
Sean
October 7th, 2011 9:45 amGood thing 1em is the same as 16px, since that’s the browser standard.
a
October 7th, 2011 2:40 amCouldn’t agree more, this is the first page ive visited, where I have not had to zoom in (normally my first action on any page).
BrianRowdy
October 7th, 2011 2:40 amYes! To every word, right up to when you dropped the word ‘weaksauce’ and then No.
tiujam
October 7th, 2011 2:42 amDo you still remember when 4 years ago it was a passing trend to use 10-11px Verdana as body copy? Yach! I know I personally get very easily tired on reading blogs that use such a small font-size (*ahem* A List Apart *ahem*) – no matter how _awesome_ the content is! My personal preference is to use a RSS reader with my own font preferences, since zooming pages annoy me even more:P
For me the small body copy problem got slightly worse when I got my first laptop. After that I started to implement bigger and bigger body fonts to my designs. Haven’t gotten to 16px yet tho’! It definitely takes something to get used to AND it is definitely not for every situation:)
Johan
October 7th, 2011 2:44 amWhile I can agree that you should define your website (and thus your font-size) by the users you aim to attract, the font you are using is hurting my eyes more than if the size was 12px.
This makes me focus more on the strain it was reading than the content you were sharing. Food for thought perhaps?
Anyway, good points but common sense. You should always design for the people who will be reading your site. What I walk away from is the stats about the differences in age and readability. Thank you.
Mike Dougan
October 7th, 2011 2:50 amYour Typography here stinks, but I agree with you 100% about 16px font’s on screen :-)
WC
October 7th, 2011 2:56 amI was reading this in Google Reader, which has about half the font size your site does. It’s quite a striking difference and convinced me to try zooming Reader in a bit. Definitely better. Not sure why I never did it before… I zoom other websites all the time.
Jacques
October 7th, 2011 3:05 am16px fonts definitely tend to be a good choice, but it really does depend on exactly what font you’re using, particularly if you’re using something like typekit where the fonts tend to have a small x-height.
If you make those too large and there’s a problem with the service, suddenly your back-up font is looking comically large.
Neil
October 7th, 2011 3:06 amI find this article hard to take seriously.
Font size should be determined by the nature of the site and the target audience.
To say 16 pixels should be the minimum for all sites is a blinkered view.
D Bnonn Tennant, Information Highwayman
October 9th, 2011 2:40 pmNeil, I agree that font size should be determined by the nature of the site and the target audience. Perhaps you could give an example of a site or target audience that you think indicates <16px for body copy?
Espen
October 7th, 2011 3:06 amSome very valid points here, but there’s a lot more to readability than just font size:
1. People need to breathe so give them the right amount of air between lines and words. The formatting in this article needs more on both, if you ask me.
2. People don’t read letters and words, they read shapes. This is why road signs, which is probably the most important texts you’ll read in your entire life, in most countries are capitalized and have been for decades. Once this shape is significantly bigger or smaller than they are used to, their ability to recognize it as a word they know is reduced. This is very individual but personally I’m feeling the font and size of it in this article is borderline of what I’m able to “read without reading”, if you know what I mean. Probably because I’m so used to reading 12-ish px sans-serif fonts on a screen.
Paul Trumble
October 7th, 2011 3:11 amNice non sequitor Scott.
As I read this article, I leaned back from the monitor and relaxed and enjoyed reading it. I’m a convert.
Chris Howard
October 8th, 2011 4:06 amThat doesn’t make sense. How can you be converted by the comfort of the font size when it was actually 19px? So, to achieve what he has, you’d actually have to set all sites you make to 19px. But then that’s a whole new debate.
Check yourself
October 11th, 2011 7:13 amYou need a lesson in typography before commenting. 16px in one font does not equal 16px in another. It’s about x height.
Tesmond
October 7th, 2011 3:12 amHi,
I strongly disagree. You simply cannot generalise in such a manner. You should setup personas and test cases based on your target and current markets, if your set of personas does not (and will intentionally) contain anyone over the age of 28 why should your design consider that user secondary?
Luckily I read this article in an RSS feed reader where I can choose my preferred font. after reading that you had intentionally styled your article in TeX Gyre Schola at 19 pixels I thought I would look at the article in its original format.
The serif font strong colours and large font gives a childish impression and so it is much harder to take the article seriously.
Additionally perhaps your lower contrast colour (#444) and hard to read serif font required you to select greater than 16px. Try #000 and a sans-serif font and I suspect that 14-15px will appear a lot more serious and having done so in both a feed reader and this webpage I can assure you it is definitely more readable.
There are far more complex factors at work than just size and lack of structure in body copy is far more important than larger font sizes in improved readability. You are correct in identifying that font sizes are often too small for specific user segments, however, generalising to the extent that you do is a step too far. Fonts and users are veried and the appropriate design decision is a complex mix of this.
PS your main heading is highly unreadable, if you changed it from its current:
48px font size, 40px line height and centre alignment
to
14px sans-serif, 18px line height and left alignment
it would be much easier to read for the majority of users…having overlapping line heights and center alignment is very poor for readability…
D Bnonn Tennant, Information Highwayman
October 9th, 2011 2:59 pmHowdy Tesmond.
Well, I do. Could you give me an example of a persona or test case where 12px is more readable than 16px?
Of course, I explicitly addressed the distinction between aesthetics and readability in this article. I also find it bizarre that you evaluate my contention not based on the facts and arguments, but on how I choose to present it. Isn’t that rather like judging a public debate based on which speaker’s voice you found sexier?
Alex Poole reviews the evidence for and against sans-serif versus serif, and finds no difference. Are you saying I should simply accept your assurance based on your personal reaction? A sample size of one? What about all the research I’ve cited that contradicts you? Should I ignore it?
Now, perhaps #000 is easier to read than #444. I don’t have any definitive research on that. The general consensus which I’ve seen says that #444 is easier on the eye, since the extra contrast of #000 on a screen (which is an active, additive color mode, remember, so glare is an issue) results in fatigue.
Of course, you can debate my choice of font. I’m sure you could find a better one than TeX Gyre Schola. That’s just the font I happened to feel was right in this case, because I wanted to mimic a more classic look, rather than a modern one.
Well certainly. You won’t get any argument from me about that—for example, setting copy in multiple columns is a pet peeve of mine, and an example of awful structure (physically speaking). And in terms of logical structure, well, that’s something I teach very vocally as well (see here, here, and even here.)
I agree re overlapping line-heights—but please check the article in Chrome to see how it should look. I have no idea what happened to it in Firefox; you’d have to ask the Smashing Mag editors. It worked when I submitted it :)
As regards center alignment, again, I’m afraid you’re wrong. For a top-level heading, center alignment is very appropriate, for two reasons: firstly, it acts as a visual cue that “this is a headline”. That is how we’re accustomed to seeing headlines formatted in newspapers and advertisements. Secondly, provided your headline is in about the right place vertically, it means it will be directly in your user’s eyepath for the initial fixation when they arrive on the page. See http://blog.kissmetrics.com/eye-tracking/
Tesmond
October 26th, 2011 2:41 am“Could you give me an example of a persona or test case where 12px is more readable than 16px?”
At no point did I suggest that 12px was a good size. However, I have seen relatively small font sizes used for label headers where the value they were describing was significantly larger and the supporting label was not normally necessary to read, more a clarifier for the first time you view and item.
However, your examples of 12pt text are perhaps important. Paperback books are normally 10-11pt which as you have cited is less than 16px.
“Alex Poole reviews the evidence for and against sans-serif versus serif, and finds no difference. Are you saying I should simply accept your assurance based on your personal reaction?”
Mis-quoting me in a discussion is poor form. I said “hard to read serif font” NOT a really well designed serif font verses any old sans-serif font.
“distinction between aesthetics and readability”
Believing that these are two very distinct elements is clearly wrong. You emotional engagement effects how will you can read an article and the emotional engagement is highly influenced by the general asthetics. They are inextricably linked elements.
If Barrack Obama delivered a speak naked doing an Irish jig with his lungs full of helium do you think your engagement with the content would not be effected?
“For a top-level heading, center alignment is very appropriate…”
A center alignment is less readable when is spans multiple lines than a left aligned heading. Your main heading and sub headings both span multiple lines, reducing their readability. The added overlapping which you have advised is an editoral blunder makes the alignment even harder to understand.
“http://blog.kissmetrics.com/eye-tracking/”
Says nothing about center aligned text. The negative aspect is the extra conscious effort to find the next line start.
Steven
October 7th, 2011 3:15 amBefore writing my response, I thought I would have a look around on the web for sites with large text size.
The one I found to be the easiest to read was apple.com, I was surprised, cos at first I didn’t agree with your article. It has now made me rethink of how I view webdesign. So thank you.
Leon
October 7th, 2011 3:15 amYou speak truth. Interesting to see how people still try and argue that big text is “ugly”.
I’d also argue that
@font-facewill make minimum size more difficult. For example, try using Minion Pro at 16px.I guess ems are another discussion…
Avinash Birammbole
October 7th, 2011 3:18 amNice article but being a UX Designer and after working in Print and online media, Ive few of my views based on my experience.
Choosing a font size is still lot more depends on various criteria s for example target audience (very important) is it slates/tabs/pads or monitors or mac displays. every screens has different properties and the display calibration levels.
Your point on viewing distance of 28ins if fine but its not the same case while using ipad or hand-held devices, and your comparison with printed book is not a valid comparison as both are different mediums. One is reflective and other non-reflective.
Fonts on different devices are rendered differently hence on reflective devices its suggested to use “em” and not “pts” or “px”, as they are scalable eily without distorting, sans serif font having 12 to 14 is very much readable on any screen.
Using “em” : Usually browse default is 16px, to make it easy, change your default to 10px that way while using “em” you can set 1.4em for 14px font.
D Bnonn Tennant, Information Highwayman
October 7th, 2011 12:55 pmGood points, and ones I’d liked to have gone into in the article…but space was a factor. Thanks Avinash.
Niels Matthijs
October 7th, 2011 3:23 amVery (very) tiring article to read. Short reading lines + big font sizes don’t allow people to sit down and read the text comfortably. You know what doesn’t happen in books … scrolling and readjusting to your eyes to find the reading line again. Which I need to do like every 20 seconds when reading this article.
Doesn’t help that the site is completely in shambles on FF7 (really??)
Ich
October 7th, 2011 3:27 amGreat article. I’ll have another look at my websites now….
Graham
October 7th, 2011 3:33 amIf you have to activate ‘Readability’ on a blog, its design has failed. I find I have to use Readability on most sites because the text is too small, so I definitely sympathise.
However, I have to agree with other comments with regard to Impact in this article. Why? What’s wrong with a slightly larger, darker Georgia for headings? Do drop caps help us read body copy? I found myself reaching for the Readability button.
andy
October 7th, 2011 3:34 amWhy are the comments on this site not 16px?
Patrick
October 7th, 2011 3:44 amOf course, I had to zoom out to read this comfortably, so…
Stijn
October 7th, 2011 3:50 amFun fact: exactly last night I changed the font-size of my body copy from 16px Georgia to 15px Helvetica. In my vue the letters are still about the same size though, as 16px Helvetica appeared a bit too bold.
Richard
October 7th, 2011 3:52 amI found this article very interesting! So interesting that I tried following your link in the footer but it’s pointing in the wrong direction :(
Jonas
October 7th, 2011 3:53 amCouldn’t agree more.
On another note, am I the only one getting some bugs with the new site design? In FF, I get Arial instead of Ostrich Sans Rounded Medium for the h2 headings (works just fine in Chrome and IE9), and even the h1 heading also looks wierd in FF.. and the header navigation is broken for me in IE9 (breaks into two lines, 1366*760px display).
Like the new look overall though.
Erwin Heiser
October 7th, 2011 4:13 amHear hear! Been saying this for years myself, on mobile it’s even worse: ever tried reading 11px arial on an iPhone? Good luck enjoying that :)
Jason
October 7th, 2011 4:14 amI’ll change my website when you change yours…
Cameron Chapman
October 7th, 2011 4:15 amPersonally, I think line height and contrast are equally important to type size when it comes to readability. My personal favorite is to start with a text height of 14px and a line height of 24px. Depending on the particular font, I might go up or down from there. But having a line height that’s roughly 1.7x the font size does way more for readability than font size alone.
James Brocklehurst
October 7th, 2011 4:57 amCouldn’t agree more.
The way this article has been styled is a prime example – bump up the line height to 32px + and it becomes much easier to read!
Chris Raymond
October 7th, 2011 6:40 amCameron: You are right on in focusing on line-height. It is too bad that Eric Meyer’s reset, which so many use slavishly, sets line-height to 1, which is so wrong, wrong, wrong for legibility and the basics of good typesetting.
Hector Hurtado
October 10th, 2011 10:33 amThat’s why it’s a reset, not the holy grail of style guides… You still have to work after that.
Jaina
October 7th, 2011 7:25 amWhole heartedly agree with the line height being just as important as font size. I find that even with a relatively small font size, if you give the lines space to breathe, it’s so much easier to do the reading.
James
October 7th, 2011 9:13 amI think line height is more important. Condensed text is extremely hard to read. Many sites fail to see this. It doesn’t matter what size your font is if your line-height is too tight it will instantly become harder to read. I think even this articles font style failed there. The line-height is far too tight.
equus
October 7th, 2011 4:15 amReadability of < 16px on what resolution screen?
Ave
October 7th, 2011 4:20 amGood point! well done .. thanks :) 16px it is! ;)
Pete
October 7th, 2011 4:29 amI agree with Tesmond. Of course there are arguments for and against but it all depends, there’s absolutely NO one magic bullet, there are a myriad of factors to take into consideration. e.g. screen resolutions.
Martin
October 7th, 2011 4:30 amThe point, and a valid one at that, is that you should split test and adjust your font to which size converts the best. Be it 10px, 16 or even 24 px.
Interesting read if you get the overall concept.
Michael Schmidt
October 7th, 2011 4:33 am+1 for larger fonts. We have already implemented it in our latest sites such as http://www.keyboard-and-mouse-sharing.com
It looked strange at first but it is clearly easier to read.
Michael
Andrew
October 7th, 2011 4:36 amI can’t believe how many of praising one-line comments the author of the article has posted himself. That cracks me up.
Gabriel Gosselin
October 7th, 2011 4:36 amFirst, I agree that body copy on the web tends to be too small. I understand comments about scrolling, but I disagree with longer line-lengths. The book convention of 60 characters to a line to holds true on the web in terms of easy scanning.
Second, comments about setting type in ems are slightly misinformed. em is a relative unit of measure. If you don’t use a reset, and browser has a default of 16px, them 1em is 16px. However, if you set style to 12px, the em follows.
Another factor others have touched upon is line-height. Heights of 1.2–1.6em tend to yield more readability (defaults are often quite tightly set).
Sandstream
October 7th, 2011 4:40 amGreat article and the first one I have read on SM leand back in my office chair! :-)
I use big fonts an all my sites and will try 16 next time!
Steve Fenton
October 7th, 2011 4:45 amThis article was great because it made me think about this issue. I might not be opting for 16px, but I don’t think we are really discussing a “rule” here, more a “guide”. I believe you can justify different font sizes based on your website and audience, but if you haven’t considered the font size in this way you have missed an important step in the design.
It has certainly made me think about opting for larger font sizes than I am used to using, which I will experiment with – so thank you for the article.
Tomo
October 7th, 2011 4:47 am+1px/year is my rule of thumb.
We need to keep increasing the font size, true. I did enjoy reading the article, and it was a better feeling reading text than it usualy is.
However, I would settle at 13-15px and provide a visualy clear way for users to increase it. Otherwise, clientwork might get realy complicated. :)
Kilian
October 7th, 2011 4:52 amI totally agree. But don’t take my word listen to these guys:
http://www.informationarchitects.jp/en/100e2r/
http://www.wilsonminer.com/posts/2008/oct/20/relative-readability/
And while we’re at it:
http://www.informationarchitects.jp/en/the-web-is-all-about-typography-period/
Matt
October 7th, 2011 4:53 am“If you don’t explain what people should do, or why they should do it, then they certainly won’t. And the only way to tell them is with text.”
For a medium that’s as interactive and visual as the internet, I found this to be really disheartening. There are many ways of conveying message and what people should do in ways other than just plain text, and usually in combinations of text, video, imagery, interaction, etc. To say it’s only done through text is shortchanging some of the best assets of the medium we choose to work with.
Alex
October 7th, 2011 4:53 amI much prefer small text to big, and as a result often find myself skipping headers (not great when you are proofreading). I find 9-10 points optimum for reading. I am very short sighted so it may be that my glasses are having an effect.
Large text is horrible for scanning, I imagine because the text takes up more space and your eyes cannot take it in all at once.
If large text is the way things are going to go, roll on 40 – only 11 years to go!
Matt Velic
October 7th, 2011 4:54 amVery nice article! I’ve switched to using an 18px standard lately. 16px if the font has a larger x-height. I’ve worn glasses for 10 years now, and I just don’t know why anyone would want to strain their eyes playing around with some tiny font.
Viktors Rotanovs
October 7th, 2011 5:08 amMacBook Airs have 135dpi and 128dpi displays. Using smaller fonts creates very real problems for MBA owners. Zooming isn’t the best solution: pinch-to-zoom doesn’t reflow the page, and Ctrl-+ breaks many interface elements (buttons, etc).
bikeboy389
October 7th, 2011 5:09 amI agree that we should all be paying greater attention to usability/readability. And I agree that 12 pixels on a high pixel density monitor is getting to look a little small.
What I do not agree with is your facile linking of small text with loss of revenue. Lots of definitive stats presented on readability and visual acuity, zero on how readability hits conversion.
If you tell me I’m losing money with small text, I expect to hear something more than vague insinuations that people *must* be turning away in droves from my otherwise awesome site because it’s in 14px body text.
Gain points for advocating readability/accessibility. Lose points for failing to prove your central claim, which is that small fonts are a “costly mistake.”
Finally: I agree that your article was actually a little hard to read. Column too narrow for your type size.
David Philippe
October 7th, 2011 5:09 amVery interesting article indeed and an enjoyable read but I don’t think a 1em font size is meant to work in every case.
Bigger font size is often associated with book for small children and/or catering for the age/disability and depending on the audience that you are trying to reach this could actually be off-putting and costing you sales.
Viktors Rotanovs
October 7th, 2011 5:15 amI’m 34 and I already find some fonts hard to read.
Rafael Trabasso
October 7th, 2011 5:10 ami totally agree with the author.
i just had a client who asked to design his site with a 11px basis. i hated it, won’t even show in my portfolio.
i’m sticking with the 100e2r (100% easy to read) stuff and it’s font sizes are sort of huge and i love them.
and i’m surprised how many people here took their time just to bitch about petty stuff .
Fred LeBlanc
October 7th, 2011 5:12 amSomething that the http://largefontcoalition.org/ has been rallying for for years! (Actually, it’s our only platform.)
iCosmin
October 7th, 2011 5:14 amI was reading this in Google Reader that has font-size set to 90% (about 13px).
When I reached the end and saw you saying “this is written in 19px”, I just had to click through the website and see for myself. Indeed, much better.
I always tend to constantly skip & search for what interests me in the body copy and this always results in eye strain. I want to read fast and font size is keeping me back.
It was a pleasure writing this comment in 22px, thanks for the eye-opener (pun intended)
mc
October 7th, 2011 5:16 amI for one hate small body text and I am 19. So much so, in fact, that I used 20 px body copy for my last design. Maybe a little excessive but I felt it worked in the context I used it. Another point to be made is increased resolution; the fact that many smaller monitors are rocking 1900 x 1200 resolutions that make everything appears significantly smaller.
Tibi Neagu
October 7th, 2011 5:16 amThe trick you used at the end is pretty clever, but completely irrelevant. After reading an 18px font for 3-5 minutes, of course 12px will seem tiny and hard to read. But using 16px fonts as a general rule is simply not feasible for every scenario – screen real estate is limited, and when you need to get your user’s attention with the content ‘above the fold’, you can’t afford to use 16px fonts. Even Amazon (with 100s of millions of users) use 12px, and 16px only for headings.
Perhaps in a blog, with lots of body text, larger text might make the users more relaxed while reading the article – in that case, I would agree with you.
D Bnonn Tennant, Information Highwayman
October 7th, 2011 1:05 pmTibi, I’ve found very little evidence that the fold has a significant affect on conversion rates. You should definitely place more valuable content above the fold—but provided your headline and lede are there (and they interest users sufficiently), then users will scroll very happily.
The myth of the fold is one of the most pervasive and long-lasting boogeyman stories on the internet.
Liz
October 7th, 2011 5:18 amHm. I actually found this article very difficult to read and had to zoom out to read comfortably. These font sizes are obnoxiously big to me, and I don’t think it’s just what I’m used to — the footer text was a welcome relief to my eyes. Maybe it’s because I’m young (mid-twenties) or maybe I’m just a special case.
I think choice of type size is largely dependent on audience. Next time I design a site aimed at middle-aged customers, I will remember this article. If I’m designing for people in my own age group, however, I’ll stick to 13 – 14 pt.
tl
October 7th, 2011 5:19 amI’d like to mention. Besides font-size, line-length is important as well – if a line is to long readers will have a hard time to focus if is to short the eye hat to travel to much. A good length is something between 50 and 70 characters per line (about 12 words).
D Bnonn Tennant, Information Highwayman
October 9th, 2011 3:09 pmIndeed. I’d like to emphasize this, since a few people have said the narrow measure of this article made it hard for them to read. I set the measure at around 60 characters per line (depending on other factors). To the best of my knowledge, that is a nearly ideal measure for readability, with decades of research to corroborate it.
adam
October 7th, 2011 5:19 amare the comments 16pt as well?
Markus
October 7th, 2011 5:21 amPlz read Olivers article again before quoting stuff he didnt wrote and didnt mean.
Next thing is, you cannot say 11-15px costs your client money. It may be true in some cases, and others it dont. It just depends and its the designers job is to know that. You obviously dont know it and repeat others ppl opinions you didnt unterstand.
D Bnonn Tennant, Information Highwayman
October 9th, 2011 3:11 pmHi Markus, please could you point out what I quoted that Oliver did not in fact say?
I am always careful to check my sources, so if you believe I have incorrectly attributed a comment to him, then I want to correct it.
Phil Middlemass
October 7th, 2011 5:28 amWith the variation in screen resolutions these day I find it hard to that this kind of research on board.
Lets look at it like this:
I have a laptop at home, 15.6″ display running at 1920×1080 = a pixel density of 141ppi
My wife has a laptop, again 15.6″ display but at 1366×768 = pixel density of 100ppi
So assuming the default dpi is set in the OS on both machines, my wife’s laptop will display browser text 40% larger than on my own laptop simply down to the resolution.
Resolution = massive factor in readability for text on screen
These kinds of articles can only be taken as a guidance, there are too many factors to consider to come to a single conclusion that can apply to all
16px on my laptop looks OK, bigger than I’m used to, but on my wife’s machine is actually difficult to read, it is all subjective, everyone has a different comfortable reading size because of their respective screen sizes and resolution.
Nate
October 9th, 2011 4:08 amAgree.
Let’s just the content, specified target audience, design direction & assets (fonts & necessary elements) “be the pilot”. And tell the machine to do the spoiling services.
Also, after the “use black-on-white text” so called-movement, fixed-vs-responsive web grids ideas, eye candy vs minimum image usage suggestions, and-the bla-bla-bla-tra-la-la of:
“let’s make all future websites looked just like plain Microsoft Words documents, so we can all go to sleep already, please?” articles.. and now, this??! 16px for body text?? Oh man.
Truly.. very elderly thoughts :-)
I have a simpler idea, if (elders’) readability is the case… let’s make a public vote, so all browsers make their built-in zoom (function) button BIGGER somewhere on their next update… case closed!
(FYI: I’m 31, and I’m still enjoying 10-12px body text IF (and only IF) a site’s content is WORTH to read & visually well designed)
Pattie
October 7th, 2011 5:37 amI agree (with a caveat). I believe designers have missed the mark on designing for readability / usability. It is hard for them to break out of the “small is tasteful” mindset. Now for the caveat – at my resolution 16 px look too big . . . I would set the minimum at 14px.
Manuel Ruiz
October 7th, 2011 5:50 amI totally agree with you. I never make my body copy smaller than 16 pixels. I find it to be the perfect size for an easy to read copy.
Kim
October 7th, 2011 5:59 amI understand the point, but this article is absolutely unreadable in the large font…and I’m over 40. The bright red makes it more painful, and if the point was readability, it’s lost in this presentation. I’m all for usability and accessibility, but this was a terrible way to present the argument.
Chris Howard
October 8th, 2011 4:10 amthat’s coz he did it at 19px, not 16px!
Stephen Eilert
October 7th, 2011 6:06 amThis text (and most of the comments) do not make any sense to me at all. I thought we were way past the age where we had to hardcode elements in *pixels*!
Monitors have different physical dimensions and screen resolutions, so all this talk of “pixels” is irrelevant.
D Bnonn Tennant, Information Highwayman
October 9th, 2011 3:17 pmStephen, if only operating systems correctly identified monitor ppi and adjusted text display appropriately, you’d be right. Then we could set body copy at 12pt and rest assured that it would display the same size for people on iPhones and people on 30″ displays.
Unfortunately we don’t live in that world, so we have to make do with rough-and-dirty guidelines instead.
baz willrun
October 7th, 2011 6:10 am“At age 40, only half the light gets through to the retina as it did at age 20. For 60-year-olds, it’s just 20%.”
So you saying that a 40 year old see things 50% darker
and you reckon to a 60 year old things appear 80% darker
you have medical proof of these “facts”
Hmmmm….should I laugh now or later
john charlton
October 7th, 2011 6:53 amLaugh now. As you get older, you won’t find it so funny.
This article should be taken in the spirit it is offered which is simply asking you to consider how small is too small.
Fact: Our vision deteriorates over the course of our lives and these changes need to be consider by designers, not brushed aside.
D Bnonn Tennant, Information Highwayman
October 9th, 2011 3:19 pmBaz, I cited a source for this fact. I believe the source is reputable and accurate.
If you think it’s wrong, perhaps you could provide a reason other than your mere incredulity?
Since the brain is remarkably good at signal processing, and can easily compensate for a lot of degradation (especially if it happens gradually), I’m not really sure what to make of your comment.
James Koster
October 7th, 2011 6:12 amI think people need to stop thinking about type size in pixels.
As has been pointed out “16px” may be perfect for some typefaces but for many it’s way too small, or too large.
The message in this post remains true but there’s more to it. Also consider the measure (IE how many words per line. “16px” is not always appropriate in, for example, a sidebar). Don’t forget the leading and most importantly the target audience.
For me, there’s no excuse not to be setting our type in ems these days.
D Bnonn Tennant, Information Highwayman
October 7th, 2011 2:06 pmGood comment, thanks James. A lot of people are taking my article as an ultimatum, which is understandable, but my concern is to get them to justify their font size decisions with facts and logic, rather than just base them on their gut.
I still use px for setting font sizes because I find it a real pain in the ass to use ems with a decent vertical grid.
James Koster
October 10th, 2011 2:35 pmHehe, it does indeed involve a few trips to the calculator and back. But it becomes second nature once you’re used to it. I can tell you how to create many pixel values in ems from a base of ~16px!
*sad*
Scott
October 7th, 2011 6:15 amDoesn’t your screen/print comparison imply we should be using *point* sizes for text, not pixels? On smaller and larger screens with different DPI pixel sizes will all be different, whereas point sizes will be the same (e.g. 12/72 of an inch).
D Bnonn Tennant, Information Highwayman
October 7th, 2011 2:07 pmHey Scott, that’s definitely a point worth considering. Especially as screen resolutions become wildly more disparate, I think points do make sense. Traditionally, designers (including myself) have resisted them because they’re not really appropriate to the pixel-based medium of the web. I wonder if that needs to be re-evaluated.
D Bnonn Tennant, Information Highwayman
October 9th, 2011 3:29 pmHi again Scott, further research shows that while this should be true, unfortunately it ain’t.
12pt on a 15″ screen is significantly smaller than 12pt on a 23″ screen.
The reason: operating systems don’t correctly identify the ppi of the displays they’re attached to. In fact, from what I’ve seen, pretty much every web-capable device created since 1995 is “hard coded” at 96 ppi—actual display resolution and size be damned.
steven
October 7th, 2011 6:16 amThe credibility of this entire article was completely lost when I scrolled down and saw the author dressed like a character from the Goonies.
Bronwyn
October 7th, 2011 6:19 amI habitually enlarge just about everything I read online. I’m still young, but 16px is on the small end of my “comfortable reading size,” since I see no point in ruining my eyes for something I can change quickly.
I’ve been using larger body text in my designs from the start, because if I don’t want to squint, why would anyone else?
DCuffman
October 7th, 2011 6:20 amInteresting read, even if you may not agree with the arguments put forth, the comparisons to print font sizes are informative.
blahedo
October 7th, 2011 6:21 amIt’s particularly charming that he writes an entire post about user-oriented design and making sure your users can see and read your content, on a page that has a huge blank bar on the left that pushes the content partially outside my window AND THEN also disables my horizontal scrollbar. I suppose this is technically Smashing’s fault, of course.
Mary
October 7th, 2011 6:28 amThank you for the article, and the research. As both a graphic designer, artist, and intellectual I often skip over completely a web site if it has illegible type, regardless of the reason, if I can’t read it you have lost me. And that for me is the bottom line. Assuming that a web site can afford to loose customers (that is their primary purpose, information and sales of some sort) is simply an indication of ignorance. As a fifty something, interested in all manner of information, I am a little baffled by the comments that it is alright to exclude me and my reading needs because I don’t fall into your supposed demographic. I might be buying for my grandchildren, what then?
David
October 7th, 2011 6:51 amI’m 56, keep my monitors at 32″ from my eyes, need significant correction from my glasses, have mild cataracts, and still prefer 12 point on screen. No neck craning here. But if your vision is typical, then I can see your point.
Avangelist
October 7th, 2011 7:00 amWhat a fantastic article! I have often felt that using the bottom end of the font size safety zone if you like (10-16) is due to needing to cram lots of unimportant copy into a small space.
Stefan Wallin
October 7th, 2011 7:09 amUsability Studies “NO difference”:
http://alexpoole.info/which-are-more-legible-serif-or-sans-serif-typefaces
Dingle
October 7th, 2011 7:12 amSpeed limit: 60mph for ALL roads.
Anything less is a costly mistake.
D Bnonn Tennant, Information Highwayman
October 7th, 2011 2:14 pmHowdy Dingle. Two things:
1. Could you explain the relevant analogies between font size and car velocity?
2. Since the article explicitly says 16 pixels for body copy wouldn’t a better analogy (assuming there is one at all) be 60 mph for all highways?
Gary
October 7th, 2011 7:12 amYou make a very good point. I can see why this works. Although on the smaller paragraph – if you set the line height correctly it isn’t as hard to read (still not as good as the 16px)
Gisella
October 7th, 2011 7:21 amI liked the post and wish the clients agree with this.
Unfortunately, in my experience so far, having worked with big companies (and especially the ones involved in kids entertainment, which usually like things being BIG), clients want to fit so much content ‘above the fold’ that using a 16pt font would be impossible. Let’s not forget MPUs and various ads, which need to be in a specific position…
Ric
October 7th, 2011 7:22 amA well written and reasoned argument, not to mention something that should have been said a long time ago.
Thank you.
Harish
October 7th, 2011 7:32 amI am a web designer and my eyes are not so great. Honestly I use a 12pixel font size for almost every single website I make.
But while reading your blog post, My eyes were more at ease. As soon as I read the author details, I felt a strain in my eyes.
Most of my clients would not like me to use a large font, but am sure going to use them on my own sites from now on.
Chris Howard
October 8th, 2011 4:13 amHarish, you will need to make sites at 19px, not 16px if you want to “enjoy” the effect the post had on your reading comfort.
Nate
October 7th, 2011 7:46 amLets not forget that to maintain credibility in a product, the design (in the web page) must still remain aesthetically pleasing.
Aesthetics are a large part of usability. (but, yes. I found this article very easy to read and agree with the 16px type).
*dislike the top nav
Holly
October 7th, 2011 7:53 amI’m a very long-sighted web designer and had to lean backwards from the screen in order to read the majority of this article, so from a personal stance, I disagree. However, my own website alternates between 12px and 14px font sizes, which sits nicely in the middle of the spectrum.
I do agree that large fonts are marginally better than the ridiculous 8 and 10px font sizes that became fashionable in the recent years, it’s a move in the right direction.
Mackenzie
October 7th, 2011 7:53 amWhy on Earth are you giving typeface sizes in PIXELS? One of my laptops has a 98 PPI screen. Another has a 135 PPI screen. Do you think 16px on these two screens is even remotely similar? No.
Font sizes should be done in points, just like typesetting.
D Bnonn Tennant, Information Highwayman
October 7th, 2011 2:16 pmMackenzie, this is something I’m researching at the moment, but my initial reaction is that I’m inclined to agree.
That said, designers have traditionally resisted using a print-based sizing system for a pixel-based format; hence my focus on pixels.
Patrick Samphire
October 7th, 2011 8:07 amOne of the things that has come up several times in the comments but which a lot of people aren’t really taking on board is that, these days, there is an awful lot more variation in displays. Most of those older websites that use 10 – 11px font size were designed when most of us were using low-resolution displays. 10 px looked pretty big on standard monitors less than 10 years ago. Hell, probably 5 years ago. Now, of course, on tablets or phones or monitors with high pixel density, they look terrible.
What this variation means, of course, is that you either have to know that the vast majority of your visitors have a limited set of display devices, or you need to use some clever media queries to adapt font sizes (e.g., increase the font size for common tablet or phone resolutions, or for very big screens). It’s going to become increasingly difficult to design something that looks great on my grandma’s old monitor and on a retina display if you choose a single design and font size for all of them.
nhavar
October 7th, 2011 8:07 amI’m an advocate for larger font sizes as well as more white space. I have seen some studies that suggest using some variability to impact readability, cognition and retention. For instance at times it may be helpful to have a small font in order to force the user to focus on something that requires understanding or retention. The difference between the small font size and the surrounding font size draws attention and the small font size makes the user focus harder and take more time to read/understand the material.
I think one unintended side effect of large font size is also a reduction in usable space. This can constrain content authors to be more concise. Constraint breeds creativity.
I have to agree with some of the other comments here about the choice of style detracting from the overall message of the article. The red, drop caps, serif font, shadow, etc., shock the user more so than the size of the text. I would have left the default styling in place and only adjusted the font size itself. This would have given the reader an easier model for comparison against other similar pages.
Alex Poole has a good post about legibility of serif versus sans serif fonts @ alexpoole.info/which-are-more-legible-serif-or-sans-serif-typefaces.
Viget is doing some studies on line length and legibility @ readability.viget.com
And I really like Joe Clarks commentary on some of the studies regarding typography, line length and readability @ blog.fawny.org/2005/09/21/measures/
D Bnonn Tennant, Information Highwayman
October 7th, 2011 2:18 pmThanks; I appreciate you sharing those resources. And no hard feelings about not liking my design ;)
Jason
October 7th, 2011 8:24 amI think it depends on context. 16px wouldn’t work for displaying, say, the homepage of the nytimes, but it makes sense for an individual article.
DJ
October 7th, 2011 8:59 amIn actuality – the front (home) page of especially a new site is designed to draw readers in to reading the full article. That’s why headlines and first paragraphs are nearly always larger in size. Our front pages should be the MOST easily scanable of anything.
Patrick Samphire
October 7th, 2011 11:07 amThe New York Times doesn’t match those criteria, though. It’s not a new site for the vast majority of its readers. They know what it is, and like, say, the BBC news website, they want to scan for interesting content.
Flux Appeal
October 7th, 2011 8:24 amI agree with using somewhat larger fonts + I focus on varying the sizing between headers, photos and other design elements to keep the eyes interested. I don’t think anybody really wants to read copious amounts of tiny copy, I know I don’t. Looking at the pages in different resolutions and zoom modes when designing helps me arrive at the best size to use for a particular typeface. Admittedly, this font is a bit large, but I suppose maybe the author is merely pushing his point…which I believe is valid.
Michelle
October 7th, 2011 8:25 amCompletely agree. I have been inching up the font size on sites I design and both clients and visitors are typically very appreciative. And though my eyesight is just fine (even though I’m 40, ha) I still find there are websites with excellent content I avoid reading just because the type is so small. And I’m a web developer! I know full well I could ctrl-+ the type up to legibility but it pisses me off that I have to, so I avoid the sites altogether.
Tom Rod
October 7th, 2011 8:28 amI find your subtitles font and the paragraph introduction font incredibly distracting.
Edward
October 7th, 2011 8:42 amTypo alert: under second photo you wrote: 12 INCH text?
Ben
October 7th, 2011 8:51 amIt’s so true that most users do not know how to increase the font size or zoom a web page. Make the site enjoyable from the start and you won’t need to rely on browser features you hope your audience is aware of. I’ve noticed that as I get older my designs and layouts are changing. More negative white space, narrow columns for body copy, brighter graphics, and yes larger font sizes. When I first started designing, small font sizes were the hotness, now those small font sizes can create a hot mess…
Over the years I’ve also become more focused on the user experience and not just the design. I often say “You can design the most beautiful thing in the world, but if no one is there to experience it, who cares?”. So I totally agree with the statement “It is about what users want and what will best achieve our client’s goals.”
Using larger font sizes will place more emphasis on typography and layout which should make for some interesting new trends (I hope!). Great article!
fearless
October 7th, 2011 8:55 amim confusing. good point :) very good
DJ
October 7th, 2011 8:56 amExcellent points. Equally as stupid as too small of font size is: putting the text in a narrow ribbon down the middle of a full sized screen. On the screen I’m viewing this on I have 2/3rd’s of it white space! (60%). What a needless waste! To me, that’s the REAL next “bleeding-edge” that web developers MUST be addressing. Lots more wasted space there than in type size – although even I, more and more, need to use the “zoom” feature the past 12 months; and it’s especially true on development sites.
Chris
October 7th, 2011 9:09 amGreat post, good points, but why does this post look so different? Isn’t familiarity a big part of legibility, esp on the web? By changing the format you’ve asked me to relearn something. Also, sup with the line-heights? A little shallow, no?
Steffen
October 7th, 2011 9:09 amI read the entire text… your absolutely right :-)
Mike
October 7th, 2011 9:10 amThis argument doesn’t really matter IMO.
How Users Read on the Web:
They don’t.
—Web usability expert Jakob Nielsen