It's tiiiiime! Once again I've gathered some of the best minds on link building to have an in depth discussion on the state of link building in 2013.
I've been doing this series since 2007 – you can check out the past editions below:
- Link Building with the Experts – 2007 Edition
- Link Building with the Experts – 2008 Edition
- Link Building with the Experts – 2010 Edition
- Link Building with the Experts – 2011 Edition
- Link Building with the Experts – 2011 Edition Addendum
- Link Building with the Experts – 2012 Edition
2013 has been the year of continued drama at the hands of black and white animals – namely the Panda and Penguin algorithm filters. And there's a ton of input on both those topics – and many others – featured in the post below.
If you're new to the Link Building with the Experts series, let me explain how it works. The people being interviewed each submit a question that they themselves want to hear input on from the other panelists. No one sees the answers by the other panelists until the interview is published (I answer the questions myself before sending them to the rest of the panel). I've always felt this makes the series interesting for two reasons…
First, we get higher quality questions. Secondly, there is no “head nodding” because no one knows how anyone else answered the questions. Each question gets an answer with no outside influence regarding “what the rest of the panel thinks” so to speak. We've all been doing this a long time, but that doesn't mean we always agree or don't have different methods of achieving the same end results.
Meet the link building interviewees:
I'd like to sincerely thank everyone below for continuing to share their knowledge and for giving their time to this series. They're all freaking awesome.
Now that you know what you're in for, grab a cup of coffee and get ready to learn about link building tactics and theories from the talented and insightful (listed and answered in alphabetical order by first name):
- Aaron Wall of SEO Book – Twitter | Google+
- Debra Mastaler of Alliance Link and the The Link Spiel – Twitter | Google+
- Eric Ward, Ericward.com Linking Strategies – Twitter | Google+
- Jim Boykin, Founder and CEO of Internet Marketing Ninjas – Twitter | Google+
- Julie Joyce, Director of Operations and Co-Founder of Linkfish Media – Twitter | Google+
- Michael Gray of the Graywolf SEO blog – Twitter | Google+
- Rae Hoffman, aka Sugarrae, CEO of PushFire – Twitter | Google+
- Rand Fishkin from Moz – Twitter | Google+
- Roger Montti, the founder and owner of martinibuster.com and publisher of Advanced Link Building Strategies – Twitter | Google+
- Todd Malicoat, aka Stuntdubl, at the helm of FishingCharters.com and SEO faculty at MarketMotive.com – Twitter | Google+
- Will Critchlow, Founder of Distilled – Twitter | Google+
With that, let's get started…
1. Are links becoming a smaller portion of Google's ranking algorithm? If so, what do you believe are the factors that are new and/or growing?
AARON: Absolutely Google is folding more aspects into their ranking algorithms. I think there is far more weighting put on aggregate usage data than is let on by Google engineers. Another big driver of rankings is localization, with more and more queries being localized over time (both the inclusion of the local Google+ pages & localizing many of the remaining results).
DEBRA: Links are still one of the dominant factors Google uses to determine how a page ranks but there are new trust and accountability signals being used to determine the value of those links and pages they point to. Google Plus, reconsideration and disavow tools are providing those signals.
ERIC: In some ways they are actually becoming a larger portion, but what's happening is the types of inbound links that Google is willing to trust and reward is becoming much more exclusive and “bad” links are now blowing sites up. If you look back over the history of Google's link devaluation process, you can see what they are doing. They devalued links coming from junk directories, devalued links coming from link networks, devalued links coming from paid brokers, etc. But what nobody talks about is the converse of this, which is that the highly trusted links that were there all along are likely sending stronger signals when looked at as a percentage of all links analyzed. Imagine you had a site that ranked #1, with 1000 inbound links, and then 990 of them were devalued, leaving you with only ten trusted links in your profile, and your site vanished from Google. Contrast that with a site that has 1,000 inbound links but 275 of them are high trust. It's a pruning process, and people are still in denial about what they need to do.
JIM: Yes links are becoming a smaller portion of the algorithm, but that being said, I still believe that they are the largest and the most important part of the algorithm today… It's certainly true that “exact match” anchor text, which was the SEO ranking key in years past, can be the poison of today. It's also true that links that once worked, but now can be spotted and mapped via patterns, are getting sites penalized by Penguin. If you've been doing too much “cheap SEO”, you might one day be getting familiar with some Disavow tools to analyze your backlinks….because if you did too much SEO in the past, you might have to disavow them in the future. Some links are great, then one day, they could be poison. Google actually doing a good job of making SEO's “grow up” in the field of marketing, because now if your links aren't “Natural”, chances are they can hurt you. The short tail game has ended, and in many cases is now poison to sites. Links are very important, but you can't “force them” today.
New and Growing Factors… All the stuff that spewed out of Google's mouth for years, has certainly come true now in the days of Penguin. Links are the most important thing…and Google's getting very good at finding links that are not given via quality people or quality sites, where it's all natural and not “paid”.
Exact match anchor text as a “bad” thing is new…google mapping the social web and getting closer to giving social signals more strength in the algo is new…analyzing things like “return visitors” (loyalty), “brand strength”, “having people think about the SEO they were doing years ago, that they stopped doing, but can hurt them now is new…thinking about “who” your connected with (including social), is fairly new. Authorship, and WHO links to you is growing in importance. Who you're connected to and how you connect to them is growing in importance. How people interact with your website is also growing in importance. ….thinking about “who are the authors for your site” and who are the people you want to get “connected with” (social or link) is fairly new..I could go on and on…next question :)
JULIE: I think that they are becoming a slightly smaller portion as other factors are being added in but not to a significant extent. I think a lot of factors like click through rate, time spent on pages, personalized signals like how many times a user visits a certain site or ignores one, authorship, and frequency of socialization of content will all be a part of the algorithm but to me, those are all tied in closely with links. Unless they rebuild the algorithm from the ground up, I don't see the importance of links drastically decreasing.
MICHAEL: I don't think saying links are becoming a smaller portion is the right way to look at it, what I do think is that links no longer stand alone. Google wants to see many signals that you are a legitimate business/website (aka a brand) and they want to see the user data and social signals that come along with links. Lets say that over the next three months you got 500 links of good or better quality, If no one was passing your url around on Twitter, Facebook, Google+, Gmail, etc that would look odd.
If over that same time period no one was typing in your domain, your company name, or your domain/company name along with a keyword phrase into a search box, that wouldn't really make any sense, and would not represent normal behavior. So Google would think those links are suspect and possibly artificially manipulated, and probably ignore them. The genius of that plan is if you are a traditional SEO with no marketing finesse, you would say hmm I guess I need more links, and you would dig yourself into a “linkhole” (I'm totally taking credit for creating that phrase) that you'll never get out.
RAE: Do I believe Google is desperately trying to find more and better ways to “validate” the authenticity of links? Absolutely. But links are still the core to ranking in the current algorithm. I think we hear the “links are dead” mantra because people that build a shitload of keyword laden links are getting hit for it – therefore “link building is dead” when in truth, it's simply that Google is getting better at identifying artificial links and giving more credit towards more “validated” links. “Easy link building” is what's dying.
Links still matter – in a huge way. But people need to realize that a natural link profile is key – and it's all about the percentages. We hear a lot about the big brand bias – when in reality, I think that big brands merely have more “legit” and 100% natural links to mask any deliberate link building they're doing.
At this point in the game, you have to spend time building 1. Links that matter from a ranking perspective and 2. Links that don't matter from a ranking perspective but keep your link profile looking as natural as possible (I.e. Developing nofollow links, image links, etc).
RAND: I don't see them diminishing massively, but as other signals are rising, they certainly feel less overwhelmingly powerful and SEO itself feels less like a link building process and more like a positive result of getting a holistic marketing strategy right. Some factors that feel like they're on the rise include:
- Branding & brand signals
- UX – especially design, speed, and mobile friendliness
- Mentions & citations (not just those containing links)
- Far more sophisticated content-based analysis
- Social signals (or the secondary signals that are often the result of social success)
ROGER: Links have been on a years-long journey away from being purely a ranking signal to becoming a negative signal in itself, something for the spam filter. It started as a ranking signal then quickly became a relevance signal once Google started depreciating links based on on-page relevance factors. The further depreciation of the value of links continued with whacking links based on the location on the page. That was the evolution of the concept of unnatural links, growing in response to methods such as expired domains (remember those!), text link brokers, and reciprocal linking. Links moved from ranking signal, to relevance signal then links as a signal moved to having one foot in the ranking algorithm and the other foot in the “un-ranking” algorithm.
The concept of unnatural links is now associated with a broader range of linking PATTERNS related to anchor text, paid links, paid advertorials, and other repetitive patterns that do not occur naturally. The common aspect to all of these signals is that they are types of links that are being used as signals of EXCLUSION. This is important to note. It gives you a clue about the algorithm. To make an anology, the algo resembles that rule created to by shopkeepers to keep out the hippies: “No shirt, no shoes, no service.” If your links are the equivalent of no shirt and no shoes… You get no service. So are links becoming a smaller portion of Google's ranking algorithm? To a CERTAIN EXTENT, links have evolved to become signals of exclusion. But web publishers are getting around these limitations by co-evolving their efforts by focusing on links as freely given citations, signals for inclusion, as described below.
Previous algorithm improvements led the SEO community to encourage “natural-looking” links. It wasn't about natural links, it was about unnatural links looking natural. While many are still plugging away at that and older methods, many are rethinking strategies and coming back to the first and original principal of building links: Build something and tell others about it. The latest iteration of this very old strategy centers on the DIRECT PROMOTION of content as an INDIRECT method of building links. You're going to be hearing more and more about Content Promotion because it dovetails nicely with traditional marketing and Search Marketing; skills in any of those two will suit the approach.
The idea behind Content Promotion is that you create a particularly good piece of content and use all available means to get the word out about it. This differs from the direct method of asking for a link. In Content Promotion there is no asking. That's not to say you can't ask, just that the marketing energy is focused on the promotion of the content- with links, referrals, and traffic following on from it.
This is similar to Viral Content Marketing, with the difference being that Viral Marketing is limited to promoting via social/emotional triggers (anger, amusement, outrage, etc.). That's not always a good fit for some companies and the results are generally mixed. In a nutshell, Content Promotion is about creating content and promoting it with social media advertising (particularly Facebook Ads,StumbleUpon paid discovery, BingAds, Zemanta and Twitter), AdWords (both keyword and content), and traditional press release marketing. The content can be virtually anything, from a limited promotion to a video tutorial. The idea is to stimulate actual citations, real links, based on the merit of the content. There is nothing more natural than that.
TODD: Links, or offsite equity, has comprised the lion's share of importance with regards to organic search engine rankings for the last decade or more. Google has made great strides to segment and devalue many types of links that have proved to be poor indicators of relevance and authority, but the power of links to rankings remains. Personally, I don't see other factors becoming more important to relevance anytime in the near future despite the problems that have always existed in a link-based algo.
Other search ranking factors that have been introduced in recent years in many ways complement offsite equity rather than replacing it. Many of these signals, such as social media and brand mentions serve to validate the quality of links rather than to completely replace the necessity of links in the algorithm.
Despite the best efforts of some enterprising spammers, Google has kept links and offsite equity as a valuable and reliable part of their algorithm since inception. These indicators are still valid, but will continue to be augmented and validated by social graph data and better user data sources.
WILL: This is a more complicated question than it seems. I think it's making increasingly less sense to talk about the “ranking algorithm”. My mental model these days looks more like a bunch of decision engines and associated separate ranking algorithms. Many of these separate ranking algorithms (local rankings, choosing the correct knowledge graph entry, SPYW-type social stuff, QDF etc.) have much less dependence on links than the regular “10 blue links web search” algorithm. At the same time, the proportion of queries that trigger something other than a traditional ranking is growing.
So: yes. The overall influence of links in the traffic you will get from Google across a broad range of queries is decreasing. I'm less interested, however, in whether they are a declining factor in “regular” search than in all the ways the search UX is changing.
Probably the biggest argument in the other direction is that the growth of Penguin and associated updates means there are more and more ways that some kinds of links factor *negatively* into the overall algorithm.
Aside from arguing semantics, the most important thing is what effect our theories have on our *actions*. On that side of things, I don't anticipate any reduction in the value I place on great links on high value pages (you know – the kind that humans click on) any time soon, but I *am* becoming more interested in link-engagement metrics than in nominal link-strength metrics.
2. Google is still trying to crack down on the buying and selling of links, and now they're sending out warnings to webmasters that they've detected unnatural links coming from their sites. Do you expect to see link buying getting harder, or will it just get more expensive?
AARON: I think it will get both harder and more expensive over time. If you are a big company who can generally do no wrong then I do not think the risks are as high as claimed (even when Interflora was hit they were only out for 11 days & were allowed to triple dip on branded keywords in AdWords while penalized), but the smaller you are the greater the risk of “poof it's gone” if you are aggressive with link buying.
For those who want to play at the spammier end of the spectrum, increasingly ranking spam pages on parasitic hosts that Google is over-trusting is going to create a higher ROI than doing bulk link buys into their own websites.
DEBRA: I see a little of both since less people are blogging and many of the previous outlets hosting paid links have been identified and devalued as such.
ERIC: I'm sorry, are you saying people actually BUY links to rank higher? This is blasphemy. LinkMoses is horrified. Ha! Link selling and buying aren't going away, people will just try to get more clever about how they create the arrangement whereby money and links can be exchanged. Here's an example that has nothing to do with SEO or search rank at all, but does involve paid links. Look at http://solarcar.engin.umich.edu/quantum-sponsors/ Now you tell me. Are those paid links or not? Yes, they are. Do some of those sponsors look like they might have been after a little juice? Yes, they do. But, are these they type paid links that should cause a penalty or devaluation? In my opinion, no they are not. And this is a great example of the slippery slope.
JIM: For those who are link building with the intent to get exact anchor text to targeted phrases, yes, it's going to get harder ..by harder I mean, Your Going to have to do some WORK. you're going to have to come up with ideas on what you can create to cause people to talk about it….and you need to come up with lots and lots of those ideas, and then you're going to have to execute those ideas and then you're going to have to market those ideas….what are you “buying for links today”…if it's “links”, then you're not going to be in this game long. If you're buying great ideas, and if you're buying the execution of those ideas, and you're doing real marketing to real people and not the Whore sites of the internet, then sure…one can call it “buying links” and it just got a whole lot more expensive because the “store” is only selling poison links simply by being virtue of a store. Linking and connections needs to be natural now. It's hard to buy someone's trust to connect to you and to mention you. Now you have to be outstanding, now you have to build your brand, now you need to get them back, now you need a great writer, now you need a community. The game of “buying links” , in my view, is “over”. It's a game of “Marketing”. Be better than your competitors, and earns peoples trust. Money doesn't work anymore.
JULIE: I think that it will get more expensive and more difficult, but I also think that will be true for many types of link building and not just paid links. I don't think the paid link will be going away any time soon. I think people buying links will get smarter and will adapt to whatever algorithm changes occur, just like SEOs do with everything else. There will always be webmasters who create sites just to make money, and if they have tons of throwaway sites, nothing will stop them from selling links. Despite how closely we follow Google's every move, there will still be webmasters and clients who don't understand the risks and they'll still want to buy and sell links.
MICHAEL: I don't think link buying will get harder, I get a dozen emails a week offering to sell me links somewhere on someones over inflated network of 400 sites. However buying quality links, that still work will get harder. For the people who do sell links, unless they are exceptionally good at how they do it, they run the risk of becoming the victim of a 3rd party drive by shooting from the competition. While Google would like you to believe they can detect bought/sold links with near 100% accuracy we all know it's false. I've seen content pieces that contained a purchased links that you had no way of identifying, as the content was that good and there were multiple commercial links in the page.
The problem is most people buying/selling links are as subtle as a Vegas Drag Queen at a convention of Tibetan Monks. People buy links and say I'm paying for this so I'm going to get the most bang for my buck and go for the best commercially valuable anchor text I can get, usually with something like this …
I was looking for a cheap rental car in Orlando, I finally found Luigi's Rental Agency.
Normal people don't link like that, they link like this …
I was looking for a cheap rental car in Orlando, I finally found Luigi's Rental Agency.
One final word on the subject, link buying/selling is a high risk tactic and against Google's Guidelines. While I don't have a problem with it, far too many take the chance without fully understanding the risks or explaining those risks to their clients. Very few sites are able to recover unscathed from a link buying incident once has been discovered by google. If you are able to convince Google you have amended for your sins and are going to stay on the straight and narrow from now on, most of your traffic will come back, but it will never be 100%. You should also know some people are doomed to the purgatory of traffic so low its irrelevant, and will never again see the light of day no matter what they do. So choose wisely.
RAE: Honestly, I've never gone the link buying route. Here's what I can tell you though – site owners are definitely more “afraid” these days. Not only about selling links, but about everything… accepting advertisements, or guest posts, or linking out to affiliate products, etc. So it only makes sense that link buying will not only be harder, but also more expensive – especially in SEO savvy industries. And if I were buying links, I'd avoid link buying / selling networks like the plague.
RAND: It will likely get both harder and more expensive (at least to do in ways that last as long as old forms of link buying used to). At some point in the future, I expect most of the webspam and manipulation that happens through link buying will be re-routed to other activities with a more reliable, less dangerous ROI.
Caveat – this is really only true in English-language markets. Outside the US/UK/Canada/NZ/Australia/South Africa, spamming & link buying is still much easier and more common, likely because Google's engineers haven't built as sophisticated detection & devaluation methodologies in those regions/languages.
ROGER: Unnatural link warnings ON SITES about paid links have expanded warnings and penalties related to UGC (User Generated Content) Spam ON SITE and in backlinks. Google has moved on from the paid link wars and opened up a new front against forum spam. So rather than discuss last years war, I'm going to address where Google is NOW. Unscrupulous SEOs have a long history of buying links for clients because it's easy to do. That strategy has become riskier and some of them have turned to outsourcing to third world forum spammers. There is a white hat subculture that believes spamming forums with “useful content” with links in the signature line is a win-win situation. This notion propagated on some “white hat” sites has created a green light for UGC spam.
I am not JUST a consultant. I have thirteen years experience creating my own sites, including award winning forums that have been featured on sites like BusinessWeek Magazine. I have actual experience being on the receiving end of spam and I research who is doing the spamming, including phone calls to the clients in order to track down the sources. In my experience, there has been a growing trend in forum spam originating from 3rd world countries on behalf of “white hat” SEO agencies. I believe this is why Google has expended warnings and penalties beyond link buying and are now making it harder for agencies who have moved on to forum spamming.
TODD: Link buying has always been a slippery slope. Since there will always be sponsorships, advertising, and promotions, there will always be some level of “link buying” going on. Google has shown little tolerance for link buying, and the enforcement has been rather severe at times.
Buying links has become much more expensive because the reward isn't nearly as high as the risk anymore. For years, that risk/reward ratio made purchasing links directly a profitable endeavor for many websites. Now the only people making money with links are the broker networks “selling the shovels” to all the people who are years late on the “let's go buy all thelinks goldrush”.
Finding a way to creatively tell, sell, and distribute your message is a far superior approach to just trying to buy your way to the top. Even if you're in the payday loan vertical, this point is being proved on a more regular basis, and now even with a specific algo filter targeting it.
The pure cost, and risk cost of buying links has relegated it to a strategy that is not viable for most businesses. It's an expensive proposition to invest heavily in a website only to have it fall to a google penalty or algorithm update.
WILL: I think that it may actually get cheaper.
I guess that's a bold claim but it's based in economics. As Google spreads FUD (justified or not) about the risks of straying, more of the deep-pocketed brands will dial down or quit their link buying. This could easily take the bottom out of the link selling market. We are definitely seeing more and more brands asking how they can wean themselves off their cash-based tactics. In some cases, this is because it's become less effective (or downright harmful) but in even more cases, it's based on changing perceived future risk.
I assume this is a big part of Google's plan.
3. How do you split your efforts between small, lightweight content and campaigns and big ticket investments where you can't know for sure if the 10x investment is going to bring 10x return?
AARON: If you are working with a new client & the relationship is based on either limited resources or limited trust then in most cases it is probably best to go with the certain low-cost option, but as the relationship is built up you can do more swinging for the fences.
If the 10x return is only going to have some x% chance of succeeding then you need to do the math on the potential returns vs the costs. So you really need 10x (x% chance of success) = $y … you need to be able to have $y be greater than your cost on the project & you need the client to be willing to run those sorts of campaigns enough times until a few of them take off.
DEBRA: You never really know if a marketing campaign will take off, it is best to invest in a little research before you start to maximize your efforts. I have three steps I follow to help me with this:
#1 Evaluate. Before starting any marketing campaign evaluate your personnel strengths and where your website stands in your industry. If you don't have the manpower to follow-up on opportunities popping up as a result of your campaign, you may lose prime links. Look to see who is ranking ahead of you and why, and then look at the sites behind you. I am of the opinion sites behind you should be watched closely, they want to move ahead and often test and implement aggressive campaigns to get there. Don't be caught off guard, know what the successful sites (those ranking in front of you) have done to be there and what hungry owners (those behind you) are doing to jump ahead.
#2 Research. If you have the staff and financial resources to implement a bigger marketing campaign, research what’s been done and try to gauge the success of the promotion (see my tool list in question #4 for help with this). Do a short survey to your customer base or if you don’t have a customer database, join a forum and ask for input. Every time I do this I get great ideas on issues I never considered and also uncover influential people to tap for help and further research. Most big ticket campaigns fail because the owner did not invest the time and research to find the right markets and communities to target.
#3 Disengage. Know when to quit or not start a promotion after going through the first two steps. Conversely, if something works, tweak and reuse as soon as possible to capitalize on the momentum.
ERIC: By making sure the client is given full disclosure about what the expectations should be for any approach they are considering. Do not be afraid to say “This is an expensive bet, and we can't know what the outcome will be”.
JULIE: I look at what the client wants and can afford, both financially and in terms of risk. If it's a big brand with a successful history and they want to try something where the ROI is uncertain, I'll try it. If it's a small business who can't afford a lot of money and risk, I'd be more cautious and do a lot more of the lighter weight work. To me it all depends on what makes sense for that one client, at that moment in time. If they've been penalized or are really suffering, I'll be a lot less likely to recommend something where I can't give them at least a 51% chance of having some success.
MICHAEL: This question is similar to a how do I find a good work life balance … the answer is different for everyone, and I don't think it's fully attainable, it's going to always stay just out of reach. I don't think you ever want to start out chasing the big money vanity keyword phrases, unless you have a good bankroll and a decent amount of resources you can throw at the problem.
Plan out your top 5 keywords, but don't directly chase them yet. develop your second and third tier keywords and start chasing those. For example if your top keyword is [El Dorado Hotel Reviews], start out building high quality pages for [Family Travel to Eldorado], [Fine Dining in Eldorado], [Best Beaches in Eldorado]. Don't worry that these aren't big money pages, what you want to do is build brand recognition and establish that your website is an expert for [El Dorado]. Then start moving up the food chain [Beachfront Hotels in Eldorado] and [Villas in Eldorado] before pushing on your big phrases. Understand that people are very seldom going to link to your commercially motivated pages, they are going to link to informational pages or pages that have human interest, like legends, ghost stories, or interesting and colorful moments in an areas history, etc.
You can try and get links or do a social media push for your hotel booking pages, but unless its exceptional it won't get a lot of love. However if you push story about a legendary mining prospector who was murdered for his gold claim by the town bullies, and who's ghost mysteriously returned from the grave to exact revenge on his killers and who's spirit is still seen today, you'll get a lot more traction.
RAE: From a consulting perspective, I tend to be a bit more cautious in aiming for the homerun so to speak. It really depends on what the client has already done and what their budget is (and what they feel comfortable spending on a “hit or miss” campaign). If they haven't covered a lot of the basics, then we'll start covering them and suggest higher investment and thus potentially higher return campaigns as they start to see results from our initial efforts.
From a personal site perspective, I tend to be a little more willing to swing and miss, haha.
RAND: We've found a healthy balance by investing in a few things daily/weekly (stuff like blogging, social sharing, community engagement, etc) and a few big projects at any given time (e.g. the Algo History, Mozcast, the Ranking Factors, etc). Based on our analyses, those big projects actually have a much larger ROI on average (even accounting for the ones that fail) than the daily stuff, which surprised me quite a bit!
ROGER: An eighty dollar bottle of wine doesn't always taste better with food than a forty dollar bottle of wine. I have a reasonable expectation of the equation between effort expended and responses received, taking into account the aptitude of those working for me. Scaling up successfully depends on the quality of those workers. You can put ten butts in the chairs and can expect ten times the return IF all ten share the same aptitude for the work. In the real world some workers are less better than others. Some people get it and others, for whatever reasons, are too distracted to do the job. If YOU have workers who can focus and are dependable, the only limitation is the strategy.
TODD: Most websites need a mixed marketing strategy based on their size, resources, and goals. It's crucial to spend some money on marketing experimentation, but much more important to make sure the majority of budget goes to time tested tactics.
Every site needs some core / cornerstone content to answer the questions of the visitors and retain someone to the site. Creating content that attracts links is another problem entirely. Webmasters and site owners are realizing that B- level content will never draw the links that a piece of real linkbait will. That B- content might not even draw the engagement needed to maintain rankings with the way Panda has filtered poor content based on user data.
Even great linkbait fails – if it was easy, everyone would do it. When you budget for linkbait, you have to budget for some failures, and also focus on the process. If you can master the process, and truly create A grade content within your vertical, it will always succeed in the end. Hoping that one article will fund the link equity for an entire site, however, can be downright foolish.
WILL: I keep reading and hearing about the 70 / 20 / 10 model which advocates placing 70% of your budget into the “same old” content – stuff that is solid, dependable and has worked for you in the past, 20% into experimental deviations from that and 10% into “moonshots” – the totally unpredictable high risk / high reward stuff.
I personally feel like 10% isn't high enough for the moonshots, but I'm looking forward to hearing others' thoughts.
4. If you're limited to 5 SEO tools in your toolset – what are they (and for what)?
AARON: #1 – Our SEO toolbar – a quick “at a glance” baseline for the competitive landscape
#2 – Ahrefs – in depth link data with great interface usability
#3 – Google AdWords keyword tool – it has some known issues (not returning data for certain keywords, defaulting to broad match, etc.) but they have more data than anyone else
#4 – Advanced Web Ranking – self hosted rank tracking
#5 – Screaming Frog SEO spider – lots of uses for on-site analysis, it is like a more modern version of Xenu Link Sleuth
DEBRA: #1 Timer – I suffer from “oooolookitdat” syndrome and tend to wander if I don’t set the timer and keep myself in check. It also helps keep track of time spent on client projects.
#2 Domainsbot – WHOIS and domain name checker.
#3 Buffer – I set up my daily “informative” tweets and let the tool do the work.
#4 LinkDetective.com – Analyzes back links by type.
#5 SocialCrawlytics – Identifies most shared content.
ERIC: #1 Link Prospector
#2 Be A Google Ninja. Know every possible operator and method for surfacing content [addition by Rae: 25 killer combos for Google's site operator]
#4 Open Site Explorer
#5 Xenu (Still great after all these years, and free)
JIM: #1 They'd all be our private tools…but, for public ones:
#2 SEMrush – Organic and PPC keyword values and rankings for client and competitors
#3 MajesticSEO – backlink analysis – competitor analysis – site analysis
#4 SEO News to keep track of the latest news
#5 Spyfu – Keyword research and competitor research
JULIE: #1 – Evernote. I know it's not a proper SEO tool but to me, it is. I document everything in there. I document what clients want, what they send me, what they've said to me on the phone, new ideas that I have for clients or myself, brainstorming sessions, notes for my articles, notes I take from what I read, etc. If I'm stuck, I read back through my notes and it almost immediately helps.
#2 – Link Research Tools. It's what I use for my analysis and my audits. If something looks weird for a site it's the first place I go.
#3 – Screaming Frog. If there's a site issue (and there usually is) you'll find it here most of the time.
#4 – Rex Swain's HTTP Header. I use this religiously to check redirects.
#5 – Majestic SEO. I love the section where you can view new and deleted backlinks and I really love the BackLink History as it's fantastic for comparisons from site to site.
MICHAEL: Narrowing it down to 5 will be tough but here we go (disclosure some of these links are [Michael's] aff links but I would have listed them anyway)
#1 WordPress – I've got a love/hate relationship with wordpress, but despite all of it's many shortcomings and annoyances, it's the fastest, cheapest, most powerful, and flexible way to get something that's within spitting distance of being a CMS.
#2 Raven Tools – It's an extremely powerful suite of tools that help you solve multiple problems (see my Raven tutorials and reviews here). That said I do miss my ranking reports :-(
#3 SEOMoz … err Moz Toolset – Again it's another suite of SEO Tools. I don't think looking at the data from one set of tools is always a good idea 2-3 will give you a much better picture.
#4 Website Auditor – There are lots of programs that will crawl websites and spit back the data, but in my experience this is the most configurable, easiest to use and export the data. When I'm doing a site audit or trying to diagnose a problem that doesn't stick out right away this is a key tool in my toolbox.
#5 Scribe SEO – Let's be honest, sometimes my grammar and spelling isn't the bestest it could be, scribe helps me clean that up. While I do try to give clear instructions to my copyrighters, sometimes they get a little verbose and flowery, and emphasize words I don't want or didn't intend to, scribe helps identify that. It helps me write and edit on topic focussed posts, and since it sits right in wordpress it's super easy to use.
RAE: For the longest time I was very resistant to using tools. Over time, I've come to accept them as valuable helpers. But, the key word there is “helper” – I recently did a rant on Google+ about people being dependent on tools. So with that in mind…
#1 – Raven's SEO Software – it actually does a lot more than “SEO” – but the SEO part is my main usage of it – especially their link manager tool. Their new website auditor tool is great for small business owners who don't live and breathe SEO as well. It also imports your Webmaster Tools data and stores it indefinitely – so you can have a much larger historical picture than the 90 day historical view Google's WMT gives you.
#2 – SEMRush – while their data is included in Raven, I still have a seperate SEMRush account because Raven only serves part of their data. I heart SEMRush. Their organic keyword data is fairly accurate and extremely helpful in competitive keyword research.
#3 – Screaming Frog – awesome crawler for more advanced SEOs. But I'd recommend having the paid version to be able to use it to its full capabilities. Getting a sample of data goes no where near as far as being able to evaluate a full dataset.
#4 – Ontolo – killer link prospecting tool and their competitor backlink metrics are pretty cool as well. The amount of time Ontolo saves you when prospecting is huge.
#5 – Open Site Explorer and Link Detective – I use these as a combo to export the external backlinks to a site from Open Site Explorer and upload them into Link Detective to get an visual overview of the types of inbound links into a site. This is often a help in helping to identify and even out obvious unatural components of backlink profiles.
RAND: #1 – Google Analytics: still the most powerful, fast, and useful web analytics platform I've found.
#2 – Twitter, Facebook, & Google+: just posting content and engaging on these three networks over the past few years has been one of the most powerful strategies to get the content I produce ranking
#3 – Open Site Explorer & Fresh Web Explorer: I know I'm biased, but using these is invaluable for me to discover content, link, PR, and marketing opportunities. Larger web indices like Majestic are super useful when I'm trying to find spam or cleanup bad links, though I don't much of that other than from a research perspective.
#4 – GetListed & Whitespark: for Local, these two are essential
#5 – ScreamingFrog: really good for experts who know how to customize for crawl data
ROGER: That said, tools too often become the strategy. Which is why consultants new to the industry and in-house SEOs gravitate to tools. They give the illusion of empowerment. Most backlink checker sites feature instructions on how to use their checker to spy on competitor backlinks. Poaching competitor backlinks is a mediocre strategy. Why bother promoting a backlink tool that way? It's can barely be called a strategy and is limited in effectiveness. And here's the problem with tools. People reach for tools because they hope they will solve a problem. But the problem is almost always strategy. Tools are not the strategy. Tools serve the strategy. So my answer to the question is, focus on strategy and your selection of tools to serve that strategy are an Internet search away.
TODD: #1 SEObook.com – Competitive Research Tool
#2 SEMRush.com – Keyword Discovery
#3 RavenTools.com – Site Auditor Tool
#4 Moz.com – Keyword Difficulty tool (with advanced report)
#5 ScreamingFrog – Indexation / IA Planning
WILL: In no particular order:
#1 – A unix command line (unless this is cheating!)
#2 – A link index (I'm most comfortable with opensiteexplorer)
#3 – Google Analytics
#4 – Webmaster Central if there is any hint of penalties or crawl issues. Web developer toolbar otherwise
#5 – Google itself – for everything from competitor research to content research to indexation checking
5. Which search or social site, outside of Google do you feel presents the most opportunity for building links and why?
AARON: In many cases this may come down to vertical. For niches like fashion or cooking Pinterest might be best. For niches like technology (and especially things like online rights) then Reddit would likely be great. For start up related stuff hacker news would be a great place to be seen. There are some verticals like video game walkthroughs where YouTube is huge. Of course both Twitter & Facebook are generally quite large.
The big trick with building links is the impact of getting direct & immediate links off of the social networks vs gaining awareness from the networks that leads to greater general awareness which helps your future content spread both within the social networks and across the broader web. In a lot of cases the social stuff will lead to more of the indirect links over time than the immediate links that count.
DEBRA: YouTube. If you’re not using YouTube as a search and discovery venue the same way you use Google to find link partners you’re leaving a lot of link equity on the table.
JULIE: Twitter, definitely. It's the fastest way to find new content which can trigger your own timely responsive content (like a rebuttal piece), help you see what people are talking about so you can jump on that popularity bandwagon, and be the first person to ask to buy/place a link on great new content. It's also easier to quickly interact and build a relationship that can lead to a link…much easier and faster than email in many cases.
MICHAEL: I think it depends on your market, if you are in a techno-centric space you will definitely get the most traction from something like Twitter, Hacker News or even Google+ (shudder). If you are writing about cupcakes and home decorating it's definitely Pinterest or Facebook. Know your audience and follow them to the sites they use.
RAE: Totally depends on the niche. If it's clothing or cooking, Pinterest is going to give you the most bang for your buck – by far. I wouldn't recommend any site who's demographic is 18 or above spend time on Instagram but if your market is tweens – then you'd be well served to get your Insta on.
That said, looking at niches overall and ignoring the “certain social networks are better for certain niches” obvious, Twitter would be my pick. For me personally – as in the owner of the Sugarrae blog, it's my highest exposure driver – and exposure typically results in links.
RAND: For me, it's been Twitter. Although Facebook is larger and Google+ has a more direct impact (especially on personalized results), Twitter's a channel for influencers, site owners, and marketers, which makes it ideal for building relationships and sharing content that leads to links.
ROGER: Facebook. The users are real people who use it to keep in touch. It's a true social AND sharing community. The sharing aspect is important because “sharing” has in my opinion become part of the process for ranking better. So while “sharing” might not count as a traditional hyperlink, it still belongs in this conversation because it's a citation, a new kind of citation. And the algo is a citation based algorithm, so sharing dovetails perfectly with that.
Facebook has the potential for becoming a top recommendation engine. It's current limitation is mobile.
TODD: The site with the most opportunity for distribution is nearly always vertical specific. Within any vertical, there are publishers, manufacturers, aggregators, retailers, and other types of sites.
If you're in tech, TechCrunch and Reddit are high at the top of the list. If you're working in finance, however, these sites may provide little to no value with distribution.
When you're searching for link opportunities ““ focus on knowing your niche, and looking for the distribution platform (site with the most traffic that you can get an article or content piece on) and work to build a relationship with them to work your way up the link authority food-chain.
I think I'll go with twitter for general link building, since it has fundamentally shifted the way a lot of link building is done, and increased the need for secondary credibility indicators, and multiple communication points before successfully achieving a link. That said, you have to be able to build a following on that channel before you can start to build links which is no simple task.
WILL: There are multiple ways I could interpret this question. I'm going to go with “which search or social site is best for reaching an audience of people who might link to you?”.
My answer: it depends.
If I had to pick blind (not knowing the website or industry) I'd go with Twitter.
More specifically, I'd choose the one with the best overlap between influencers and target audience. For many B2B sites, that means LinkedIn. For some consumer verticals, Facebook. For internet marketers, Google+ :)
6. The Link Removal Tool – How do you think it works and what results should users expect?
AARON: I imagine that Google is using it as a data source to flag sites for review. If some of your sites that have been hit share common backlinks with some of your sites that have not yet been hit, you might be begging Google to drop the hammer on those other sites.
DEBRA: If you’re at the point of disavowing links with Google and Bing, evaluate the potential for the site going forward and if it’s not looking good, shut it down. Take your good links and content and start over on a fresh domain or find a used website and rebrand. If you decide to use the tool, read this post on the Naylor blog and take some of the percentages and situations into consideration. The actions of an email account could be a guideline for what you disavow.
ERIC: The concept of a tool that could magically remove all bad links sounds wonderful, but there are parts of the process where any automated tool will break down. The point of breakdown will always be if the tool scrapes sites looking for contact email addresses. Having sent out out a couple hundred thousand link requests over the past 18+ years, I know beyond a shadow of a doubt that email addresses such as webmaster@, help@, info@, contactus@, questions@ are rarely received by the same people who make the editorial decisions about site content. I learned this early on. Those generic email addresses that appear on web sites are also highly likely to be hit with spam. I've done tests where I put an email address on my site just to see how long it takes before I start getting spam. It isn't long. But the larger point here is that if you are sending link takedown requests based on an automated program that is supposed to identify a decision maker, your success rate will be low. You need to take the time to find the actual person who has the power to edit pages.
JIM: The Google Disavow Tool….I could talk about this for hours…that past 2 weeks I've done about 20 disavow documents. I use my own tools (one will be live probably in a few days)….how does it work..well, download your backlinks from google, sort the backlinks, look for patterns, look at everything…remove what you feel got you in trouble…that's the short of it….I'll say there's no push button answer, there is no tool where you can see, “ah, I'll just disavow those”..that is not the solution…everything must be looked at and you have to ask yourself “is that natural?”
What results to expect..well, it comes down to this…When you strip away the SEO links, what do you have left? I've had some people where I've told them “Well, I disavowed the “patterns”….I'm not sure if that will work….because you didn't have 1 natural link at all…if I disavowed the unnatural linked you wouldn't be left with anything at all….so I may give them a disavow first with the “patterns”…if my disavowing of patterns doesn't work, I'd have to disavow everything else on their site…so any recovery would be disavowing everything, and starting over again. When people first ask about recoveries I always say “I'll make a guess at what that may look like after I see what's left after I finish my disavowing”.
JULIE: I assume this is referring to Google's Disavow Tool. I've never used it myself but as far as I know, you list links from sites that you've unsuccessfully tried to remove from a profile, then upload it to Google so they know you've put in a decent effort and that these are links you don't want to “matter” in your profile, as you have no trust in them and don't want them there. I do know people who've used this process and unfortunately it's taken much longer than they expected, as after submitting the file and doing reinclusion requests they still got the message that there were many other unnatural links in the profile. I don't think it's a magic bullet by any stretch but I think a lot of people see it as one and think that if they disavow all their problem links, they no longer have a problem, and I don't believe that.
MICHAEL: I've worked with the tool on two different sites and so far haven't seen it have any effect. Both sites in question were being heavily scraped, between 70-80% of all content, it was basically a joke. After an update last year, we tried to get the links taken down no luck, tried filing a DMCA takedown they were on NON-US hosts so that went no where, so we went the disavow route. Now what really sucks with this tool is the same thing that sucks with a lot of Google tools, you have no idea they got the information you sent them and chose to act on it. If I log into Webmaster Central I would really like to see some indication that those links have been disavowed. However Google's opinion on the matter is … “These aren't the links your looking for … you can go about your business”
Filing a re-inclusion request is similarly a joke, if I've taken the time to cite 3-4 specific examples where Google is giving credit to a scraper, sending back a form letter that doesn't address any of those points, shows the person on the other end either completely ignored my email or is just arrogant and rude. Bad form Google, bad form.
RAE: First off, I resent that I should need to disavow any links at all. If Google knows what they consider crappy links to my site, then just freaking ignore them. Why do I have to go in and tell you “this link shouldn't matter” – if you're penalizing me (let's ignore semantics here) for it then assume I don't want you to count it. But that wouldn't allow Google to get me to tell them about all the artificial links to my site that they CAN'T identify and collect said information en masse.
But I digress. At this moment, I don't jump to disavowing links. Usually, I'm more concerned about ensuring the percentages and naturalness of your backlink profile is on point. Everyone has shitty links – I don't care if you've never done an ounce of link building in your entire life – your site has some shitty links pointed towards it. I don't worry about scrapers or any kind of other naturally accrued shitty links.
The kind of links you need to worry about are the unnaturally, purposefully obtained shitty links. The ones you actually went out and developed (or that your competitors developed for you). But even then, I wouldn't jump to disavowing until I exhausted naturalizing the backlink profile. And even then, I'd probably limit it to paid links and infected sites. But that's me.
RAND: There's a lot of conflicting information about how automated it is or isn't, so I won't speculate there. My general sense is that it can work, but there's a lot of additional hoops to jump through (including the need for having a very comprehensive, truly regretful-sounding reconsideration request). As a user, I wouldn't “expect” anything, but I would certainly use it if you're feeling nervous about some links you know you shouldn't have.
ROGER: The problem with the Link Removal Tool has been one of unrealistic expectations associated with the use of the tool. There seems to be an expectation that previous ranking will return. Whie this happens sometimes, it's not guaranteed and I will explain why. Once unnatural links are removed and others disavowed, the site is going to be ranking where it should be ranking had those links never existed. Often those keyword positions are where the site currently is ranking, regardless if the tool was used or not. If the site has quality links but also acquired unnatural links on top of that, then the chances are higher for that site returning to ranking well again.
TODD: My experience has been limited with the tool, but I think it has really been an admission from Google that there ARE links that can potentially hurt your site. The more devaluation and penalization there is of links, the more this becomes true. These bad links could have been placed by a competitor, contractor, or in-house employee depending on the situation.
The tool is meant to provide aid to those being unfairly targeted by a competitor as it's primary function in my opinion. Reinclusion requests have been around much longer than the tool, and should be included as part of a link disavow if there were problems created by internal SEO mistakes within the company.
I've had a few sites that were filtered by penguin, and when attempting to bring these sites back, I've done my best to change the anchor text, and remove links by hand, or through other online software options like removeem.com or rmooov before resorting to the disavow tool. Another technique I've seen employed successfully is to just 404 pages that may have been overlinked with heavy commercial anchor text (or “overoptimized”).
WILL: You mean disavow? I believe that Google is using it for two big reasons:
#1 – it does seem to factor directly in as an input into the algorithm – but it appears that this happens only on either data refresh or when those links are recrawled rather than when you submit the disavow (this is speculation and best guess and even if it's right, it's hard to tell which of these is true)
#2 – as an input into the machine learning algorithms and possibly even the manual review “dashboard” for reviewing the sites on which those links appear – we've seen little to no direct evidence of this, but I'd be shocked if it wasn't happening.
7. Many online marketers have suggested that we are ‘past' link building & onto social or some such, but when we look at where Google distributes fear in the SEO industry, they mostly seem to be targeting links while they promote spamming social (via ranking spam Facebook Notes & doing things like buying out Frommer's & selling it off for a song while keeping & rebranding the Twitter account). Does the ongoing FUD campaign on links indicate links are still foundation? Is it that links are an area where Google feels more policing is needed? Is Google afraid that if they distribute too much fear on the social front that Google+ will never take off? What gives? Why is the fear so concentrated on links?
AARON: I think part of the reason that Google can't distribute a lot of fear on the social media front is those sites are huge & powerful entities outside of search. Another issue there would be the duality of messaging. How would Google be able to go after large social networks for having some spam on them when Eric Schmidt publicly states the following in interviews:
“We [YouTube] can't review every submission, so basically the crowd marks it if it is a problem post publication. … You have a different model, right. You require human editors.” on Wikileaks vs YouTube
I think links are still quite important & that is indeed reflected in the air time the issue receives. We don't see Google spending a lot of time on meta keywords tags these days.
DEBRA: There is fear because link building is still the cornerstone to online success and Google is making it harder to obtain easy links The question around Google Plus isn't “if” it will take off but when, they've invested too much and integrated the program too heavily to allow it to fail.
ERIC: Google's continued vigilance and public notices against certain types of links speaks pretty loudly to me. And I recall recently listening to an interview where someone from Google said links still have many years of value left. Imagine you are going to build a search engine and money is no object, and you have access to the smartest programmers in the world. You tell them to go create a search engine using whatever signals they feel will produce a satisfactory searcher/user experience today. What can they count? Links, Likes, Plusses, Tweets (which are links anyway). And it's not just about the counting of data points. There's associations (found by links) between entities, there's citations without links. I'm as big a fan of social as anyone. But if the searcher is an 88 year old looking for denture info versus an 28 year old looking for iphone apps versus an 8 year old who is looking for the newest skylander, where would you want to look for signals for each of these. Should intent drive signal selection? Can we every really know intent?
JIM: Google give fear about links, because links are still the key to rankings. Period.
JULIE: I think it's concentrated on links because they are indeed the foundation of the algorithm. Scaring people into not linking or wanting to link is the only thing they can do, in their eyes I suppose. I think it says a lot about them that they are happy to do this though, happy to scare and punish people for being smart enough (or unaware) to outwit their algorithm, and that's what it is…they're being outwitted. They always will be. More policing may be needed but scaring people into not doing something that they've been encouraged to do since the beginning of Google, which is to create sites that users like (and linking is part of that, as good links take you where you want to go), is not the way to go about it.
MICHAEL: I think links are the core of the ranking algorithm, but its not the whole magilla. If you run a site and just concentrate on links you'll fail miserably in the long run. You need to take a holistic approach to growing your website. Is google trying to steer you away from manipulating links … sure, but I think they are also trying to nudge you in the right direction and give you clues to what they want to see and what works.
RAE: Bottom line is you don't spend heavy amounts of time and energy policing things that don't matter. Matt [Cutts, head of Google WebSpam] clearly stated at both SMX West and SMX Advanced that links still matter, they matter a lot and that he saw that being the case for quite a while. What's changing is the type of links that matter, Google's increasing ability to algorithmically validate links and become better at spotting link manipulation.
RAND: I think the focus continues to be on links because links are still a massive part of the ranking systems and will be for a long time to come. Making people afraid to link spam because the penalties are so harsh and the process so Kafkaesque is a smart strategy for Google.
ROGER: Links are still the most important Internet currency. Acquiring links is money. Googles algorithm continues to be citation based, although it is true that Google has been moving away from that in certain ways. This partially relates to user intent, a concern that comes straight from Larry Page. It's this focus on User Intent that has changed the game in many niches, which to a certain extent increased the importance of on-page factors, particularly for short two word queries. This is why for certain niches you might see only a certain kind of site ranking. That's because for those phrases it was determined that a certain kind of page (often non-commercial) serves the user better (than a commercial page). That's one example. Nevertheless, links are still an underlying factor, it's just that with the addition of user intent influences, it can take more than links to fit into what Google wants to display.
TODD: Whoa – how many questions was that? One question – in six parts? Most hardcore marketing types know and understand that links are important to rankings. You probably wouldn't be reading this article if you didn't believe that they are important.
If everyone is out getting the same links, they start to become less important and less of a signal. Google knows how important links are as well, and doesn't want everyone turning into hardcore marketing types sending more unsolicited email link requests.
Corporations tend do what's best for business, as do the individual business units within the corporation. Defending the engine from having low quality results due to high volume of artificial links is good practice for G, and we will no doubt see the anti-link sentiment continue. My advice is don't take your SEO advice from a search engine.
WILL: Tin-foil-hat answer is that Google makes money from links-like products (adwords and adsense) and not (yet) from social advertising.
My real opinion is simply that link spam remains a much larger problem for them at the moment. The link economy (that they largely created) is orders of magnitude larger than the social economy. I think they'd be just as philosophically opposed to all these things, but haven't targeted engineering or PR resource on attacking them yet.
8. Is there such a thing as “safe” links that did not come naturally?
AARON: I think “naturally” is a bit of a loaded term. Almost every large successful online publisher played in the gray area in some ecosystem at some point.
- YouTube was founded with violating copyright as a core principal, as highlighted by one of the co-founders repeatedly uploading copyright content.
- TripAdvisor rented a boatload of links back in the day.
- Path shot up the iTunes rankings by buying boatloads of Spanish language exposure on Facebook.
- etc etc etc
With that being said, I think that Google tends to look at link profiles as a whole. If you come under review for any reason and all your backlinks are from splog networks Google isn't going to care much about your site. If your site has a bunch of media coverage & other links that would generally seem to be hard to manipulate, then Google is less likely to care that there is a little bit of gray stuff in the mix somewhere. After all, just by ranking well a site will automatically pick up some spam links from scraper sites & such.
Matt Cutts made a video a while back that mentioned that the Yahoo! Directory was considered a fine link source even though it is paid, in part due to the editorial review process that goes with it. So the key thing with links is it is best to look like the link was naturally placed by the webmaster, look like it is on topic, and to look like there is some editorial review that goes along with it. Some industry trade organizations & vertical directories might be seen as clean link sources as long as they have editorial integrity.
DEBRA: I worry less about the link being “safe” and more about the page it sits on. All links are good links provided they work, even those using the nofollow attribute may have a measure of merit depending on the site overall. Links using keyword anchors and sitting on authority pages are considered helpful, put the same link on a low quality page and it screams “SEO”. Ding! I try hard to create links that appear to be a natural extension of the page or part of a promotion that fits the content, that way I don't have to worry about looking over my shoulder all the time.
ERIC: Yes, Absolutely. Imagine there is a professor who curates a link list about tsunami information. The Discovery Channel web site puts out new content devoted to tsunamis. I contact this professor to let him know about this new content/URL, and he checks it out. He likes it. He adds a link to it. I cannot see how anyone could call that unnatural. At worst it is speeding a process that could possibly have happened anyway. The question is would that professor ever have seen that new content if I hadn't told him, and this we cannot know. But once he has seen the content because I alerted him to it, does that make his judgment of the quality of the content any decision to include a link to it less natural? To me, no it doesn't.
JIM: Um… maybe only % safe :) …well, I guess a nofollowed link is “safe”.
JULIE: Absolutely, just as there are very dangerous links that did come naturally. How can anyone tell whether a link was given naturally or not, 100% of the time? If a site gets penalized or deindexed but they link to you naturally for whatever reason, that becomes a potentially dangerous link, and a potential problem for the site it links to through no fault of their own. If you know the right person at the right site, you can buy a link that looks completely natural and will probably never get identified as an unnatural link.
MICHAEL: Sure the media can manipulate attention and links, for example a few years ago Stephen Colbert got a huge influx of links for “Greatest Living American” after he asked for them on his show, IMHO that's not natural. While google didn't penalize him for the links they did adjust the algorithm so he no longer ranked for the term.
RAE: In my opinion? Absolutely. I believe that the majority of links accrued while attempting to get in front of your target audience as the primary motive and accrue links as a secondary motive are pretty safe bets. Is that a little “unicorns prancing over rainbows”? Probably. But the most natural link profile you can possibly have is via marketing as if Google didn't exist. It's all about keeping the percentages in check. The more natural links you have, the safer it becomes to obtain the occasional not so natural link.
RAND: It depends on how one defines “naturally.” :-) If the question were, “is there such a thing as links that will never be considered manipulative yet do not come through a trust-deserving, editorial process of some kind, then my answer would be no. I don't think users want results that rank from those kinds of signals and I don't think engines want to reward them, so eventually, they'll die down.
ROGER: What is safe? The algorithm is partly about pattern matching. The search engineers do not want to penalize legitimate activities. So you look at the patterns used by legitimate sites, like most news sites, magazines, university publications, that sort of thing. Good information that is generally published without an SEO goal is the kind of information the search engineers will make sure does not trigger a false positive match. The use of anchor text, the way bylines are structured, that's the kind of thing that makes a link safe.
On the other side of the ranking coin are the kinds of things the search engineers DO want to catch. That means a pattern of inbound links with aggressive anchor text- pretty much all the things that have historically been associated as SEO Friendly are pretty much off the table. SEO Friendly = Penalty Bait.
Right now the safest link is a freely given citation on a page that does not appear to be paid for or sponsored. Conforming to customary non-SEO patterns are going to be safe. Which goes back to creating content for users not search engines, patterns that make sense for site visitors.
Creating great content and telling others about it is perhaps the safest method for attaining links. Here are some ideas:
#A People will share your content if you make it easy to do it on popular social media sites.
#B Engage with others on popular social media sites that are relevant to your niche. For some it's Facebook for others it's LinkedIn. Understand where your site visitors like to hang out and find them there.
#C People will share your content if it cannot be interacted with on a mobile device. Make sure that “touch events” are activated on your site so that tablet users can interact with it. Here is a web page that discusses enabling touch events on your sites: http://touchpunch.furf.com/
#D People will share your content if you advertise and actively promote it using traditional online (and sometimes offline) advertising methods.
TODD: There's such a thing as a safe balance of links. There are constantly examples of artificial links that work, and expecting that G can stay on top of ALL the bad links and devalue them is rather silly, but they have proved very aggressive in their policing of overly-aggressive link building tactics.
I think penguin has finally proved to SEO's that it's a lot easier to do things right the first time rather than try to benefit from the latest link scheme. Rather than chase the old techniques, build some new ones that apply to your business. Jon has more than a few good ideas on his site. If you can't generate links from a few of these things – your link development and SEO isn't doomed – it's quite likely that your business is.
“Safe linking” is a healthy mix of content creation, and outreach via different channels…as for what comes “natural” – I think that's a bit like asking if food is “natural or healthy”. Every link has a cost and a value – that doesn't mean it has to be paid, solicited, or “un-natural”.
WILL: Unfortunately, I'm not certain there is such a thing as a “safe” link even if it did come naturally. The one thing I've consistently always said about manipulation is that it's hard to prove (or even really define) intent and so unfortunately I think that how the link arose is actually irrelevant to its safety. I've seen cases of links that were added by independent third party editors with no incentivisation be called out as problematic in manual reviews.
I do, however, think that there is such a thing as a (relatively) safe link profile. Most of the issues I have seen where innocent links have been highlighted have been as part of link profiles that were anything but squeaky clean. It's actually this varying application of the rules that I find hardest to stomach. I'm a fan of bright lines cleanly enforced. I'd love to see a much more transparent highlighting of the causes of penalties and ranking issues.
9. You have a brand new site. Let's say it's a blog for a specific niche (personally branded or topically branded) to narrow it down a bit. What are the FIRST five things you do to start procuring links – and why?
AARON: #1 – Get a decent domain name. Get a decent logo & get a theme that is color matched to it. People pass judgement quickly, particularly on unknown outsiders, so looking good from day one can save you 6 months or more of slogging it out.
#2 – Write a few strong posts on your own site, along with a decent “about us” page.
#3 – Participate on some of the topical forums & blogs that I would want exposure from.
#4 – Try to interview some of the thought leaders in the space.
#5 – Try to guest post on some of the sites of those thought leaders.
DEBRA: #1 – Create a Twitter account, tweet each post on the blog plus the posts of other authority blogs in the niche to attract attention, open guest blogging opportunities and gain followers.
#2 – Register on a related forum and start posting. Make note of super users, develop working partnerships.
#3 – Turn all posts into podcasts and upload to iTunes. Promote iTunes account to customers, on site and via press release.
#4 – Find all journalists in my industry on Twitter, follow/engage
#5 – Use Facebook ads to send traffic to evergreen content on blog, include email sign up form with content.
ERIC: Not to evade the question, but the steps I'm going to take are going to depend on what specific niche/topic and who the blogger is. Going back to my earlier example, if the blogger is an unknown senior citizen who writes about the issues that face senior citizens, versus a 28 year old well known guru who writes about the most cool new Android apps, I'm going to put together a different strategy for each of those. Both will involve identifying others who will have an interest in the content, and both will involve outreach, but as to the First Five Things, I would say they will be quite different.
JIM: #1 Write a great blog post about my history and credentials and keep writing all the time – why? To help establish trust and to gain friends who enjoy reading what you write.
#2 Start making friends of social networks in my industry – why? These friends will connect with you in the map of the web of people.
#3 Create a tools/widgets/applications for my industry for free – why? To get them coming back and to get them to mention me.
#4 Create a community – why? To get return visitors and to establish trust
#5 Brand my name or company to my community
JULIE: #1 – Set up social share buttons on each post. You'd be surprised at how many people do not do this. The easier it is to socialize the content, the more chances you'll have to get your content out there on the social sites, leading to better chances for links.
#2 – Set up Google authorship for the blog and a G+ account for it (and actually use it.)
#3 – Create a content calendar and recruit other contributors. That will keep you on schedule, keep them on schedule, and give you a bigger pool of people who will promote the site.
#4 – Go for the easy wins like resource pages (ones that aren't full of spam or irrelevant links), Chamber of Commerce and other local sites, social profile links, engine-specific local listings like Google Places for Business, Bing Places for Business, etc.
#5 – Start building relationships on social media. Participate in forums, answer questions on Quora or Twitter, make relevant comments on other related blogs, and just generally get your name out there.
MICHAEL: First get a professional looking design, it doesn't need to be expensive and it doesn't need to be complicated, It just needs to be different enough from a standard template and look professional. Why … NO ONE is going to link to the out of the box default wordpress theme.
Next fill in the core informational, and how to pages, fill in your boiler plate pages like contact/about/privacy etc. It's about making yourself look legit, if your About us page says “Need Content Here” you fail.
Now start by creating some core content that's linkbait. Hottest peppers, most poisonous fish, infamous double agent civil war generals from the area, etc. find topics that are just a step to the side of your core focus, but have a strong viral quality. You need to get people to start recognizing and remembering you brand/domain when they think that niche. These aren't the kind of things you can crank out in an afternoon and cross of your to-do list, it needs to really be good, and will take some time and money to create. No one can guarantee anything will go viral, but if you study what people do like you can increase your chances. Lather Rinse Repeat every 2-6 weeks.
Make some friends. Chat with other noteworthy people in your space, link to/tweet about/facebook share their stuff. If you promote the stuff that's important to them they are like to link to/tweet about/facebook share the stuff that's important to you. Give more than you receive.
Can we talk about guest posts? Listen most of you suck at this you don't have a clue about how to do it. Your pitch goes like this “Hey I see you write a lot about Glass Hull Icebreaker Ships we can create a piece free content for you about this topic, all we ask in return is credit in our author bio”. Thats what I call creating a wrapper for your link. What you want to do is tell me how you are going to create a great piece of content, and show me samples of some of your other articles. YOU WANT TO CREATE VALUE FOR THE PERSON YOU ARE GIVING THE ARTICLE TO. Why … because if it's a great piece and it gets that person links guess what you get … (queue Jeopardy music) … you get a stronger link back to you. Sure it's more work but everybody wins.
RAE: #1 – As much as it may pain you to publish content without an audience, you need to launch with some good content under your belt. You want your first impression to be a site with great content, not a new site who may or may not be about to post great content. You also need to have several pieces of flagship content (really in depth, completely kick ass pieces) ready to go – I typically shoot to create four pre launch – one every two weeks after the site launch.
#2 – Procure a few guest columnists with already established profiles within the industry to create new content for you post launch. Columnists are often happy to promote their work – which puts you square in front of your target audience. Make sure their G+ authorship is also attached to their work on your site. You're likely going to have to pay for said columnists to produce content. Suck it up buttercup. :)
#3 – Find out which social networks will mesh the best with your niche and begin building up profiles on them while your site is still in development. For instance, I'm currently in dev mode on a site, but I've already begun building its social profiles on the two networks best for its niche – Pinterest (16 boards, over 200 pins) and Facebook (and running ads to accrue likes). This way when you launch, you already have some sort of social base and your content / promotion isn't the first thing you're dropping on them.
#4 – Procure a few high quality guest posting opportunities. Not bullshit 300 word articles on a default WordPress theme blog – guest posting opportunities on related sites with actual audiences that can help you drive exposure first, links second.
#5 – Start accruing subscribers to an email list from day one. DAY. ONE. Email marketing is an extremely powerful way to remind people about your site (and make them aware of any amazing content you publish) without them needing to remember to check your site on the regular. Launch with a “subscribe bribe” and be sure to write your emails with the same effort as if you have 10,000 subscribers when you only have 10.
RAND: #1 – Research the influencers and content-creators in the space. Learn what they do well and where there's content opportunities.
#2 – Invest in 1 or 2 “big content” style projects that can help launch the new site/brand – something people in the field clearly want/need and existing resources are non-existant, low quality, or too commercial.
#3 – Start participating heavily in social media around those spheres, particularly on Twitter & Google+, but also through Pinterest/YouTube/Facebook if it's possible and makes sense.
#4 – Once your content is done and you have at least a small audience that includes a few influencers on social and through early email relationships, get feedback (before launching), tweak based on that feedback, then launch and promote.
#5 – Try to go to some real-world, in-person events in your niche. Travel costs are one of the highest-ROI link building methods I've seen (so long as you can be a good people-person at an event).
ROGER: #1 – Content Bait. Create web pages that are useful to a specific kind of site visitor, for which you know there are many sites that would link to it. Build the page then promote it to the sites that will find it to be useful. Here's an example. Suppose you have a site about Real Estate in Podunk, USA. A content bait page featuring a checklist for parents buying a home. On it you discuss possible issues such as asbestos in the ceiling popcorn, lead paint, how many rooms is the bare minimum, the dangers of wood floors, etcetera. Then you promote it to where parents hang out on the web. Another example would be creating a Content Bait page about how to choose housing for senior citizens, considerations for disabled home buyers- then target senior citizen sites or advocates for the disabled.
#2 – Advertise. The why part is because advertising is an effective manner of telling others about your site. StumbleUpon is a no-brainer. Get creative, don't let StumbleUpon be your default.
#3 – Create social media profiles. Facebook can be a source of referrals, a way to build awareness, good will, etcetera.
#4 – Light reciprocal linking. In the blog world it's known as blogrolling. There's nothing wrong with acquiring a few reciprocals from good quality websites.
#5 – Build relationships with important organizations. The why is nothing legitimizes a site and makes it appear important like being seen in the company of important sites.
TODD: #1 – Reach out to existing friends/ webmasters with semi-relevant sites, and ask them to link to your site.
#2 – Create an infographic and submit to a few of the better IG directories (as well as use for future outreach purposes). Rinse and repeat as resources allow.
#3 – Start a facebook page and twitter account to start building relationships with new webmasters for linking opportunities
#4 – Write A LOT ““ every single day if possible. Quality guest posts can get you a long way, but most people aren't willing to put in the time.
#5 – Link out to others often. Build a repository of resources within your niche, and prove the abundance mentality of the site. A great example of this was the extensive blogroll on Search Engine Land when the site was just being started.
WILL: #1 – Backfill some content. I'd want the site to look “lived in” before I started in earnest
#2 – Kick off a bigger-ticket piece of content to help me start getting it on the map – something impressive enough that I could begin cold outreach
#3 – Pick off the low-hanging fruit – I was just doing this the other day for my dad's taxi firm – getting him listed on the websites of hotels that he works closely with, making sure he's listed in tourist information etc.
#4 – Mine my network for overlap with the new industry (e.g. say I was working on something in HR, I'd look to my CEO contacts)
#5 – Start getting to know the influencers in the space – particularly those who control the meta coverage (the equivalent of the Search Engine Land editors in our space)
10. There's been a lot of talk about Google having a brand bias, from an anchor text/link perspective. Do you think this exists and what should website owners do show Google what they are looking for?
AARON: I think if your anchor text profile looks a lot like the branded sites then you are less likely to get flagged for a penalty algorithmically & are more likely to pass manual reviews too.
I should add that my view of the brand bias is a bit more broadly encompassing than just the anchor text profile. In addition to the links aspect, I think brands benefit from:
- folding of usage data (if you have broad awareness then your aggregate usage data with repeat visits is going to be stronger than a lesser known site)
- being less likely to be hit by manual or algorithmic penalties & having shorter recovery periods when hit by manual penalties
DEBRA: I do think there is brand bias when a site is under manual review, what search engine reviewer will recommend a brand page be devalued over a page from keyword-keyword.com? Chances are it won’t happen so when humans are involved, I say yes, there is brand bias.
But I think there is less brand bias when algorithms are involved. Algorithms are programmed to detect certain factors, most brands exceed the thresholds and are insulated as a result. People are searching for them by name (a plus), clicking their results (another plus) and linking to them with trust (ultimate plus). Brands have large staffs to focus on sales, promotions, media relations, social outreach and most importantly – paid advertising. All of those things weigh heavily into the reason why a page ranks and why brands dominate.
ERIC: I don't know that it's a bias as much as it is a natural occurrence that larger brands tend to be discussed more on the web, and then have larger budgets to further encourage even more discussion on the web, in social media, etc. That can look like a bias in the search results. I'm not saying it's fair, but it is what it is. The smaller guy has to try and find alternative tactics. If I decide to launch a cola flavored drink named LinkMoses Cola, I'm never going to outrank Coke or Pepsi, or even Sugarrae Cola, and let's be serious, nobody is creating links using the anchor text “refreshing cola flavored beverage”, see http://goo.gl/879ey although this next one surprised me https://www.google.com/search?q=%22refreshing+cola+flavored+beverage%22&num=100&lr=&safe=images&hl=en&as_qdr=all&filter=0&biw=1508&bih=840 :)
JIM: Yes, there's a big brand bias….you need to become a brand if you're going to survive into the future. It is what it is.
JULIE: I definitely think they have a big brand bias. For one thing, I think webmasters feel safer linking to a big brand than they do linking to one they're unfamiliar with. For another thing, big brands have more money to get more visibility in every possible way which automatically puts them at an advantage for generating more links.
With Google's whole idea of “would you trust this site with your credit card number?” I think it's natural that they're going to favor big brands because people do, too. Big brands also have bigger link profiles and can withstand bad links better than a profile made up of 100 domains only. I think that whatever your offering is online, you should look at the big brands in your niche, see how they're building links, see how they're writing content and promoting it, and figure out what you can do better.
If you're a small brand selling outdoor furniture and you're competing with Amazon, Home Depot, and Lowes, for example, you could take advantage of the fact that those sites may rank for key money phrases, but people still have to dig to find what they want because those sites sell so many other things. With your site, you can show users what they want faster and with fewer clicks, so capitalize on your snippets and make sure that when your result appears, it stands out even if it's further down the page. You'd also be wise to realize that to compete, offering an incentive of some sort is probably necessary. People tend to become very loyal to a brand very quickly especially when savings and great customer service are in place.
Give first time buyers a coupon good for the first TWO purchases, not just the first one, for example. You might not have as many customers as Big Brand X but your reviews and online sentiment might be ten times better.
MICHAEL: Ok yes I do think Google has a brand bias, that said I don't think there is some master white list double encrypted in some secret partition on Matt Cutts hard drive (actually he carries it on self destructing USB jump drive). What I do think is Google has identified signals that brands have that less trustworthy websites don't.
How many people a day do you think go to any search engine and type [Cheap-Hotel-Rooms-Las-Vegas-Strip.com]? How many people type in [Pricline] or [Priceline.com] or [Priceline Hotels Las Vegas] or [Pricline Las Vegas Hotels on Sale]? Do you think people link to Priceline using [Cheap Hotels on Las Vegas Strip] as anchor text or do you think it's [Priceline Hotels] or [Priceline Hotels Las Vegas]? How much information content, research or press worthy content is [Cheap-Hotel-Rooms-Las-Vegas-Strip.com] generating and are they getting links to those pages? Just because you are an SEO doesn't mean you need to act like one … in fact it's probably best you don't.
RAE: I think Google has a brand bias. However, I don't think it matters if the brand is large or small. All true brands tend to exhibit the same signals. The difference – to me – between large brands and small brands is that large brands accrue so many natural links without any effort that they're able to put more effort into purposefully obtaining links – and those links become somewhat camouflaged so to speak by their overall link profile.
However, overall – brands, regardless of size, exhibit many of the same signals. People search for the brand, the brand tends to have way more “brand term” links than it does keyword links, they accure nofollow links, image links, redirected links, co-citation, co-occurence, deep links, the brand is mentioned on related sites (linked or not), brands are typically active on their social networks that their target demographic is, etc. AKA, they're attempting to be more than merely a website that ranks in Google. Do the same.
RAND: I do think there are lots of elements of Google's ranking systems that intentionally reward brands. My opinion isn't that there's specific tactics one should pursue to “imitate” brand signals, but rather we should focus on “becoming a/the brand” in the space. That may mean a lot of investments outside of traditional SEO and link building tactics, but I think that's where the future is going anyway, and over-reliance on Google as a traffic source is a dangerous thing.
ROGER: No, it does not exist. This kind of thinking comes from not understanding the underlying principals for why search algorithms are ranking certain sites. I will explain what is being being missed by those who claim there is a brand bias. The algorithm is looking for PATTERNS indicative of manipulating the algorithm. The traditional recipe for ranking well, particularly with anchor text, has become a fingerprint of unnatural links. In creating the algorithm, the search engineers will want to exclude false positives. So, all the things that high quality sites tend to do naturally will become the fingerprint of natural links, including outbound and inbound link patterns.
TODD: We'll likely continue to see more “brand bias”, which is really just another factor to validate if a company is “real”, and the sentiment associated with that company. G is tired of having crappy small companies rank for big important terms. Rule number one of SEO should be don't make google look bad by having an irrelevant page rank for an important term.
It's important for a site to have people doing navigational brand searches for their company alongside the keywords. “How to build a brand” is definitely beyond the scope of this conversation, but social media mentions, branded searches, and non-targeted brand anchor text are certainly important factors in the current state of search.
WILL: This is definitely one of those areas where I would strongly advise not chasing the algorithm. The power of a real brand is so great and the benefits so wide-reaching outside pure search that I'd far rather focussing on building an actual brand than trying to exhibit the same signals that real brands do.
Aside from the real benefits of building actual brands, I also think that Google's algorithm is in its infancy here and algorithm chasing is likely to lead down blind-alleys over the next 12-24 months.
To specifically answer the question, though, aside from the obvious natural patterns they will expect to see in anchor text, I think the strongest brand signals are branded search and query refinement – e.g. seeing people going from [widget] to [widget example.com] searches. It is those signals that I believe Google will use to set the limits on link metrics they would expect to see for a brand of a given size.
If they need more, I would expect to see them use signals like non-linking citations and brand mentions, indications of existing in the real world and employing people and so forth.
11. Under what specific criteria/circumstances do you feel it's best to advise a client to kill their site completely rather than try to recover?
AARON: I think for most smaller companies it probably makes sense to have a back up site in advance for insurance reasons. That way they potentially can transition more smoothly to promoting the newer site while trying to recover the older site.
I don't ever think it is worth killing a site, particularly if it has non-search awareness (like from old links around the web & an email subscriber list) or if is still doing decent in Bing. The holding cost of a website can be pretty close to zero, so the real issue is figuring out how much resources to pour into the backup site vs the old site. It might in some cases mean that 100% of the resources go into the backup site for a while, but the web changes over time & some penalties might not last forever. Even if you do decide to go away from the old site, it is still worth leaving it up for a bit to consider what forms of cross marketing you can do in terms of trying to get current customers & current linkers to switch some of their links over to another site.
DEBRA: #1 – If the site is new/has no real equity
#2 – Uses a glaring EMD (exact match domain) and you have another URL/site waiting that is better
#3 – All the inbound links use keyword anchors that match the EMD
#4 – Wasn’t listed under the G+ account
ERIC: This is the most painful decision a marketer faces right now. If your site has been penalized, and if you know in your heart your link profile is filled with junk you helped create…If you have received warnings in GWMT, and if you have have attempted to remove all the bad links and disavowed the ones you couldn't get taken down, then a reinclusion request is either going to work or not. In my opinion, if all of the above have taken place, and you have sent two reinclusion requests, and neither worked, it might just be time to kill the site. I've heard some folks build a new site as a replacement even as they try to salvage the old site, just as a precaution. That sounds smart, but be careful with duplicate content issues and keep that new site out of the indexes until you know for sure you are going live with it.
JIM: When I do a disavow analysis and I see that there is nothing natural at all. Then it's either start over with a new url, or disavow all links and start over again with a new strategy.
JULIE: I've never had to advise that for anyone yet, thankfully. If the site was deindexed and the link profile was full of nothing but spam, I'd say kill it and start over. Actually properly cleaning up enough links to get back into Google's index can take you ages and cost a fortune in outreach labor, and you can go through countless rounds of reinclusion requests and still get denied. If you're penalized manually or downgraded through an algorithmic change AND you have a lot of quality links, I'd keep moving ahead with the site, cleaning it up and resubmitting it. You have a chance then, but a deindexing plus a profile built on crap sites doesn't stand much of a chance of recovery from what I've seen.
MICHAEL: There are a lot of factors to consider, site age, profitability, money invested, future earnings potential, offense committed, penance required. IMHO the two most important are future earnings potential and penance required. If your Zune website got spanked in the latest penguin algo update, please recognize this is a blessing in disguise and let it go.
If after 5 years Google has finally busted your 5 way link exchange network of 300 sites cloaking all of their content, and using hacked link injections from WordPress plugins disguised as charity donations for orphans and puppies … just let it go. Sure those examples were extreme, but I want you to really grasp the concept of when you are over the line without taking it personally . It's a lot easier to make the call on a clients site than one of your own. For your websites it's best to get a second opinion from a trusted friend or colleague.
RAE: Advising someone to kill a site is only something I'd do under very extreme circumstances. I.e. if you own buy-cheap-widgets-discount-widget.com and you have nothing but blogspam and paid links pointed at it – then, yeah, I'd say kill out and start over (either via a new domain or disavowing everything if the domain is tied to your brand) – and stop trying to take the easy way out (or “in” to top Google ranks).
For the most part, the typical site owner rarely ends up past the point of no return. Even if I had to disavow every link ever built, ask for reinclusion and start over – if I had a real business and a real brand, I wouldn't be willing to give that up. Nor should any other business owner.
RAND: If your site has been clearly penalized for manipulation/spam in such a way that the vast majority of ranking value is gone, AND the site doesn't have a strong brand presence in the space, I'd give up and start over with a focus on building that brand and earning, rather than “acquiring” links.
ROGER: When the site is burned by excessive unnatural linking patterns that more or less cannot be undone. The part about not being able to undo what was done by the site owner is probably the chief criteria, at least for me. Other scenarios can be bounced back from. Self-inflicted.
TODD: If it takes you longer to clean up your link mess than it does to start a new site and build new ones, it's probably time to call it quits. G really doesn't like repeated offenders, and the likelihood that you'll recover a site multiple times I rather low. Killing off a site is a pretty drastic tactic, and I would certainly be forced to question the rest of the client's tactics before giving such an extreme recommendation.
WILL: I'm very wary and conservative when it comes to this kind of brutal option. Assuming that there was value in the original site (branding investment, some good links etc) I'd probably only ever kill it after starting building a new presence. In other words I'd keep the penalised example.com running while building up exampleindustries.com and only if example.com stayed looking hopeless and exampleindustries.com was gaining traction would I pull the plug.
Disavow ability could become useful in this situation – I'd far rather go as far as a complete reset, disavowing all links, and start over with the same brand than have to start on a new domain. That relies on disavow becoming effective for that purpose – at present, we haven't seen it work that reliably and dependably.
Thanks so much again for everyone taking their time to give amazing input as always. If you liked this article, you can show your thanks by sharing it and following the amazing folks who contributed to it via the social media channels listed at the top of this post. :)