So, I asked a few of the more well known guys on links to do a kind of “group interview” on link development a few months ago and I've been admittedly slow about getting the final result up. But, today is the day, so get ready to learn a bit about link development methods and theories from:
– Eric Ward, the Link Moses behind URL Wire
– Rand Fishkin from SEOMoz
– Roger Montti aka martinibuster, the VP of Verticals at Best of the Web
– Todd Malicoat of Stuntdubl and SEO Class
– and myself, Rae Hoffman of Sugarrae and SEO Class
If you've been under a rock and don't know any of the above names, click through to their sites and grab their feeds now. Seriously… go… now.
What follows is over 6600 words of link development experience and theory packaged into one handy dandy document. I'm sure the guys will be adding their own thoughts on the final result over at their blogs in the coming days and I'll link to them here as they do.
An important thing to remember about this group interview is that no one saw anyone else's answers before writing their own. This isn't about a single answer followed by four head nods. Any agreements come from true beliefs and any contrary opinions came from the same. We're all good at what we do, but it doesn't mean we always agree. ;-) I only saw the other answers as I put this post together and the others will see each other's opinions on the questions that were posed for the first time when this is published.
So, without further ado… ten questions (each interviewee contributed two questions) answered by all five of the above… let's rock and roll…
Question One: What are the factors you would take into account before buying a link from a directly contacted site (not a link broker or seller)?
It is 100% dependent on the outcome I'm expecting that link will cause, and the signal of intent that link may send to the search engines. That's a fancy way of saying I never buy links with the intent of improving search rank, because I just don't believe the risk is worth the reward. I know it's old fashioned, but I don't think it's helpful to make Google's job harder than it already is by seeding a bunch of paid links in nooks and crannies that might be credited as natural.
I think there are numerous factors to take into account and it is hard to make a final decision on a link buy unless you're looking at it. But, as a rule of thumb, I'll take the following into account (in no particular order):
The age of the domain and the site that lives on that domain (i.e. if the domain was registered in 96, but has only been live for a year, in my opinion, it has a different value than one live for ten years), the quality of the site, the placement the link will have (in content vs. in sidebar vs. semi-invisible spots like the footers), the amount of text I can surround it with, the topic of the site (strongly related, somewhat related, I can create a relationship via a certain sub section or not related), the quality of the inbound links to the site/page I would be buying from, the price and the number of visitors it can actually send me. If you pay close attention, you'll notice what I do *not* take into account.
I'd look at a lot of different items, but here are the primary 3:
1. Rank for the target search term(s) at the major engines
2. Relevance to the site/page in question from a human perspective, i.e. would people on that page be likely to click-through and convert?
3. How well do pages at the site generally rank against their competition?
Backlinks. Take a look at the backlinks then note any from the same site. Remove those similar sites and search again. So if you're in Yahoo and you're doing this:
1. linkdomain:linkseller.com -site:linkseller.com
2. And find they have a bunch of inbounds from anotherlinkseller.com, do this:
3. linkdomain:linkseller.com -site:linkseller.com -site:anotherlinkseller.com
Sometimes you'll see the backlink count go down an astounding amount, like ten or thirty thousand backlinks. That smells fishy. I call it link kiting, where several link sellers link to each other to get their Google Toolbar green to a certain level. Also might want to visit some of the sites linking to them to make sure it's not from one entity's network, or from too few sites in their backlinks. So quantity of backlinks from different sites is good to understand, and hopefully you won't easily spot signs of a network or common ownership.
When checking the backlinks, screen out links from .info domains. Sometimes the scraper sites may pump up the backlink count.
Also verify the quality of the content. Cheap bad content is the smell that lets you know something might be bad elsewhere.
That's affirming the negative. You can affirm the positive by restricting your search to the dot edu and dot gov spaces: linkdomain:linkseller.com site:.edu
A site with a significant amount of backlinks from dot edu and dot gov sites must be doing something right?; right?
Testing the value of a link is critical. You really need to know the type of link popularity that is going to be passed to determine a monetary figure to associate with the link. This way you know what kind of deal you're getting. There's a few ways to associate a value with the link. The really easy way – use Rand's page strength tool, or Text Link Ads Link calculator, to determine some type of value associated with what to spend on the link. It's even more important that you understand what type of criteria they are using to come up with those values. I use these 10 tips to determine the value of a link, but generally concern myself with the most important 5 criteria to a link's value.
Question Two: In sectors where links are particularly tough to come by (retail, B2B in boring industries, etc.), what are content tactics you've taken to appeal to relevant sites in those industries (specific examples would rock)?
I was working with an online retailer that specialized in veterinary supplies and medicines. I asked if they did much business with vet schools, city or state governments and turns out they did tons. They'd never created content designed for those particular audiences. Aside from the high trust links such content would engender, it's just good business to target such audiences. The potential link equity is a bonus.
An example I can give is that I have a specific site in a finance industry that it on a very boring topic (think taxes, all the boring legal financial stuff) that I've take a few different approaches with:
One is that we didn't have a forum, and I didn't want one, so I approached a larger site that did. I offered to integrate them into my site as the “forum” link in my navigation to the specific sub forum that was relevant to us. In return (this site was a big board with *huge* traffic numbers) they gave us a sticky at the top of the forum welcoming our members and linking to some sections of the site. We get a lot of traffic from them, they had increased membership from us and it really is an example of a new age, relevant and non search engine reciprocal link relationship.
One of the other things we did on this site was to turn some of the government forms into printable calculators. People can play with their numbers over and over until they're happy, print out the sheet and copy it into their legal forms… a lot of forums and sites link to these calculators.
I'll give you another example of a one way link a lawyer managed to get from us via this same site. He answers two questions a month submitted by readers. We get great content that actually comes from a lawyer and to advertise (and stand out) with our “ask a lawyer” section and he gets links and targeted traffic from his bio that is listed on each page for every month he does them. He gets a one way link and it takes him a few minutes to do each month. We offered him the partnership and thus far, it has been working out well for both of us.
I've seen a couple that I really liked, including:
Cosmetics – http://www.campaignforrealbeauty.com/home_films_evolution_v2.swf (though it wasn't well executed)
Wine – http://corkd.com/ – appeals to wine geeks and regular geeks ;-)
Most sites that are in “boring industries” have organizations and associations that support the “boring industries” professional. There are often even local chapters of “boring industries” professional organizations. Google talks about expert directories but these are expert sites, too.
You can also do the corporate blog. Just have to approach it with the understanding that you are reaching out to an existing community of “boring industries” professionals. Best of the Web's Blog directory at Blogs.BOTW.org is not only a good place to list your blog (if it's good enough), it's also a great place to start your search for other “boring industries” blogs to reach out in the comments section, as well as linking to from your blog roll. Check out who the best blogs are linking to and add those to your feed reader and be sure to regularly comment on those blogs you want attention from. Remember, you are creating community.
Lastly there is video. Video content is becoming more and more important. Some boring industries can pick up buzz or links from incorporating video into what they do. Make it useful. Like blogging, don't do it for it's own sake, make it useful, make it with a community in mind. Of course you can even combine the two into videoblogging and become the “boring industries” most dynamic site for information all about your product.
My favorite answer to this question is a reference to Michael's "Getting Your Non-Tech Site Into Digg". Which is an fantastic and timeless post. I've seen half a dozen people ask Michael this question, and he always has amazing answers, because it's the way he thinks. If you're really lazy, just hire him to come up with an idea for you – we work in an idea economy now. I always remember a quote from Mike Grehan that he says, "asking a site for a link is like asking to do business", and it's 100% true. You need to think like the person you are requesting to do business with.
A specific example would be if I had an site about accounting (I don't) – even accountants have a sense of humor to appeal to. Rather than writing a tutorial on how to file a 1099 – why not write the 23 ways most people cheat on their taxes, or the top 12 worst audits of all time. In boring industries you have to think twice as hard, and put in ten times the effort on creating interesting compelling content. The hardest sell on this stuff is the championing of a project like this internally because it takes some real stones to explain an "off the wall" concept like this. Think about who might link to you and create something compelling for them. Your users will most likely enjoy it too.
Question Three: What is the current climate with paid links, and how can I navigate the paid links landscape so that I come out smelling like a rose on the other side?
Don't do it. Or if you do, make sure your intent is to buy links for direct click traffic, i.e., advertising, rather than for any potential ranking boost. And don't kid yourself. You know your intent anytime you seek a link.
My stance on paid links is pretty simple and anyone who has ever seen me do a site clinic at a PubCon has likely heard me say this before: If I can tell that you're links are paid within five minutes of doing a “link domain” in Yahoo, you're link buying strategy needs to be revamped. If you want to smell like a rose, take the time to create a link buying campaign that either appears stealth unless under very close scrutiny or one that is aimed to bring you traffic and therefore looks more like an advertising campaign. For the latter, throwing a few redirect buys and links that aren't direct in there might help to make your overall appearance (and intent) seem more legitimate.
I think that if you stick to personal link sales, i.e. those where the relationship between buyer and seller is direct, you can do very well with paid links. The problem is that link buyers expect it will be quicker and easier to buy links than to acquire them through non-monetized systems. In many cases, though, there's an equal amount of work involved. I wrote about this a bit more here.
I suspect Google learned from the SearchKing debacle that it's better to quietly cut a site's ability to pass PageRank and leave the toolbar green alone than to go out and algorithmically remove sites from Google's index that are selling links. Cutting or dampening the ability to pass PageRank will not affect a website's ability to sell advertising to advertisers who simply want exposure to a demographic. It also accomplishes the goal of preventing paid links from skewing their algorithmic scoring. Paid links still work, but the word is that it's working less and less in terms of helping ranking.
Google's approach is to attack this algorithmically so the wiggle room in my view is approaching site owners selling advertising and negotiating a deal for advertising and sponsorship. Monetizing a site can be tough, so many site owners are open to selling some advertising to supplement their income.
The paid link debate is pretty frustrating to me – the easiest answer is only buy links from relevant places. Every link has an inherent monetary value. Understanding the monetary value of any given link is the basic fundamental of most internet marketing. The best thing you can do when buying links is to stay on topic.
Question Four: Automating the link acquisition process: How far do you recommend going with it? What are the benefits and pitfalls of automation?
The only parts of the link building process that I will employ automation for are those that do not involve human contact. I'll do link audits using proprietary scripts, but once I have a database of target sites I do the rest myself. Target site evaluation, contact info, making the requests – I always do those personally.
I am an old timer in this regard. I believe in doing as much by hand as you can. How do you send out hundreds of requests by hand? Well, I don't. I target my link requests. Media intros (and old tip from MB he said over at webmasterworld.com several years ago), strategic reciprocal alliances (like the one I mentioned in the finance sector above) and one way requests all have a better success rate for me when done by hand, on a personal level, with someone very well spoken and intelligent representing your site in the equation.
I have no problem locating potential links automatically or researching the competitions backlinks via automation (a mechnical grunt is a good thing) – it is the contacting of those possible link sources that needs to be natural. Benefits of automation is obviously speed, with the downfalls being noticeable footprints – mainly for those who use automated sources to display those links as well as those who use the same automated technology and therefore have the same “paths” as many other sites using the same programs in their niche.
If you're looking for a churn and burn, there are aggressive link farms that usually work for a limited amount of time before dying out. They're harder to come by at Google, but you can certainly rank well at MSN & Yahoo! with them in a short period of time. If you're a big brand or if you want to grow into even a small brand with some permanence, my advice would be to avoid link spam acquisition entirely. Go the organic route, build up a true industry and authoritative presence, get links naturally and you'll be far better off in the long run. It's not worth the short-term loss for anything that's going to carry permanence.
I prefer to do things by hand as much as possible. There are too many nuances that can't be automated. If my team in India has a hard time sorting the good from the bad, I don't think an automated process will be able to screen it much better. I can see it doing some processes well, like doing a search and returning only the sites that contain an email link or a phrase like suggest a link. But it's still going to leave many legitimate link prospects behind. Everything has it's upside and downside so you just have to choose which downsides you're willing to live with.
For instance, email blasting software is cool but one screw up in the database and you've just emailed the wrong personalization information for each of the prospects in your list, lol.
There really aren't that many shortcuts if you're looking for quality links outside of SEO neighborhoods and link buddy networks. But for every choice you make whether automated or not, it's going to have it's downside and upside.
Link "discovery" can certainly be automated as much as possible. An example would be crafting creative queries. You can find sites with a high likelihood for success – but automated link request e-mails will almost never garner success without substantial detriment to the site's branding. Automated link request e-mails are not a good idea, but improving the process of discovering sites that may have a high conversion rate is one of the things that can make this painful process a little bit easier.
Question Five: Name 5 specific techniques (queries, tools, process, or starting points) you would use to get links for a statewide real estate agent.
Since the overwhelming majority of real estate agent sites are, well, nothing more than online billboards with little content, or gateways to MLS databases, the linking potential is very similar for most of them, and the link quality potential is low. Yes, there are some to be had, but it depends on the state, the agent's focus, and the agent's site. For example, this site http://www.lakehouse.com/ which focuses on lake property will attract links that this site http://www.century21thewrightchoice.net/ never could. Have a look.
Death to whoever came up with this question (kidding). Well, again, without seeing the specific site, it is hard to make very specific suggestions, but the top five that pop into my head are:
1. The first starting point is making sure he site is ship shape from an SEO perspective and that she has general content worth linking to.
2. Creation of specific content meant to be interesting and attention grabbingâ€¦ something like “Top Eight State Name Communities You Didn't Know You Wanted to Live In Until Now”.
3. Creating a portal for specific housing developments within cities and towns is another smart tactic. If your area has a gated community, consider “sponsoring” a site that gives information about it. Not only will you get inbound links from the community site, but since these sites are often the only ones focusing on a community within a town, they rank fairly easily and can send you a lot of targeted leads who see you as “the” real estate agent for the community. Scrapers and made for Adsense site builders can scrape city names, but information or even knowledge of “Sugar Woods”, a gated community within a town, is a little harder for people who aren't local to come by – again, making competition for those niches within a town almost non-existent.
4. Creating a local presence on national sites is also an important step. The more obvious: making sure they're listed in the regional sections of directories like Dmoz, Yahoo, the BOTW web directory, etc. Also making sure they have any and all available listings within the local search arena like those listed in my Local Small Business SEO Guide.
5. I'd also start to backtrack their competition by using the (linkdomain:competitor1.com linkdomain:competitor2.com) -linkdomain:myclient.com to see who is linking to several other local agents, without linking to the site I am working with and figure out how we can get them those links.
Google maps mashups – they're all the rage and if you can put a good one together, you'll be getting links from all over with the perfect relevance and anchor text that you want. Some other good ones include the very basic but often overlooked design and user experience element. If I go to your site and the experience of surfing it isn't at least as enjoyable and useful as a site like Zillow.com in real estate or Wufoo.com in forms or Digg.com in user-submitted news, you have an opportunity to improve. A site that looks stunning and performs with an almost sexy aesthetic is a great way to insure that you're maximizing the number of new inbound links and viral spread per visit. Sorry I didn't do all five.
1. Video Walkthroughs. Host them on your site.
2. Upload Video Walkthroughs to YouTube and Google Video
3. Send out press releases and pre-written articles about your Video Walkthroughs directly to the real estate editors at newspapers across your country (and maybe even the world). Then send out press releases the to technology editors and the business editors. Do it in phases, spread it out.
4. Start a how-to blog or website that is useful to first time or even aspiring home buyers. Give them the tips, be sure to brand yourself as the Real Estate Agent you can trust. Everyone wants a superstar Real Estate Agent on their side. Be a superstar, don't just say it! Then get links to the site or blog. How-to sites, especially ones without an obvious commercial component (like AdSense or banners) are the easiest for attracting links to.
5. Chambers of commerce memberships
1. Google directory
2. Search query combination tool – state or city names + add url (or better more creative queries)
3. Yahoo site explorer – search "state + real estate" (the top phrase) – and mine the backlinks of the top 3 sites ranking
4. Search geo-targeted areas for local opportunities "cityname + sponsors"
5. Install Hubfinder on your own site (so you don't run out of api calls), or using some of Aaron's other tools
Question Six: What are the top 3 over-rated and under-rated criteria
for determining how valuable an individual link will potentially be to
your site's search engine rankings?
Over-rated: .edu based IBLs, keyword laden IBL anchor text, pagerank. Under-rated: Your own site's previously earned trust.
Overrated: Google toolbar pagerank, direct links being the only type of link with value, a link being on a homepage. Underrated: Redirects, traffic value of the link, topic on the page level of the page linking to you.
Over-rated – PageRank (of both the domain and the specific page), Link location in the site (a lot of folks worry if it's not on the homepage, it won't mean as much, but I like deeper links; they're more natural), and number of outbound links on the page (unless it's significantly more than 2-3 dozen, I don't think it's a big issue unless your anchor text is the same as everyone else's). Under-rated – Ranking at the search engines for the target term(s), ranking at the search engines for competitive terms and phrases in the title tag and quality/relevance of click-through traffic.
Overrated: PageRank, especially a minimum of PageRank 4 (Even though people acknowledge the invalidity of using PR as a measure they will, often within the same breath, acknowledge they don't know of any other way to judge for quality), Anchor Text (Believe it or not, some webmasters will actually walk away from a link simply because it's not a text link. In their minds, banners are not links), Digg. Underrated: Blogs, Videos, Digg.
Over-rated: Toolbar Pagerank, surrounding text, Toolbar Pagerank. Under-rated: Site trust, relevance, and anchor text that's used, which page it's placed on within the site (see the strongest page tool) and outbound links on a page.
Question Seven: If you were in control of the search engine algorithm at Google, what are the top three changes you would make specific to how Google values/counts links?
It pains me to say this, because I'm a 7 year DMOZ editor, but if Google is giving algorithmic credit for DMOZ links, it's time to stop. A best in category content site launching today will never get in, because orphaned categories are everywhere.
The 301 redirect is abused by people buying old sites and funneling link juice. Do something.
Quit nursing the no-name directory industry.
Wow, nice and tough question. As much as we bitch about the way they do things, they're the only ones who have had trial and error at the volume to know why they do or don't implement certain things. But, I think they could still stand to see some improvement in some areas:
First: Paid links should be treated the same as any other links. That is simply the way the world works. Some people buy their marketing, some create it via guerilla methods (like developing organic links) – but it all boils down to engineered results from marketing efforts. Spend more time learn to separate crap links from links from good sites instead of paid from organic. And I say this as someone who does not pay for more than 5-10% of the links across my entire legion of sites, and the number would be a lot lower if I didn't consider a directory a paid link.
Second: A lot of old sites ride on scraper links that they get as a result of being/having been number one on Google for several years. If an old domain isn't getting new trusted or quality links occasionally, they shouldn't be able to hold that number one ranking based primarily on age of the site and age of the inbound trusted links.
And lastly, for the love of god, turn down the authority knob on Wikipedia, about.com, Technorati and other “top” domains, please. For both their own rankings and their ability to pass value to other links.
[rant] While we're at it, here is an unrelated shout out to about.com: As the world's original made for Adsense site, please redesign that beast and make it functional. You can have the best information in the world, but with your clunky layout and *click on me* in your face ads, your usability goes to hell, users get frustrated and you end up surviving on volume from an authority knob turned way to high instead of value. [end rant]
Wikipedia is far too prevalent, even with some very low quality pages on their site. I would almost certainly look into reducing that site's strangehold over so many SERPs.
My second change would be to make a careful review of most major media portals' – TV, radio and print – online presences. Many tend to mention websites without linking out (link data that search engines would love to have) and will often have paid links or network links on the sidebars, which can really throw off accuracy. If possible, I'd consider utilizing something where URL mentions in text can actually be counted as links in some instances where it's valuable.
Third – This isn't specifically related to link analysis, but in the Webmaster console, I might actually consider pointing out links or pages of “low or suspicious quality.” This would be a great way to wake up webmasters to their site's link-building or content-building efforts that Google doesn't particularly like. Providing feedback like this might save thousands of questions and help to make the web a better place overall :-)
Less emphasis on anchor text.
Introduce a filter to differentiate between local type sites and score sites better not according to quantity nor quality, but the KIND of links they're getting, so that it indeed fits the profile of the typical Joe Locksmith who puts up a website, instead of massive nationwide service industries.
Introduce several thresholds of statistical probability that a site is manipulating links by forum spamming. This doesn't mean penalizing a site for forum spamming, but it should take the form of automatically not counting links from any forum if it's determined the links fit the profile of spam.
1. Work on classifying every type of link in a semantic ontology
2. Work on tracking user behavior with regards to link
3. Work on interpreting user behavior with regards to links
Question Eight: What's the most common mistake you see people making in their link building activities?
I'll paraphrase what I wrote here. People ignore the fundamental, logical, natural and unique inbound link potential that their site has that no other site does. There are links that any web site can get, and there are links that your web site can get that other sites will never get. Somewhere in-between lays the perfect inbound link profile for your site, for both algorithms and people.
Paying attention to the Google toolbar pagerank numbers for a site tied with clinging to the link development tactics that worked in 2002 (“but, but, but my competitors are ranking with nothing but reciprocals with their 8 year old site!”). Using TBPR to determine value and reciprocal linking as a *sole* method of marketing is dead (I said sole method; I swear, somewhere there will be a quote tomorrow that I said it was dead in general). Let it go. Please? Seriously, please. Let it go.
Buying on link networks that are way too obvious. If you've been in the SEO game for a year or two, you start to know what the various paid link ad networks look like and when clients complain that they can't afford SEO services but are spending $10K/month on paid links that only help them out at MSN, it's a very frustrating experience.
Focusing on PageRank instead of better quality metrics. The other is not thinking of enough free one-way inbound opportunities and ways to make one's site accommodating for that.
Several to list:
1. Focusing on quantity over quality – or getting too many directory links
2. Not understanding the inherent value of a link – and how to identify it
3. Not searching effectively
4. Not building relationships and starting conversations within their respective online community
5. Not prioritizing link development in their marketing efforts
Question Nine: How do you think that nofollow links and redirected links are treated by the various search engines and do you see any value in obtaining these types of links?
Regardless of how the engines treat the nofollow attribute or redirects, you still need to evaluate each potential link target for its potential to help your business. I pursue nofollow links every day. Is a wikipedia link any less valuable just because its links are nofollowed now? Algorithmically, perhaps, but who cares? The topical link in front of an interested audience is the primary benefit anyway.
Warning, I am about to be very long winded. ;-) I think that each engine probably treats both nofollows and redirects in their own way. The engines never said they would completely ignore these types of links. The nofollow link was used to identify that you as a webmaster were not vouching for the link, but never did *all* of the engines state unequivocally that they would completely ignore nofollow links, but take that how you will. I think that fact that nofollow links show as backlinks in Yahoo illustrates that point. How each engine treats those links is up to them.
For me as a webmaster and an SEO, I view a good link as a good link, regardless of whether or not it has a nofollow attached to it. Sure, I'd like to get one without a nofollow attached to it, but if CNN offers you a link, are you going to turn it down if it has a nofollow attached to it? Certainly not. As far as redirects go, I generally have the same opinion. They show in Yahoo's backlink searches too and I think engines do see and evaluate these links. How much value they see in them remains the knowledge of the engine.
But I'll take a redirect from Yahoo Tech redirected or not, because bottom line is that it brings traffic. I also pay for several ads on sites that are redirected for the same reason. The more people that come across your site, the more of a chance someone will stumble upon it, find it worthy and you'll get even more links (direct, nofollow and redirected) as a result.
Nofollow doesn't appear, to me, to be passing any anchor text or ranking weight value at any of the engines. The only cases I've seen it have any effect is with test keywords that don't exist anywhere else, and to my mind, those tests don't, in this case, provide proof positive that nofollow links actually can help with rankings. The value in obtaining these links, however, can be tremendous in terms of actual visitor traffic, but you can't get it by automated systems and spamming, it has to be truly valuable links that people want to click.
Yahoo and MSN still seem to be counting them so it follows that if they still work pick them up- just don't rely on them as the cornerstone of your link building strategy. Redirected links, if there's traffic then fine. But there's still the lingering fear of the redirect turning into a hijack.
I haven't studied nofollow at any scientific level – though my gut feel is it's away to track publisher intention to some extent. The worst case scenario is that nofollow links don't pass "link juice", but may send some direct traffic. The important thing to remember is not to confuse the two distinct goals of 1. getting links for higher search rankings and 2. obtaining links for direct click through value. For the most part, the two goals should remain mutually exclusive.
Question Ten: So many people complain about competitors who are ranking solely on reciprocal links. Do you believe reciprocal links still works as a complete strategy or do you believe age, trust and grandfathering (for lack of a better word) are taken into account? How important do you feel age is to a link?
To believe that reciprocal links are a dead strategy is silly. Likewise, to believe that you can recip your way to position 1 is nuts. The reality is that many sites rank #1 with zero recips, while others rank #1 with pages and pages of them. There are many other factors that go into it. No one factor will apply to all sites. A site that has had a page full of recips since 1996 obviously never intended those recips to influence search results, since engines back then didn't analyze links. On the other hand, a site that has been around since 1996 but didn't have a reciprocal links page until this month may still benefit the recips, due to previously earned trust. Or, to complicate matters more, if that same site that's earned a level of trust is found to be engaging in link buying or link farming, then their good reputation might go down the drain, and their site down the rankings. It's really a Gestalt kind of thing. It's about the sum of the parts, the signals of intent, the complete history of site, traffic, linking patterns over time, and on and on. Every site comes with its own unique linking history, and what works for one site cannot work for all sites.
Sigh. I love moderating at webmasterworld.com, but this one question is one I see over and over and sometimes you just want to grab people and shake them: “It is not 2002 anymore!” and hope they finally hear you. Age, trust, grandfathering – I think they all play a part. Reciprocal linking is not dead, but it's ability to be a sole marketing vehicle that will build *ranks to last* in my opinion, is. I also think any site, regardless of age, grandfathering or trust is a ticking time bomb if they have a 40 page reciprocal link directory that links out to anything and everything under the sun. I also think the age of a link matters quite a bit – especially in Google and Yahoo.
The bottom line in my opinion is that link development has become the ancestor to traffic development. This is a new age in link development. We will still have standards and low lying fruit links that have value in obtaining them, and we still have a need for experts who can think of angles and help execute marketing plans to gain links. But, the real deal is that it is time to become a marketer instead of a “simply” a link developer or don't let the door hit you in the ass on the way out.
Reciprocal links as an actual strategy (i.e. you link to me on your page of 400 links and I'll link to you on mine) is very low, but links from domains that point to each other are not necessarily losing much value at all. I'd say with SEOmoz, a full 20-30% of the sites that have ever linked to us are sites we've linked to (and that percentage may be closer to 50), but I don't think those links have lost value simply because we later pointed to their site or page on the blog or in an article. Age of a link – not nearly as important as age and trust of a site. A year-old link from a year-old site (all other things being equal) is, in my opinion, far less valuable than a day-old link from a 10-year-old site.
I think this is discussing two activities with different goals that intersect at one point. The first one, about reciprocals, is about invalidating a link and gets into patterns and statistical probability that a site is engaging in an activity to manipulate their links.
The other one is about validating a link. Whether age actually is playing a role might be hard to say because often you've already moved on and other links may be playing a role with the rankings. Nevertheless, supposing you haven't picked up links apart from your reciprocals, I think there are enough measurements for invalidating links that may rule out any pluses from the link-validating controls.
Just have to remind that this is isolating one measure out of many other measures, and we're really simplifying things to get at just one piece of the puzzle, one piece that is affected by others. There are plenty of other variables in play outside of links that can add more in your favor to validate your site as a whole.
I think the answer was given in the question on this one. Ranking mainly on reciprocal links is not something you're going to manage to pull off with a new site. I would not suggest anyone actively involving their site in a reciprocal link campaign – and good SEO's have been saying this for years. The main reason older sites maintain their rankings is due to the age of the links. The age of those links makes them somewhat trusted, despite the fact that they are reciprocals. My opinion is that age of links is a very important criteria to the value of a link.
I would imagine that trying to clear out all the people who ranked well in the past due to recips would likely cause a lot of collateral damage in the serps. I think the search engine's concern is that reciprocal linking does not continue to be an effective technique. With that goal, I think they have succeeded. For those that are against folks with a lot of reciprocal links that are ranking well – you will need to find a new technique to compete with them. There are plenty of techniques still available for those who are not lazy, and can get a bit creative – I would suggest linkbaiting.
Edited to add: We had a request for target jumps. Ask and ye shall receive: