How Google Might Turn Down The Knob On Exact Match Domains

by on December 13, 2010 | posted in SEO Theory

There’s recently been a lot of pubsubhubbub about Google’s overuse of exact match domains in the SERPs. They rank too highly, they say. They’re spam riddled pumpernickel, they say. They’re SPAM (sites placed above mine) period, they say. And mostly, they’re right. There’s more to the debate, of course, but even if a little extra oomph should be applied to these domains, it might be universally agreed that, at best (or worst, depending on what webmaster you ask), they pass a little too much weight. So, Google and Co. need to do something about it.

Definitely, the talk has been loud and impactful, but there hasn’t been much in the way of hypothesis as to how Google might go about devaluing the domains. It’s harder to talk about than it is to implement, of course. Is solving the exact match problem the same kind of feat that solving world peace or world hunger would be? Probably not, but the point remains – the job we have, as SEOs, webmasters and concerned parties – is doing some strict analysis to see where this thing is going, how Google might adapt, and adapt back.

So, I bring you – my hypothesis on how Google might be – could – or is – solving the exact match domain problem.

Sophisticated vs. Unsophisticated Linking

Webmasters take on two identities – sophisticated linkers, and unsophisticated linkers. Sophisticated linkers are aware that an exact match domain such as also doubles as an extremely competitive, commercial text, so they don’t want to link to it in a way that makes it appear as just that – commercial anchor text – which is a sight for sore eyes to their readers. Their equal consideration is whether or not this choice might be confusing to those who can’t identify it as a brand, and think the writer is talking about cheap flights, and not Cheap Flights.

As such, they frequently use “” as the preferred anchor text of choice, because it makes it clear they are talking about the web company Cheap Flights, and not cheap flights.

For those linkers who have this level of sophistication, or otherwise, aren’t too concerned about what their users think – they simply capitalize the anchor text – because they are still conscious of Cheap Flights as a brand, and because of that, won’t ever link to the website as “cheap flights” without the capital words if they identify it as such. After all, even the dumbest, most inane person won’t ever call Pepsi pepsi or Coke coke – if they have any grasp on the English language – or proper formatting – at all.

Exact Match Preference and Google’s Rationale

Exact match domains are given preferential weight because Google knows that many times, people are attempting navigational queries, so when people input “cheap flights” looking for, they need to return that result, or they’ve failed at their job. The biggest difficulty, though, is differentiating between those times that these domains are domains people are actually looking for – and throwaway domains that just want to rank for a monetizable keyword.

On those less competitive SERPs, these things aren’t a problem, and most exact match domains will be returned rather easily (and correctly). On the more competitive SERPs, it becomes more of an issue sorting out the proper alignment, and Google clearly has had problems sorting through the mud and knowing when a domain needs to be returned on the first page for a navigational query – and when it does not.

Exact Match Domains – A “Proper” Solution

When spam domains do link building for their properties, they’re often completely ignoring any brand their domain might project, and as such, they choose anchor texts like “cheap flights” instead of “Cheap Flights”, and do this en masse. When webmasters choose to link to them or otherwise are manipulatively summoned to do so, they come to the same conclusion, and use rather plain, uncapitalized anchor text to describe a product or a service, cheap flights or business insurance, rather than a brand, Cheap Flights or Business Insurance. As such, they are giving no indication that or is a brand at all – and, in reality, not worthy of a navigational query. They’re simply giving a nod that these anchors were frequently obtained in a manipulative fashion – and nothing more.

If we put 2 and 2 together, this means that Google would be apt to devalue, or otherwise, completely ignore these uncapitalized anchor texts, and only give a worthy, navigational boost to domain matches when they appear frequently in the following, example fashions:

  • Cheap Flights
  • Cheapflights


  • cheap flights


  • Cheap flights
  • cheapflights

This would return those results that have been clearly identified as some kind of brand. The less competitive the SERP, the less it would matter. But on the decent sized ones, an indicative volume of Proper Noun, capitalized anchor text would be a best indicator of whether a domain was worthy of being returned sooner rather than later. Exactly how you would attribute value to each indicator would be hard to say, but surely, the one that would offer up the least indication a domain is a brand is the spaced, uncapitalized version (cheap flights). And the largest indication a domain is a brand would be the capitalized, spaced version (Cheap Flights),  followed closely by the “” version (

It’s also possible that Google could pick up how the domain is used in a sentence, such as with “I love going to Travelstart for cheap flights” being a positive brand boost as opposed to “I need business insurance”. It seems like this would be an even better indicator of brand significance than the anchor text signals, but the level of sophistication the Google machine would require to pick up the idiosyncrasies of language therein make it difficult for me to currently rationalize as a powerful ranking signal.

On the opposite side of the spectrum, an intense volume of anchor text not capitalized would be a pretty good indicator that an exact match domain was manipulatively obtaining their links – so, all it would take would be to apply some weight to this, and once it hit a certain threshold, ding and/or devalue these domains. These uncapitalized anchors, of course, will still happen, so it doesn’t make sense to give them no weight and/or penalize immediately – but if they match exactly with the domain described, it makes sense that they should be used in a capitalized or “” fashion – or there’s some indicator that manipulation was involved, especially over a large enough sample of links – a large enough sample that would be needed to require ranking on competitive, and important, SERPs.

Brand Mapping and Competitive Research

SEOMOz’s Open Site Explorer webapp view doesn’t currently show anchor text broken down by capitalized or non-capitalized, but if you export to CSV, it shows how the anchors differentiate. Non-surprisingly, when I did this for links to Cheap Flights’ homepage (and only got a small sample of 10k links, they have hundreds of thousands of links) – they had a disproportionate amount of “Cheap flights”, “Cheap Flights”,, and etc in comparison to “cheap flights” – because they are a brand.

If you do this analysis on the spam domains, I would say, pretty confidently in fact, that this would not be the case.

Moving forward, if this appears to be a factor that’s become reality, it might be worth it for Rand Fishkin and co to implement this in the “Anchor Text Distribution” view of their webapp. There might be other webapps, too, that already do this – but I don’t use them. If they do, it might be a worthy reason to switch app usage.

Present, Past, Future – And A Warning

I have no indicative data sample to say if something like this is being used. Or, if it’s going to be used. But it makes sense, to me. Strong sense. But, again, and I warn – I am no algorithmic expert. I do not shine in the intricacies of information retrieval, so I am not one to say whether Google picking up something like sentence structure is even happening, or how, exactly, Google might pick up or interpret things as small as whether the anchor text is capitalized.

But I do excel in common sense, and these things pop as clear indicators of manipulative, spammy domains – so even if it’s beyond Google’s reach of implementation, at least short term, it’s something to worry about in the long term. Of course, you should be building a brand, anyways – but we know that’s hard, and takes resources. Either way, brood deeply on this, look at how it applies to what you’re doing and how you’re doing it, and adjust. Because it’s one of those awesome things – that even if you end up being wrong – it almost certainly doesn’t hurt you anyways.

But if you end up being right.. well.. hello #1.

  • Pingback: Tweets that mention For your Monday morning/mid-afternoon/evening reading - How Google Might Turn Down The Knob On Exact Match Domains --

  • Kyle Alm

    Intent is tough to determine. It’s tough to tell if someone is searching for or a “cheap flight.” A lot of older internet users just type in the domain name into the search bar. One person I know, his father doesn’t even know how to open a browser without doing a search in the Google Desktop sidebar.

    Domain names can be just another form of spam in a lot of instances.

    I suppose that the easy answer is for Google just to take over the net completely and return their services for any kind of search that has a possibility of initiating a transaction.

    • Book Flight With Us

      The years is 2013 and google has implemented various EMD, panda and penguin updates, exact match domains are now penalised severly for instance is an almost exact name match for the website for south african airways… its penalised

  • Kyle Alm
  • Kyle Alm

    That one, sorry last link was the corrected suggested search. See Google knows what you meant ;)

  • Colin

    I guess in a way it all comes down to variations in the anchor text. If you have anchor text which is mainly “perfect” and there are too few “no follows” then this is likely to be an artificial link profile.

    • Ross Hudgens

      I wouldn’t say that, Colin. I would say that obviously it matters. But I would argue that having too few nofollows is ridiculous, this has to be pure variance and there’s no way Google has ever dinged anybody for it. It may have existed on profiles that got dinged, but they never set up a signal that said “hey this profile has too few nofollows! Let’s ding them!”. It just seems ridiculous. And who’s to say Google even RECORDS nofollows? Their index doesn’t profile them (most of the time – see Twitter) so what use is it if they aren’t using it to ding people as I mention?

  • Webmaster

    What is really boils down to is for Webmasters to stop crying about being beat out in the SERPS by those who can do it better.

    Every since 99, Webmasters have been crying to Google about their sites are the best and deserve to be above the others whether they are an exact match or not.

    There are only so many positions on the front page and competition has got a lot more stiff. So, whiners need to suck it up and start working at beating the sites that are beating them instead of whining to Google and hope they do it for them since we all know Google so loves us Webmasters and does everything they can to please us.

  • Peter Unitt

    Great post and very in depth analysis and solutions. I’ve been talking about exact match domains for a while now and it seems there are 2 opinions, one- they are good and 2-they are bad. See my post on SEOMoz for some of the reactions-

    What I’ve learnt from this is that people automatically misunderstand the reasoning for noting the weighting of exact match domains, they think that we are moaning because exact match domains are above our sites, this is simply not the case! Any good SEO can get above an exact match if they really wanted to.

    The problem is how they appear to users in the results. I know not all exact match sites can be categorised as a spammy tactics (i know many that have great content) however I see a large amount that rank purely because of it’s domain, the content of the site is nowhere near the quality of the sites around it. That’s really hat is the concern here, not whether it works for SEO, it clearly does, just a little too well

  • Guy from a Website

    Content is king. I like generic keyword domains but I don’t like spam, and I’ve tested this with a few domains I own and the fact is that content is king. Google and other SE’s would not be doing any good to themselves and/or the average person who performs a search by separating exact match domains based on Capitalization or otherwise. The fact is that exact match domains have higher CTR and always will and the separating of capitalization and link anchor structure would not change this.

  • Audi Guy

    I hope the Exact match craze is not over. All though Im sure it is with my luck. Always late to the party. But this maybe a good thing since proper SEO will end up winning out!

  • Kiwi

    Google is smart, 2 years from now they won’t give a crap about keywords in the domain name. Those people who got lucky with nice keyword rich domains are going to need to be beating your door down Ross! I love it.

  • Meagan Andersen

    The problem with this is you can’t control all the people who wants to link to your website and how they link it. But making your anchor texts something like: “” – makes it more of a branding than targetting for the keyword “Cheap Flights,” isn’t it?

    • Ross Hudgens

      Meagan, my answer to this is my answer to all of these dilemmas: you are right of course, but at scale, things will even out. It’s the same thing with bad links. No one bad link will hurt you. But thousands will. So at the thousands level, or at the link building level where you have done this every time, Google will see it through and potentially make an impact.

  • Pingback: Le keyword nel dominio varranno di meno?

  • Pingback: Le keyword nel dominio varranno di meno? | Big Marketer

  • Pingback: Exact Match Domains: This is Why They Can Still Rank | Search Engine Journal

  • Pingback: Anonymous

Previous post:

Next post: