Infographics on the web are so bad and so broken. They are everywhere, yet few actually do a decent job of conveying information (click on the one at left to see what I mean). Some even argue that they are ruining the Internet. They tend to be formulaic and overreaching, often cobbling together too much information instead of focusing on the one or two nuggets that are truly useful. (How much better would most infographics be if they pulled out the most salient chart or set of stats and discarded the rest?)
There are many reasons why they suck. Primary among those is that they take too long to make, and the underlying data is difficult to assemble. Today, they are driven more by marketing budgets than editorial discretion. Most of them are created by companies and distributed for “free” to blogs and media outlets as a form of PR and viral marketing. The publishers eat them up because it is free content that would otherwise be expensive to produce. Design shops sometimes charge a few thousand dollars to create a single infographic.
I know we can do better, which is why I’ve applied for a grant from the Knight News Challenge on Data to build a data visualization platform to address some of these shortcomings (please like it or reblog it on Tumblr). You can also learn more in this Q&A about the project I did with Jeff Davis.
As I note in my Knight News application:
Infographics are very popular on the web, but most of them aren’t very good. The information inside them is trapped. They tend to be flat files, unsearchable, and most are not interactive.
And yet people love them because humans are visual creatures. We can absorb more data more quickly by glancing at a chart than scanning the same numbers in a table, or reading through a few paragraphs. Publishers love infographics because readers can’t stop themselves from clicking on them. (A whole sub-meme exists for infographics about infographics, including the one below by Think Brilliant, which is actually a rare example of an effective infographic).
What publishers need is a better way to create and present visual data. We need a tool to produce well-designed infographics on our own—quickly, efficiently, and cheaply. And not just one-size-fits-all infographics—all kinds of data visualizations, from simple bar charts to interactive maps and timelines.
But wait. Aren’t there a growing number of startups already tackling this infographic-creation problem? Yes, companies like Visual.ly, Infogr.am, Vizualize.me, Tableau Software, and iCharts are creating tools in this general area. And that’s great. If they can do a better job creating these visualizations, I’d love to work with them. However, all of them currently embrace a bring-your-own-data approach.
Producing a great interactive chart is only half the battle. A platform that taps into existing data and makes it instantly chartable is what is missing. Finding the right sources of data in a chartable form is the hard part. There is lots of compelling data all over the Internet: social data (Facebook and Twitter), company data (CrunchBase), financial data (SEC, Yahoo Finance), geo data (Foursquare, Factual), government data (data.gov), product data (Amazon). It all exists in various silos, and most of it cannot be browsed visually.
The big idea here is to create a data visualization platform where data providers can plug into one end and data visualizers can plug into the other. It will be open in that anyone will be able to import or create their own infographic and charting templates. Some of the data and charts will be free, and some will be for sale. But the more open, the better.
Initially, the platform will be geared towards bloggers and news organizations, but could expand to other industries and types of data. Again, from my Knight News Challenge proposal:
We are solving this problem for publishers. First, we will create a library of interactive chart templates, which can be expanded and contributed to by others. By creating templates, we will make it possible to produce high-quality data visualizations in an efficient, repetitive fashion which can be embedded anywhere.
We will also connect existing databases and work with data providers to offer a growing menu of chartable data sets geared towards journalists. Journalists will be able to bring their own data, but over time they will be able to find more of what they need baked into the platform.
The best data visualizations out there today are bespoke and almost hand-crafted. That doesn’t scale for web publishing in terms of either economics or speed. Most blogs and web news organizations don’t have an art department. A platform for creating decent looking interactive infographics is certainly something I would use, and I suspect other bloggers and news publishers would embrace it as well.
If you think it’s a good candidate for the Knight News Data Challenge, please support it by “hearting” the application or reblogging it. If you are a data provider or a company creating data visualizations, let me know what is the best way to work with you. And if you are a programmer or information designer and would like to get involved, please contact me (erickschonfeld at gmail).
We are awash in data, but we can’t even see it. The data visualization platform I envision would be a step towards fixing infographics so that they actually tell us something new.
Hi Erick, we at Visually are creating a platform that connects customers with our community of visualization creators to build quality infographics and data visualizations. Most of your desired solutions are being created by us and other leaders in the industry. There are a lot of talented people doing amazing things in data visualization. As the largest platform in the world, we are at the center of this innovation. We strive to grow and bring this young industry mainstream for the greater good.
Quality infographics have 30X the traffic and sharing of comparable text articles. I would hardly call this the ruin of the Internet. If anything they are energizing the publishing and advertising industries. We work with orgs like UNICEF and Earth Hour who attribute infographics to helping educate a new audience to their causes. My experience at Mint.com as VP Marketing and at Visually as CEO shows that quality visualizations only benefit users.
Just because people click on them doesn’t make them good.
I do applaud your efforts to make them better, however. And I look forward to seeing you launch a more general-purpose visualization engine.
Like I said, the more options the better. My platform will be more of a marketplace for both data and visualizations, packaged together. Think of it as a marketing and distribution platform for Visual.ly, married with data. Happy to set up a call and tell you more.
Thanks for the comment.
Sent from my iPhone
hey eric –
if clicking isn’t a good metric for infographic quality, then what is?
light a candle, don’t rail against the darkness…
– dave mcclure (full disclosure: investor in Visually, bcz they’re friggin’ awesome)
ps – and please no “I know it when I see it” pornographic definitions 😉
The problem is that clicking only means that the ‘surface’ of the infographic (it’s visual appeal is good. You clicked to look at it more deeply. But that doesn’t tell if the payoff was good. And for infographics that’s what counts. For example if I click and am disappointed the true result should be a net loss (since all future infographics from this site are now at a bigger disadvantage than they were since I am less likely to waste my time on them). Whereas if I liked it the win is a lot bigger than the one click count.
Sharing would be a better metric – or discussion.
Oh – and nothing against Visual.ly – I know one of the founders. But I would say they are still at a pre-alpha stage – there is a very LONG way to go – and again, nothing wrong with that – good if anything – they have had the fortitude to take on a very, very hard problem.
Wow, really? I almost can’t think of a *worse* metric for “quality” than “clicks”. How about “number of correct and accurate footnotes citing actual sources that exist and contain the same information as the infographic”? I can’t count the number of infographics I’ve seen with either no sources, or a list of sources that had nothing to do with the information they present. (I can come up with other possible criteria, if you like.)
If “clicking” is a good metric for quality of a visual, then (in my experience as a sysadmin) those trick advertisements that look like Windows95 windows that say “Your PC might be infected!” are the highest quality advertisements ever. And our best literature is the novels and tabloids found in supermarket check-out lines.
Every other field in the world has long ago come to the conclusion that “number of (initial) sales” are a terrible proxy metric for anything else. For measuring quality, we use professional reviewers, and don’t make up excuses about reviewers simply regurgitating “I know it when I see it”.
When I want to know what the best movies were last year, I might look at what movies won Academy Awards, or look up Ebert’s Ten Best list, or possibly even check IMDB ratings. I’m certainly *not* going to look up which had the most ticket sales.
look, you have to offer a practical metric. views & clicks are at least measurable, whereas the things you mention are terribly impractical & long feedback loop to measure. we can’t realistically send the infographics to the Nobel Prize judges or laureates for feedback. in addition, I’d prefer a populist rather than an elitist / academic metric…
Porn has way more than 30X the traffic of any text article. Does that automatically make it good, too? Porn has energized the advertising industry way more than infographics ever will.
UNICEF should start using porn to spread their message. I guarantee they would never have trouble with money or publicity ever again.
well to be honest, porn is in the eye of the beholder… so yes, I would say that’s a good metric. people probably pay a lot more for it too.
Having been in the business of helping distribute them, via our PR and SM efforts, I cannot agree with you more. For me personally, looking at the selections I was to present, this exercise ended up with me being frustrated.
From my editorial and writing side, asking someone like you to put CRAP up, was just not in the cards. I look at Pinterest now (every morning actually) only to find my screen crammed with distribution my fellow PR’s have SPAMMED the webwaves with.
Now, as the Editor of TechCrunch signifies with this article, it appears the water has been spoiled. This is kinda sad for me, because I like the little informative bits, especially the interactive ones. Your prompting to get a better mousetrap here, is highly appropriate. No fuzzy love ether Eric, people look to you to get the value there.
I don’t mind people creating bad infographics makes the good ones stand out more.
Except that you don’t know it’s terrible until you’ve already wasted time looking at it. I’ve become incredibly wary of infographics and now, anytime somebody proposes making one, I shield myself.
The problem is that people are all to often looking solely to drive traffic and get links as opposed to provide any semblance of value to the consumer of the graphic.
And there is a great point. Nothing worse than becoming engaged in something, only to find time wasted away at the end. What Eric is suggesting in this conversation is actually quite powerful. Back when, all the tech gurus considered object oriented development, something beyond drag and drop, into the real time, the next gen. I think with some effort, we may be on the cusp of something like this.
Dave up there asks a very pertinent question tho – “what, if not clicks?” Seems to me, measuring the “drill down” and bounce may be in the neighborhood of an answer. Now we are approaching this from multiple angles tho. Point of my comment is, “value first” as you suggest Ryan.
“It will be open in that anyone will be able to import or create their own infographic and charting templates.”
“…a step towards fixing infographics so that they actually tell us something new.”
An open platform is more likely to produce a greater number of useless (and/or repetitive) infographics as a greater number of people attempt to jump on the infographic-as-marketing-vehicle bandwagon. Simply because the data sources are presumed to be reliable and tons of neato templates have been created, even truly great ones, it doesn’t mean that all of the sudden the quality to crap ratio will do a 180. It more likely means there will be 1,000 times more infographics. that while relatively accurate from a facts and figures perspective, don’t necessarily offer any really valuable or well-thought material. But fortunately we’ll all be even more inclined to give them a little more time on picture because the templates look so good?
That said, advancement of data viz tools and data access as described will also result in a greater number of truly meaningful outputs.
The solution, IMO, is that the most respected media sources we trust not to put drivel text-based content in front of us need to up the quality control. When less of them pass muster, get posted, and garner traffic and links, the fad will die down to a reasonable level.
Can someone just create an infographic about this story so I don’t have to read all of it?
If only there was a platform that let you make one easily.
But that might ruin the internet.
the poor, poor, internets! (lulz 🙂
For the low low price of $2700, not only will I create one, I will send it to Erick to see if he will publish it 🙂 Half kidding.
Make it $27 and you’ve got a deal.
Sent from my iPhone
LOL: Fully with you on that Erick, there ought to be a way to make these things by dragging and dropping, fill in data, blah blah. I will work on the $27 solution. Currently peeps charge a bit to work even crummy ones out. Ugh.
If you have to search for data to graph, your graph sucks to begin with. Infographics are best when they’re used to present new data, not color up the old. You may be adding to the bad-graph phenomenon.
You need to find data before you can graph it. Today, that data is static. What I am proposing is a place where you can find data feeds which can keep updating your charts. So really what you would be doing when you browse the data is looking for interesting relationships that tell you something new or surprising about the world.
Maybe it’s a feed of financial data: Apple’s profit margins versus the rest of the mobile phone industry. Once you find that relationship and the underlying data, your chart can keep updating every quarter automatically as new data flows into it.
Thanks for the clarification. I love the idea of a dynamic data feed for infographics. Very cool idea!
And the data will be somehow embedded in the infographics that get passed around or posted on various social media sites?
It’s not just the infographics that are broken. Nor the ‘marketing’ approach. The biggest problem is a lack of understanding numbers and data and especially statistics. Ideally that is what the infographic is for. To explain quickly and clearly what the numbers alone obscure. But most infographics are either ‘explaining’ something obvious and not hidden by the numbers, or they reveal that the numbers were misunderstood in the first place.
again, seems like the infographic itself is a hammer — it’s neither good nor evil, smart nor dumb, it’s just a tool, a paintbrush for its author/artist. that artist might be Da Vinci, or it might be a 8-year-old with limited skills. I fail to see the infographic itself as the Grand Evil, rather than simply the limitations of the person behind the brush.
Absolutely. I think what Eric is saying though is that just at the moment that ‘hammer’ has become very trendy and everyone is seeing their problem as a nail because of that.
One of the reasons Edward Tufte is so good at this stuff is that he remembers that and looks for the best WAY to illuminate the data. His first book should be required reading for everyone interested in this topic (the rest start to go down the rabbithole rather and are less impactful)
Tufte is WAY overrated. some of his visualizatins are just as bad at conveying information as the crap erick is complaining about. just because he wrote a book and a bunch of academics think his stuff is great doesn’t mean the average person’s understanding is better. I’d suggest there are a number of more modern approaches that are quite a bit better at making infographics easy to understand.
if it takes writing 3 books over 200+ pages long to explain best practices for infographics, you’ve already failed. I don’t want a tome on best practices, I want a few simple examples that illustrate useful techniques for communicating information briefly and clearly.
If you’re experiencing a plague of infographics, it’s possible you’re reading the wrong websites.
Is the purpose of infographics to communicate data? I thought there was just a fun piece of snackable content that was fun to look at — art in the same way that memes are art.
No – that would be a graphic – without the info…..
fun is good communication, and good art 🙂
Bad infographics are like bad anything – the lower the barrier to entry and the higher the incentive to engage, the more crap you will get. Infographics are easy to build and gratifying – you can see the visual result and share it easily, even if it gets lost in a sea of mediocrity. The web is experiencing a shift in priority from features to design principles and infographics are an outgrowth of that trend. They are not good or bad per se, but are an effective or ineffective communication tool for something else.
I’m most intrigued by the prospect of connecting to a data visualization platform to access meaningful data. In one way, this is great because it makes it easier to find useful data that adds value in an infographic. In another way, it exacerbates the issue because it makes it even easier to create infographics, giving the false impression that they are an intrinsically valuable medium that don’t require a (human) design sense.
Simply put – the biggest issue is not a lack of connection between infographics designers and good data – good designers will find the data. A good design sense and an analytical mind is a rare combination, nothing can change that. Infographics (and potentially pedestrian data visualization platforms) are sending a false impression that anyone can do it, like GeoCities convinced everyone that they could – and should – build a personal website!
And Erick – the day you create a tool that causes marketers and PR flacks to stop creating 90% swill and focus on adding real value and meaning to the conversation, you will be a rich, rich man. Still, I see value in your idea as long as you can convince designers who understand how numbers and trends interact to plug into your platform. Best of luck!
exactly. the painting is only as good as the artist.
don’t blame the canvas.
Lower the barrier for good visuals, and more people will be able to find their inner artist (see Instagram).
Let the crowd find its inner artist and you get the new Gap logo or VitaminWater Connect more often than not. Rather than increasing the incentive to produce, find a way to increase the incentive for quality of production (raise the bar rather than lowering the barrier).
Lowering the barrier is putting grown-up tools in the hands of toddlers. In terms of complexity, Instagram is a hammer, sometimes it works out – infographics are guns, much higher chance something goes terribly wrong.
Pingback: Infographics Are Broken. We Can Do Better. | Swipping It
At metaLayer we originally set out to do something very similar, also focused on news creators and journalists. However, very early on we spotted a bigger problem to solve which we decided to focus on. It’s a problem that we think make us complimentary to an effort like yours.
We are creating easy (think drag and drop) methods for people work with unstructured and semistructured data in a single environment so that they can output it into other platforms (whether they be visualization platforms, spreadsheets, or other web apps etc.)
In short, we’re building what one might call “the Excel for unstructured data”, making it possible for non-experts to do incredible things with the world’s information.
That said, If you guys are successful in creating a marketplace of visualizations as well as a market place of data sets (feeds or static) I can think of several synergies for working with you.
We’re more than open to working with new projects like yourselves if you don’t find us competitive. Check how people are using our products today at – http://metalayer.com
Oh BTW, we also have an entry in the KNC – newschallenge.tumblr.com/post/24195512269/metalayer-turns-anyone-into-a-data-scientist
Would love to learn more about what you have planned. I just signed up.
“Most blogs and web news organizations don’t have an art department.”
They don’t? I’m not in the industry, but I would assume any legitimate new organization would have an art department.
“A platform for creating decent looking interactive infographics is certainly something I would use, and I suspect other bloggers and news publishers would embrace it as well.”
Is creating infographics such a problem for you? Why don’t you just use one of the other services (that you mention) then?
“However, all of them currently embrace a bring-your-own-data approach….The best data visualizations out there today are bespoke and almost hand-crafted. That doesn’t scale for web publishing in terms of either economics or speed…We need a tool to produce well-designed infographics on our own—quickly, efficiently, and cheaply.”
Ok so let me get this straight. Publishers want standardized templates with standardized data married together. What’s the differentiator between publishers then? If I am publisher A and I want to show a pre-canned infographic of a publicly available data set, then what makes me any different from publisher B?
“nfographics on the web are so bad and so broken. They are everywhere, yet few actually do a decent job of conveying information (click on the one at left to see what I mean).”
Conveying information to who? The end-user/reader, right? Unfortunately, the reader is normally not the one paying, the advertiser is.
“Publishers love infographics because readers can’t stop themselves from clicking on them.”
So who’s issue are you trying to solve then?
“What publishers need is a better way to create and present visual data.”
If publishers sourcing model is to attract eye balls and any infographic (bad or good) attracts those eyeballs, then where is the issue?
“I would assume any legitimate new [sic] organization would have an art department.”
You assume wrong. Most bloggers do their own visuals for every post. Sorry to break it to you.
“Is creating infographics such a problem for you?”
Yes, obviously. If the tools existed, I’d happily use them. What’s out there now for the most part only lets you create a few limited infographics based on a handful of data sets like your Facebook or Twitter data. Or it’s still desktop software.
“What’s the differentiator between publishers then?”
You can start with the same data and come up with different insights and graphics. A simple example. If I had basic financial data for all public companies, I could compare the revenue growth of Apple and Google, while you might want to compare the profits of Apple and Samsung. Same underlying data, different graphics. It is still up to the reporter/publisher to pull insights from the data and show that to the reader. And, of course, publishers will always be able to import their own data to differentiate even further. Today there isn’t even a good starting point.
“If publishers sourcing model is to attract eye balls and any infographic (bad or good) attracts those eyeballs, then where is the issue?”
That is the issue right there. Not every eyeball is the same. Raising the bar will attract better eyeballs.
“You assume wrong. Most bloggers do their own visuals for every post. Sorry to break it to you.”
Bloggers != News Organization. Sorry to be pedantic, but i’m having a hard time understanding exactly who your market is. Is it Joe Shmoe blogger, a world-class news organization, or a large publishing house? Your post mentions all three.
“Yes, obviously. If the tools existed, I’d happily use them.”
So what’s wrong with Visual.ly for example? On their frontpage, I see a myriad of infographics that are not only a good starting point, but present data visualizations comparable to your example. I get the impression you have limited experience with any of these tools. And even if that’s not the case, I think a much more worthwhile discussion would be pointing out line by line why they don’t work. Rather than just a blanket “it’s broken!” statement.
“That is the issue right there. Not every eyeball is the same. Raising the bar will attract better eyeballs.”
I’m not sure I understand the conclusion there. Every eyeball is the same to the advertiser. Advertisers don’t discriminate between a better or less eyeball. Give me a relevant eyeball and I’ll pay for it. Again, back to my point, are you trying to solve the publisher’s dilemma or the readers?
There are plenty of blogs that are news organizations (TechCrunch, GigaOm, Venturebeat, AllThingsD, The Verge, Business Insider, The Next Web, Politico, etc.). But anyone can break news these days, even people who aren’t professional bloggers or journalists. The market extends from professional news outlets to individual bloggers—anyone who cares about presenting information in a visually compelling way online.
Visual.ly is great, but all those infographics on its front page were not created using its tools. They are curated from around the Web. Right now, its own tools are limited to visualizing your Twitter or Facebook feeds: http://create.visual.ly/. They are working on more general purpose tools, but those have not yet launched.
Every eyeball is not the same to advertisers. There is a reason they pay $20 CPMs for media that attracts a desirable demographic and $2 CPMs for crap.
Best illustration of how the same data can result in different graphics – and by extension how to lie with infographics.
My biggest problem with the whole field is that most people can’t see these kind of issues when presented with just one graphic.
Seriously – read this. http://www.fastcodesign.com/1669222/a-case-study-in-how-infographics-can-bend-the-truth
Eric, there are several startups like BuzzData, Junar, Juice Analytics and Timetric that offer social data collaboration (the GitHub for datasets model, if you will) that are making great strides towards connecting the underlying data to visualization tools. That’s not to say that this is a solved technical issue, but there isn’t a vacuum either.
My own opinion is that the folks who make most infographics don’t understand how to create a meaningful narrative, regardless of their artistic skills or the availability of data or tools. It’s not a problem that you can just tell someone to “read more Tufte”.
“A platform that taps into existing data and makes it instantly chartable is what is missing”
This is exactly what InfoCaptor is trying to build “a platform” for Data visualization. Data can be finger-typed or pulled directly from spreadsheets or database. The data and visualizations are abstracted so any visualization can work with any data
I agree Mashable is broken.
Gapminder (http://www.gapminder.org/) does a good job of visualising epidemiological data from various sources – you may want to check out how they glue things together.
Gosh, I love this post. Yes, clicks are one measure of ‘success’ – but the real success should be measured by repeat visits. Do visitors now see the infographic producer as a content expert? That’s the gold marketers should be striving for.
Disney has best iv’e seen
Pingback: Collective intelligence online: How does it look? « wikimobs
Pingback: Basta de 'esas' infografías (II)
Pete nailed it; most infographics don’t tell a good story. I’ll take it one further, as I’ve questioned that many marketers and business communicators don’t tell a good story – to the right audience, they don’t adequately brand the graphic, and as you mention don’t make it active enough to generate the kind of traffic they want for ‘success.’ Many I’ve seen cite outdated, even conflicting data – and you’re right, often way too much info to be useful. I like the idea of ‘untrapping’ the data from flat files, finding better platforms to accurately chart, visualize current information in ways more useable to the reader, more accessible to communicator. FWIW.
Pingback: Persuasive interfaces in open data | Manufactura Independente
Pingback: Investor Presentation Infographics « GroMobi
Pingback: Liens vagabonds (10 juil) | Metamedia
Pingback: Can We Please Stop With the Info-Free Infographics? |
Pingback: Being Tired Of Infographics Is A Lazy Argument | SMS
Pingback: Remember when infographics were actually… oh, informative? | Credible Copywriting by Francesca StaAna
I couldn’t agree more. Though some publications are fine with going after clicks and views by any means necessary, others value quality content and won’t settle for bad infographics. I believe we’re at a transition point in the evolution of the web where static numbers and flat images just aren’t going to cut it. Nice to see that others are just as interested in finding better ways to show information on the web.
Pingback: The Art of Visual Storytelling: Using Media as a Storytelling Technique
Pingback: Online Marketplace
Pingback: Time to Diagnose and Cure InfographICK | SKANOWANIE co z czym i jak
It’s hard to find your page in google. I found
it on 15 spot, you should build quality backlinks , it will help you to get more visitors.
I know how to help you, just type in google – k2 seo
Wonderful goods from you, man. I have understand your stuff previous to and you are just extremely magnificent.
I really like what you’ve acquired here, really like what you’re stating and the way in which you say
it. You make it entertaining and you still take care of to keep it sensible.
I can’t wait to read much more from you. This is really a
Pingback: PR Industry Report July 27 | Polaris Marketing & PR
Pingback: Infographics in financial services: who does it well? | Bluechip Communication
Pingback: Can We Please Stop With the Info-Free Infographics? | THINKInk