SEO is often cited as a reason for pushing press releases over the wires. The thinking goes like this: since Google thinks of a link as a “vote” for your site, widespread wire distribution “gets out the vote” by generating backlinks from as many different domains as possible. Google then counts those votes and you win in a landslide. Could it really be that simple?
Unfortunately, no. Just as not all editorial coverage is of high quality, not all links - from press releases or other sources - are high quality. Google knows this, and it affects you where SEO matters most - on the SERP (Search Engine Results Page).
Links are a Meritocracy, not a Democracy
The “a link is like a vote” analogy is fatally flawed. Google no longer simply tallies up the ballots from all the precincts and declares a winner based on a simple count. Google’s algorithm is always changing, of course, but it consistently focuses on reputation, quality of content, and quality of engagement.
In that regard, it’s less like counting votes in an election, and more like deciding which friends to trust with a restaurant recommendation. We all have some friends who we trust to guide us to a once-in-a-lifetime dining nirvana. We also have a few well-intentioned friends who will steer us astray and send us to a regrettably overpriced and underwhelming tourist trap. Google is now in the business of helping us know the difference.
And the biggest single differentiator? Experience. Has your friend actually been to the restaurant, or are they just repeating hearsay? In the same way, Google pays attention, not to what you say by linking, but to what visitors do with those links. If visitors follow the links and engage with the content, Google gives it a higher weight. If the links are never seen or followed, it won’t help you – and, as you’ll see in the toxic links section below, it could actually work against you.
Why SEOs Are Asking the Wrong Question
SEOs, wire services, and others have gone back and forth in recent years on the question “do press releases help SEO or not.” Matt Cutts, Google’s pre-eminent search expert, has often said that they don’t. In fact, he views them as paid promotion and recommends treating them in the same way. To avoid penalty, Cutts suggests, treat press release links as “nofollow” links - explicitly instruct search engine robots to ignore them.
Some SEO experts have countered that, based on their analysis, press release links sometimes do boost SEO results. Even if some tests show positive results, Matt Cutts’ statements are clear, and it’s hard to argue that you can beat Google in the long term. “Did a test show a positive result, contrary to Google’s claim” is, at best, a short-term question that has little to with whether those links are actually helping visitors you care about discover content that they’re looking for. Moreover, those tests aren’t conducted in a vacuum – press release activity is usually coordinated with other promotions and activity that may, directly or indirectly, boost search results.
The right question, given Google’s meritocracy-of-links focus, is “are these links creating trust and engagement among searchers.” A legitimate link from press release on a media site that web visitors can find and follow may do so. An awful lot of press release backlinks, unfortunately, don’t meet that criterion.
Press Release Links: An Example
We recently analyzed a set of 196 links from a press release that was issued by a client. Of those 196,
- 41 links were non-searchable content buried on static pages. A link that can’t be found or followed is of even less consequence than a tree that falls in the forest. The Lorax cares about every tree that falls, but there is no Lorax obsessing over unclicked, unseen links.
- 2 links led to 404 errors (page not found). Google views a reasonable number of 404 errors as a natural part of the ever-changing Web. However, if a page returns a 404, hours or days, after the press release with the link was posted, it was probably never a legitimate link to begin with.
- 27 were essentially duplicate domains. As examples, there were five entries from businessstandard.com, three entries on indiainfo.com, and many similar ones. Most of them were low-value domains.
- 7 links were duplicate content – the same domain and almost the same url.
That cuts the number of legitimate backlinks by almost half. Of the remainder, the majority were on sites you’d be unlikely to seek out and visit (investorplace.com, insurancethoughtleadership.com, moneycontrol.com), obscure subdomains (economictimes.indiatimes.com, softwaredev.itbusinessnet.com), or Press Release sections of news sites – many of which were local markets unrelated to the markets our client was targeting. All in all, if the majority these links were recommending a restaurant to you, you’d probably want to check Yelp or Eater instead.
Are Some Links Toxic?
So, is there anything wrong with low-value links that visitors don’t find or follow? Often not – as with 404 errors, Google considers them part of the ever-changing nature of the Web. However, a term that you may be hearing more often is “toxic link.” Search Engine Land notes that “toxic links are easy to identify” and are characterized by being from sites that lack audience or engagement, devoid of human value (designed only for Web crawlers) pointing to potentially irrevelant industries or geographies, etc.
Although that sounds like some of the links in our analysis, that doesn’t mean most press release-generated links are toxic. Some may be, though. At times, Google judges sites like we judge people – by the company they keep. When a link is sufficiently low-value, deceptive, or from a publication without audience or engagement, Google may assess a penalty. To avoid that, you’ll want to disavow that link, just as you’d disavow a disreputable person by saying “he’s not really with me.” Disavowing a toxic link is relatively simple – here’s information on how to identify and disavow them.
Why We Count Hits (and Links) the Way We Do
With all that said, there are still reasons that many organizations issue press releases as a part of (rather than substitute for) a PR strategy. The analytics provided by a wire service differ from our PR analytics because we count different things. A wire service will count every possible link generated, regardless of whether it has audience, influence, or even toxicity. TrendKite, on the other hand, is built on the premise of measuring business impact. If a “placement” is simply a link on a hidden page that no one reads, the impact is zero – even if that page is hidden on a high-authority domain like WSJ.com. Therefore, our counts will be different, because we are counting different things.
Summing it Up
You will hear many different perspectives on this topic – some driven by an agenda, some driven by opinion, some driven by data. However, I don’t think anyone will differ with the statement that Google evaluates links and content based on how useful they are to visitors, as measured by audience and engagement. Consider that as you decide how to most effectively use PR to drive SEO.