Last Thursday, Bryan Eisenberg, one of the smartest people in the world of conversions & analytics, wrote a post called - The Web's Old Wives Tale: People Don't Read Online. An excerpt:
Web developers like to say it. Designers love to say it. Web execs feel good saying it to justify investing as little as possible–and in the lowest quality content they can get away with.
People Don't Read Online? Bull-crap!
If only I had a nickel for every time I've heard that statement, I'd make Bill Gates look like a pauper. Think about it. What's the first thing most people do when they get online… read their email. What's better yet, new research from the Poynter Institute's Eyetrack study released at the American Society of Newspaper Editors (proves the point)...
The Poynter Institute report is enlightening and absolutely valuable, but I think it's dangerous to read this post and assume that all web content can be structured in article format or that people are going to start taking the time to truly read everything on your pages. I still strongly believe in Steve Krug's rule:
People are going to read pages like this:
Article from CNN/Fortune
The topic is interesting, the lead-in makes you want to read the article, it's applicable to you (at least, to most of us) and chances are, if you've reached the page, you're going to want to get the value provided by the writer's story. However, on pages like this:
CNN's Gadget & Tech News Page
Or pages like this:
Grokdotcom's Home Page
And even pages like this (where it really does pay to read and the text blurbs are very short):
SEOmoz's Tool Page
... you won't get users to "read."
In these instances, and virtually all pages that navigate you to an article or blog entry or informational pages that provides exactly what you're looking for, your users are not going to read through the page and decide, based on all the information, what to click. They're going to follow that same anti-reading, quick-scan format described so brilliantly by Mr. Krug. This is the same reason you see so little traffic from Google if you're #8 vs. #1, even if your content, title and description are clearly the better choice for users. It's the same reason the browse rate on all but the best designed websites (and the worst with the most passionate users like MySpace) is so incredibly low. The online experience is all about speed and convenience and only in the rarest of cases will your users really take the time to carefully examine the page - by the time they've gotten to that point, it's because they're so incredibly frustrated that they're willing to give it a shot.
In fact, there's a perfect analogy here - ever watched a guy (yes, I'm stereotyping, but really both genders do this) try to put together a piece of furniture or work a remote control or fix an issue in a piece of unfamiliar software?
Rand: Oh, I've done this before. This is going to be easy.
Mystery Guest: Maybe we should read the directions?
Rand: No, no, trust me, I've done this a million times.
Mystery Guest: OK...Half an Hour Later
Rand: @#$%!
Mystery Guest: Maybe we should check the directions?
Rand: Fine! But they're just gonna tell me to do the same stupid thing I've been trying to do the whole time
Mystery Guest: ...
Rand: Oh, wait... Now I've got it.
Mystery Guest: Why did I ever agree to marry this guy?
All of us think we know how to use the web - and when web designers and developers don't follow the conventions we're accustomed to and make it blazingly obvious how to interact with their navigation and content, we wind up with scenarios like the one above. Tragic? Yes. Avoidable? Totally.
p.s. Not suggesting that Bryan Eisenberg was arguing with this point; I think he was trying to make a different one, but I'm worried that someone could misinterpret and wind up with some very high abandonment rates.
I really liked the post by Andy Rutledge about the Poynter study. If you have any control over redesigning a site, don't worry about eye tracking. But if you're looking at SERPs or an already built site, then they matter. If you can see where users are focusing their attention, then you might have a better idea where to place that link to the cool new and profitable feature or product.
Bud, you beat me to the punch, Andy Rutledge is a great read and one of the few blogs I check regularly because I WANT to. I'd recommend him to everyone who has anything to do with usability/design on the web.
Thanks for the link Bud. I hadn't seen that article and it does a good job refuting the Poynter study and many other eye tracking studies in general.
We were discussing this the other day in relation to a 'delete now' confirmation / warning step. We are already saying 'are you sure' as a separate step, but we were worried someone would still do it 'just to see what happened'.
Happened once early on after we built our first CMS - a client wanting to know where their pages had gone. Turned out they had deleted about half their pages. When they clicked the delete button, it said 'you are about to delete a page, are you sure?', but still, when we spoke to them, they had no idea they had deleted pages - they just knew they'd been 'trying things out'.
Yay for backups.
Hello Willcritchlow,
I found this out to be true to when testing usability within our CMS system. We even tried adding a graphic that flashed WARNING - WARNING. To my amazement they would still clicked the delete button. We sent out a memo explaining to the internal staff how important it was to read the CMS messages. Still it would happen over and over…
We ended up allowing them to delete, however the system would move it to a holding area. Now only the system admin can official delete an entry.
I believe user have been conditioned to click “YES” just to keep moving along. After a while those steps become background noise.
Yeah - a recycle bin was the route we went down as well...
You might have hit the nail with the point that users are conditioned to click "YES." What if you phrased the question the opposite way? Something like "You are about to delete the page. Would you rather save it?" so that "NO" means delete.
Another thought is what's the default. If people just hit return will the systems delete or keep the page. Maybe it's just a matter of making the default the option of keeping the page or perhaps even no default so users have to take an action other than hitting "Enter" to delete the page.
I think we're actually more conditioned to just hit "Enter" to slerts and trust that the default was what we wanted.
Very good points. There is also a lot to be said for the primary option being the verb (i.e. 'delete' or 'cancel'). But you're right that hitting enter shouldn't confirm.
This is a superb post. Thanks Rand. While it has long been accepted that 'more =less' online in terms of text, and while I have repeatedly told clients that that was not necesseraily the case, there was never any 'proof'. I am waiting hungrily for that Poynter report :-).
For my own site, the more info I have put on PRODUCT pages, the better my conversions have been. I'm going to revert to my old rant that online is less 'trustworthy' than going to a shop...
two REALLY basic points - People are:
1. not sure of refund policies - once they commit to a purchase they REALLY commit
2. wanting to make sure this product/sevice REALLY fits their needs, and is as good-as-perfect a match to provide them with the desired solution.
How do they satisfy this need online? By making sure.
How do they do that? BY reviewing all the material on hand regarding their intended product purchase thoroughly. the more you can provide that is relevant, topical and authoritative, the more likely the potential customer is to be convined that what you are offering online is a GENUINE solution to their need.
Product Pages vs Home pages - well it depends. My home page is split into clear sections with teaser copy about each main product, with copious CSS links and pretty pictures. It seems to work as my bounce rate is pretty good/low. I focus the majority of my product info, research and information on the specific product pages which is really where the product specific quetions will be asked in the visitors mind. There's no point juct cluttering page for the sake of including reams of text.
Product Page Copy - this I split via relevant teaser or descriptive headings, I use a lot of bullet points, and divide the text to follow logically, guiding the reader through the page with questions and solutions. i am constantly working on it, and I've found (via testing on a few sites) that the best way to answer a customer thought query, is to actually physically print-ask the question on the page in best matching 'query language' and then answer it , leading the visitor to the next section of information - like visual hand-holding by gently drawing the eye along. The more the visitor feels the site is an authority, that the owners have put effort into it in terms of design and content, the more they will 'decide' that the site is authoritative, legitimate, and the more likley they will be to trust the message being sold.
So in my humble opionion, more does NOT = less online - in fact good quality, defined, relevant, topical, easy to navigate text that exhibits logical sequencing is likely to engender trust, and online I have found that trust is likely to win the sale.
Obviously this does not necessarily apply to big brands as the brand name is usually trusted, and so the same rules don't apply. In big brand cases - less may be more when ability to perform a desired action may be of more importance, exspecially to brand loyal consumers - eg - view a new [car] model online, book a [hotel] room online. Clear navigation leading to desired action based on brand recognition and user requirement is likely to be of more importance to visitors than reading a whole bunch of irrelevant text. ofcourse big brand sites are all about buzz generation, speed, complementing other media and maximizing online sales... but that's for another post I think...
appreciate your brand notation. might be nice to see a few more threads on brand vs entrepreneur sites.
I'm going to agree with the consenses regarding 'it depends'. home/index pages can be very different in both appeal and KPIs to the rest of the site. Perhaps a little more on that too...
I hear lots of people talking about paying $4 per article for website content. No wonder people do not read! We now live in the age of BS Content. Writing is now something that is sold as a commodity. BS is sold by the ton and so is website content.
We seem to have an ongoing clash in the usability field between the laboratory and the "field". I'm an ex-academic, and I think research has a lot of value, but I know too many academics who don't see that a generalized outcome in a highly-controlled environment may not apply to the real world as much as they like to think. On the other hand, I also know too many businesspeople who turn up their noses at academics and assume lab results are a lot of mumbo-jumbo.
The hard reality is that the science of usability quickly becomes "art" once you start applying it. Even if conditions are perfect, we have to balance other factors (SEO, marketing, client requirements, profitability, etc.). I think we've got to be willing to get our hands dirty, which means doing our best to understand the research, trying to apply it, and seeing what happens. Like you and Rand both point out, the answer is almost always "it depends".
I think people in general just don't read the information in front of them.
We run an insurance site where people print an insurance policy after they pay and maybe 20% don't print or save it and then have to ring to ask for a duplicate.
This is a very good topic. One worth looking into if you are one of those content publishers that doesn't know how different page structures affect your user interaction and/or length of time your users spend reading.
It's funny, I get comments all the time from readers on one of my blogs where they ask me questions and the answers to their question was right above their comment or all over the site within text.
I think it depends a lot on the topic they are looking for and the original search query. Some market users just don't read. They call, email and might even drive down to your location, but they dont read. Other market users read everything! It just depends and who you are and what you do, IMO.
It is so easy to generalize on the web with broad, sweeping statements. Those statements will always be correct... and false.
We often forget that there are many elements that affect visitors:
The challenge then is identifying whether a page needs to target various elements or focus on elements, but can hopefully funnel visitors to other pages based on their needs..... more information or less.
I think it all depends on the reader's mode and mood. What type of info is he/she looking for? Is he/she buying something? Is he/she doing a scientific research? What is his/her interest?
I would agree with this sentiment; it isn't an on or off question whether people read or not. Are they reading for pleasure, shopping, work, hobby research? I think people read in several modes depending on their level of engagement with the topic matter and their purpose for reading.
For example help, support, and technical documentation it's pretty well a given people skim looking for the one sentence or paragraph that will solve their problem. (Unless they're really nerdy and read technical documentation just to learn ... ahem) If they don't find it they give up relatively quickly and switch to another document. They do this regardless of how the information is presented to them, though obviously the format can lengthen/shorten this reading time regardless. I think the main factor in these cases is tolerance, that is, the reader's tolerance for scanning the document for what they want to find. This kind of reading is usually unpleasant, the engagement level is relatively low and the purpose is relatively specific. Reading for pleasure is nearly the opposite; if the reader is compelled by the topic they will actually read the entire thing (within reason).
I think that if you can identify the "mode" of your content you can tailor it to the kind of reader who will be visiting your site.
EDIT-Rand, there's a typo in the first sentence of the second last paragraph, "All of think we know" should probably be "All of us think we know".
I was thinking much the same. I don't think how much people read online is cut and dry. A lot depends on the specific person and their particular motivation on the page. It also depends on the page itself.
A main section page is probably there more to direct traffic so the moment you identify where you want to go you stop reading. If the heading is enough you don't need to read the description below.
An article is meant to be read and will most likely get read a lot more.
I also think just like there are different stages in the buying process there are different stages in the reading process. When you first encounter a page you might be scanning it to determine if it's worth the investment in time to read it.
Assuming you've determined from your scan you do want to read more you'll probably start reading more in depth from the top. As you read you are probably re-evaluating your decision to read and either increase your focus or start skimming again.
And of course that process will vary between different people and thier current motivation.
I think a lot also depends on the length of the article. I know that I've started reading posts or articles, gotten six paragraphs in and realized that I have 9/10 of the page to go. Generally, I don't have the time to spend half an hour reading at work, or home. At that point I give it up as a lost cause and skim on to something else.
Interestingly, I think I read magazines and newspapers the same way; if they don't catch me in the first paragraph or so, I move on. If they successfully summarize the entire article (or at least seem to) in the first paragraph, I tend not to read on there either. You need a hook to keep people reading.
This is an interesting point.
Print journalists are often wary of giving too much away in the intro (or standfirst as they call it) for fear of people not bothering to read as they think that they know what the articles is about.
Online, if you don't summarise the article in the intro so that people will even bother to start reading it, then it won't get read at all!
I think it is an interesting study. I agree that "people don't read online" is rubbish. I know when I am looking for something online, whether it be a quick bit of info, a product or some other resource, I scan things. I scan the results, then I scan the page itself. If I am intrigued by what I've read while scanning, I will read the whole thing. If it is a news story or blog, and I feel I've gotten all the info in the first paragraph, I don't read any further and simply hit the back button.
I think this is one of the challenges with doing anything online. Regardless, people ARE reading online.
Really interesting and true... this only confirms what I have learned through crazyegg in the sites I manage... if there is an image at the upper left side people try to click it.. always!.. even if it is the logo.. that takes you to the homepage.. and hey!.. they are already at the homepage... then they start going down.. trying to click every image .. that attracts far more attention than text... even if it is a landscape photo.. :)
There's now an interesting follow up post on the Wordblog site which raises some interesting doubts about the validity of the study, or at least the conclusions reached from it...
We are "opting-in" to read the content when we read online (didn't Seth Godin just blog about this...? I read far too much online *winks*) so I think we do "read" at news/blog sites and yet I think when shopping we may go in to visual mode for awhile - looking at the pretty pictures.
So your examples play in to that - sometimes we window shop, sometimes we research.
We just never always do one or the other.
Bloody-minded humans that we are *winks*
"People don't read online" is just complete garbage. People engage with content differently online -- but they certainly do read.
Maybe not everyone, but not everyone actually reads everything in a newspaper or magazine. If the article grabs your attention -- you read the whole thing. If not, you skim it. Same online.
I really need to find the recent study I heard about that covered this exact issue.
And -- even if people aren't reading it online -- the content is bringing people to the site and they could be printing it out to read. How many people are testing for that variation?
i think that in most cases the conclusion is ok. but it depends a lot!
it depends on the reader/consumer and the phase he is in...
let's look at you exampla with the mistery guest. in the fist stage you were realxed, you didn't have any pressure to do thething right, you thought you knw it all. but in the second stage, you felt the pressure, you needed to read to understand... an you did it. and you would have read it even it the instructions were written in the most complicated way :)
I learned this from operating an e-comm store. On a daily basis I would get emails such as "do you offer free shipping?", "do you ship to <insert country here>?", "does that price include one set?". My response was always polite but in my mind I was always thinking "Don't you people friggen READ??".
My point is, that sometimes it's just not enough to simply put the info on the page. If it's not interesting/compelling/attractive enough... it won't get read anyways.
If you really want something to be read, my advice is to put "don't read this!" in bold!
Great point Crash, working in customer service a while back I swear people would intentionally ignore any warning or advice we gave and call simply to see what they could get away with.
DONT READ THE NEXT SENTENCE!
Homer Simpson Loves Donuts.
See now if you followed directions, you wouldn't be hearing Homer say, "MMMM donuts" in your head right now.
Got me there! I read in order:
I'm afraid I read the "Homer..." line then the "Don't...." line then the first paragraph then the last one.
Yay for dyslexia
The original proposition was never that users don't read *at all*, just that they don't read as carefully and thoroughly as they would in print. Which is exactly what you're getting at here. If it's "blah blah blah" stuff, they will ignore it. If it's relevant and useful and seems to lead towards their goal then they will.
Jakob Nielsen has a lot of research about this as well. He talks about how it is more difficult to read online than it is in print. For that reason, users may scan more on a screen than they would in a printed page. This is one reason why it's important to have good print stylesheets for longer content (people may prefer to print it out to read it).
Sorry I commented on the wrong post.
Rand,
Thank you for making some excellent points and the flattery. It was much appreciated. The caution is whenever someone grabs onto a piece of data and assumes it is a rule. At the end of the day, this just shows that people may read online. The may contains lots of facotrs. Is it relevant to them (easier than it sounds), is it written to engage, is it formatted for screen, do you have supporting imagery, etc. This does not mean vomit all your words on a page and hope the read what stuck. You need to work very hard, like you do at SEOMoz to create amazing content that people desire to read. it's all in the motivations.
Rand
Great post....or I think it might be, if I took the time to read it.
I take one exception, which is the RTFM example (Read The F.....g Manual). Even the instructions should be written to be scanned, not read. When was the last time you read the manual for say, a telephone?.
I believe most manuals are written by cryptographers. That is the only explanation for the hundreds of pages, 6pt font and undecipherable vocabulary in 10 different languages.
That is why I applaud the savvy companies that have a 'Quick Setup' or 'Quick Install' one page pamphlet alongside the technical manuals.
Edit: Recently I spent 2 futile hours of a Sunday trying to install my plasma tv into the separate tv stand without reading the manual on 'how-to-insert-a-tv-into-a-stand'. Reading the manual would not have helped me install the stand, but it would have told me the company sent me the wrong bloody tv stand (for an older model).
I think that the important line in the study quoted is
But of course, what this doesn't cover is that if people don't choose to read something, or don't read a search result that makes them want to click through, then you will have a readership of 0%. Whereas the people with the newspaper, have already made a decision to buy that paper and so at least some readership is guaranteed.
Writing for the web is an incredibly important skill, and I think that Megan summed it up when she said that the original point has alwasy just been that people read differently online.
Writers just need to work a little harder, and I find that telling people that no-one will read their story if it isn't written well, is a good way of getting them to pay attention.
Proof that people go to sites, not to read, but to find what they want. And they'll look around for it, rather than reading to find it.
I actually had the "read" the post twice before I fully understood what the post was about. I'm not retarded, just that skim through the things I read most of the time and when I finally get to a great article, i find it hard to slow down.
I blame my skim reading to crappy content online (I admit it, I've created crappy content myself) and my extremely tight schedule. If we can somehow put the breaks on for users when the time comes for them to actually "read" then life would be awesome.
I guess all you really need to remember is "Less is more" when it comes to designing and "Quality of Quantity" when it comes to content
A lot of good points are brought up here. I agree with Rand in noting that the title (whether it be intriguing and pertinent to the reader) is important, as well as the layout of the page. If the text is tightly packed together, the reader will automatically feel as if it will be a laborious task to read (though the content may be very informative and useful).
Breaking things up with spacing, charts, graphs, pictures, etc. helps as long as the whole orchestration does not look too packed together as well.
I also agree with Mr. Wailes in noting that people "find what they want" rather than "reading to find it." I think the proliferation of the Web causes some anxiety in browsers and they get the feeling that they cannot spend too much time in one place because there is so much information available.
Adding something of my own: I think garnering a sense of esteem and respect for your site plays a huge part in perusal vs. skim. It also depends on the nature of the site. For instance, I read the Moz and Search Engine Land meticulously everyday. It is rather difficult to read off a monitor than from a book but I do it because I know of the value I will acquire from the sites. I have a reverence for them and know that it is worth my time. If I am looking through the Web for information about surfboards and I don't know of any reliable site per se, than I will skim through multiple sites until I find one that I find to be exceptional and I will pledge my allegiance there.
I think blogs and forums play a part in readership because people of similar interests can turn each other on to worthwhile sites.