Earlier this week, in an interview with Eric Enge, Matt Cutts made a statement about devaluing infographics. Naturally, this set Twitter buzzing, myself included. A number of SEOs have spoken up with a wide variety of opinions on it, so I guess I might as well jump in.
Nobody should be surprised by Cutts’ comment. The writing has been on the wall for a really long time. I wrote a post back in December called “Keep Your Infographics In Check”, to talk about why I felt this was on its way.
The Why vs. The How
One item we should be careful about is using Matt’s comments about “why” they want to devalue infographics to conclude “how” they plan to devalue them. I’ve seen a number of comments about how Google can’t fact check or gauge quality at scale, and even comments that boldly claim that infographics can’t get devalued at all.
Link spam analysis is a topic I’m very interested in, and have written at length about it. One thing about webspam analysis is that it’s not an exact science. They don’t have to get it “just right”. They just have to find a way to devalue with a degree of certainty, with a classifier that improves search quality just enough that they’re satisfied.
Panda and Pengiun are good examples of this. The why it was done, and the how it was done, were very different. Google cannot directly gauge content quality at scale, but they can find a lot of signals that serve as proxies, test them against a sample of quality rated examples, apply some machine learning and scale it out (Panda). All it takes is finding a set of signals that determines a measurable improvement in the direction they would like.
Always think of spam analysis in language such as “likely to”, “probability”, and “degree of certainty”. Panda did not have to completely remove content to be effective, it only had to find enough signals within a degree of certainty to shift content down a few positions down across a large set of longtail keywords to effectively drop a site’s visibility and increase searcher satisfaction.
Looking at What Cutts Did Say
“infographics can be created and that represent an OK form of promotion, but the challenge is that as soon as I say something like that, people are going to use this as justification to do whatever it’s they want to do.”
Google spam control is just as much a PR machine as it is a technical machine. Don’t read too much into Google not wanting to call infographics a good strategy. They have to avoid encouraging them. That doesn’t mean they think they’re all bad.
“infographics you create will do better if they’re closely related to your business”
We’re talking increased relevance scrutiny when crawlers discover an infographic embed pattern. We know they can do this (even if it’s not perfect). They can turn up the dial when they encounter infographic-like content.
“there is reason to believe that the link is more about the barter to get the infographic than a real endorsement of your site”
This could be a philosophical stance from Google on the relative value of links. Footer links mean less than content links, links closer to the top may count more, and site-wide sidebar may be indicative of paid links. Google doesn’t have to determine the intent behind every one of those subtypes, just prove statistically that when those types are devalued, search quality goes up. A footer link may be a great editorial link, but it’s just more effective to profile all footer links as lower value.
If Cutts is discussing it publicly, they’ve likely done some tests already to show that when these types of links are devalued, search quality goes up. Matt tweeted about the infographic plague back in December. I doubt they simply shelved the issue since then and are just now considering it.
The link is often embedded in the infographic in a way that people don’t realize, vs. a true endorsement of your site.
Replace the word widget with infographic.
I’d look out for this to be the type of behavior they really come after.
How Might They Devalue?
Not only can a profile be judged in aggregate by the percentage break down of your link profile, but the relative value of individual links, and new incoming links, can be gauged against both your current profile and distribution of new inbound links.
To demonstrate mass estimation:
(The example is an exaggeration to a degree.)
Above are two sites with very different link portfolios. Site One may be a lead generation site that is heavily propped up on infographic links, with very little brand equity, while Site Two is a more legitimate site with a more diverse portfolio.
An inbound link, from an infographic, to Site One may have its link value reduced based off the degree of backlinks that come from infographic sources already. This helps minimize the likelihood that a site is going to rank for a term not because it’s the best in the market, but because it’s the best at producing linkbait. That site’s skill is content production, not the value the rest of the site provides to users.
However, Site Two has a defensible profile. The distribution of link sources provides corroborative signals that it may include inherent value beyond its content generation. In that circumstance, search engines can feel more confident that the content of its infographics is more factually accurate than Site One.
It can be said that Site One’s content might be more factual, and even better, than Site Two’s. However, that doesn’t really matter. Site One doesn’t have the signals to backup its breath of expertise in the vertical.
I also believe, although I have no evidence, that links can be retroactively reevaluated. If Site One’s link profile increases trust / brand signals, the links Google was uncertain about at first look may go up in value.
For Site Two, an infographic devaluation is also trivial, while for Site One, it would be catastrophic. You always want to be in the position of Site Two, where no specific tactic devaluation can take out your business.
This type of devaluation is also similar to Truncated PageRank evaluation. If you calculate a domain’s authority without infographic links and divide that by the overall domain authority with infographic links, you get a ratio that’s indicative of that site’s dependency on a tactic. The closer that ratio is to zero, the less likely your site is to have used infographics as a manipulative link building tactic.
Commercial Anchor Devaluation
It’s safe to guess that Google is going to apply its exact match, commercial anchor text classifiers with heavier scrutiny on infographic footprints. We’ve seen this growing for several years, and more recently with Penguin. There are a number of papers from Google where that talk about evaluating a term’s commercial value and its likelihood to be spammed.
If you’re using infographics to sneak exact, and potentially phrase match, anchors of highly valuable keywords, I could see those devalued.
Irrelevant Embedded Links
The next big item they’re likely to devalue is when an infographic is being used deceptively. This was the concern with widgets back in 2008, and Matt Cutts language in his recent interview mirrors this concern.
A “Safe” Infographic
The infographic above is relatively “safe”. The degree of devaluation should be less significant. The infographic properly cites itself, linking back to its original source page. The ALT attribute used is the title of the infographic, which closely matches the content found on the other end of the link. The secondary link is a branded link, back to the homepage of the website that created it.
A “Risky” Infographic
The infographic above is inherently more risky. It’s pushed the degree of relevance, links to a page where the infographic cannot be found, using a commercial ALT text that matches the keyword targeting on the “money” page. Following that is a “Via” link that uses commercial anchor text to the homepage or another highly valuable page, which is also highly targeted for the keyword.
This pattern is pretty easy to match.
Degrees of Risk
It’s worth mentioning that there are degrees of risk and degrees of devaluation. The first example above might still be devalued based of Google’s philosophical stance that infographic links should count as lesser citations, in the same way they consider footer links to be lesser citations. The second example is going to be the type of thing they’ll want to devalue to a significant degree.
It’s also worth mentioning that infographics that 301 redirect the linked to URL after a period of time, reposting the infographic up at a new URL, could end up on the chopping block as well.
Infographic SpamRank & Co-Citation
This type of analysis is so simple, yet classic, and has the ability to be devastating to a lot of infographics. Search engines can build a list of infographic submission sites, which are effectively article content farms with infographics, and crawl out from that point.
Like this paper where Microsoft used forums to let web spammers expose themselves.
That list of sites you hammer away at every time you launch an infographic – Google can build that list too.
Once Google has identified a set of offending infographics, that have been rated negatively by quality raters, they can crawl out from that list and find common co-citations of places that post low quality infographics. These sources may have their ability to pass value to infographics reduced, and the infographics mentioned in these sources could have their ability to garner equity from other sources (even if trusted) minimized.
We’ve seen them do this with article directories and web directories.
So What Next?
It’s good to spend time thinking about how they might devalue infographics, instead of simply saying “they can’t” or “make good content”. Because A) I’m not going to bet against Google on a long enough time-frame (and I like to risk mitigate) and B) well, obviously.
With an idea of how they might do this in mind, we can start to think about how to adjust tactics to mitigate future and current risk, while helping to improve the future effectiveness of the tactic.
#1 Start to Diversify Now
If your profile is significantly comprised of infographic backlinks, I’d press pause and spend the next few months using every other tactic in the book that looks nothing like infographic links.
This should A) minimize the likelihood that your infographics will ever get devalued, B) keep your site from falling over if they do. A 10% dip hurts a lot less than 60%.
#2 Stop the Risky Execution
This is such obvious advice, I hate writing it. However, I know at least one person reading this has linked their infographic to a page the infographic isn’t on, or has 301 redirected their infographic, or launched an infographic on a non-commercial site with embed code going to a commercial site, or has stuffed exact match anchor text into the “Via” link.
All of those are pretty shady.
Sidenote: I’m not disapproving of particular tactics. I don’t have a stance. The people who know what they’re doing will keep on keeping on. No judgment. However, for the few who are just good enough to be dangerous, this could hurt you.
#3 Shift Your Relevance
Start being more critical about how well an infographic’s topic relates to the website. I’m guilty of this. I’ve gone after topics just because I knew I could get links, not because it was relevant.
If a site’s link portfolio is robust, trustworthy, and relevant enough, it can likely withstand a small number of off-topic content pieces that help increase domain diversity, but don’t get too carried away.
#4 Read This Post
I’m not going to spell it out, but read this post.
#5 Make Soup
Go watch this video from Kris Roadruck and listen to his analogy on making soup. It’s not about the perfect carrot, it’s the perfect combination. Don’t use infographics for your hardcore commercial anchors, use them for your domain diversity, brand equity, and social signals.
#6 Lead With Quality Citations First
Many link builders start their infographic campaigns by hammering that list of 25 to 100 sites they can submit to via forms, or pay a fee to get listed on. The very first sets of links start to follow a discernible pattern.
My advice for every infographic is to identify a list of high quality sites that you would like to get placed on. Then go read my post on Content-Based Outreach. Work through that list one by one, investing time, to get someone to agree to placement in ADVANCE of your publication. The very first set of signals will come from high quality sources. Also, if you work on these relationships in advance, and incentivize the coverage through items like customizations, time-gating, and interviews, you can negotiate link placement in a way that does not use a standard embed. They can write them in as editorial links.
Treat infographics like a PR consultant would, not an SEO.
#7 Repurpose, Repurpose, Repurpose
• Print it as a poster and give away.
• Chunk it up into smaller consumable infographics to use in different ways.
• Chunk it up and write guest posts featuring parts of the infographic. A lot of blogs may have never seen it, so from their perspective, you’ve simply created great images as part of a more robust post.
• Use it as leverage in a guest post pitch, and include the link in the body copy of a guest post.
• Use it as the basis of building relationships for resource content. “We created this great infographic, but as a follow up we created this great resource piece of content.” – sneak in the infographic link as part of the resource pitch.
• Pair with other types of outreach, such as giveaways and contests.
The goal of all of this is maximizing the number of non-infographic-type links to the infographic.
#8 Avoid the Infographic Directories / Blogs (or grab them last)
Don’t depend on infographic sites for links. These types of sites care less about exclusivity or getting an infographic after it’s gone “hot”. Having an infographic start here can actually hinder the success of an infographic because more reputable sites don’t want to publish content that’s already been published many other places.
If Google does relative link valuation, you want Google to gauge these links as additional links to a piece of content that already has authoritative links pointing to it.
In addition, simply listing yourself here places you in a neighborhood that doesn’t have a great reputation. This is something a site with existing brand equity can withstand, but could be harmful to a less authoritative site.
Keep Doing Infographics
Infographics are great ways to create content that is easy to share and digest. They also carry a lot of value outside of embed links. Even with devaluation, they will never become valueless.
As a tactic, it may become less effective for those who use them in very particular ways, but for those who use them as a content strategy, social strategy, brand strategy, or domain diversity strategy, nothing really changes. Google’s philosophical stance may mean they have less value passed per link, but that doesn’t mean no value passed.
If the content is good, and your readers enjoy the content, then forget about Google.
The secondary benefits are more than worth it.