Why EMD’s should stay

Yet again parts of our industry are under fire from algorithmic changes at Google.

The focus right now are Exact Match Domain’s (EMD), something that us SEO’s have complained about for years.

I’ve lost count of the amount of times that I’ve seen an SEO moan that they site can’t outrank “buycheapbluewidgets.com” or whatever it may be, just because its an exact match.

Thing is, those of us complaining about it were only ever doing so, as our issue was our loss of profit – to a site we considered to be less worthy than our own!

That’s a heck of a double standard if you ask me..

So – what happens now then?


Well, early evidence suggests that the average position for EMDs across a large dataset has dropped quite significantly on average. The thing with that dataset (mozcast.com incidentally) is that it only checks the top 10 results for any given key phrase so its harder to get a more accurate picture from it, but the graph above does illustrate a fairly big swing.

Its as if your EMD was just nuked by Google.

Now this is something I have a problem with.

Most of my good domains historically have always been close, if not exact matches. I’ve always however married them to good content, and have enough of a grasp on technical SEO to ensure no bloopers, and as for linkbuilding… well – I can throw a link or two around when I need to.

What’s really bothering me though is the absolute intent here. The sites that have benefitted from this are pure informational sites, Wikipedia has (somehow) managed to increase in visibility on my dataset, and lots of really random blog based sites seem to be getting more juice out of this.

A similar pattern occurred in 2010 with MayDay, which was pretty much reversed by early October – but it proved that returning informational sites only was a mistake.

Listen Google:

A lot of the time, people doing searches where EMDs will appear, were looking for that exact service or product. Duh.

So someone went to the effort of buying a domain and building a site around it? Good for them!!
They are probably trying to monetize it by providing the service that is relevant to what the person was looking for. That being the case its your job Google to rank that site, and rank it well.

If its not the case, and they’ve done a shitty job, then fine, but use metrics you’ve actually got a better grip on – like monitoring toolbar usage for obvious signs of doorway sites and just upvote the destination site, that’d sort out the .info spammy EMD’s – if you just counted them like a link.

Alternatively, just set the trigger on people returning to search and refining their term after clicking through on that result. Actually, if we could have more of that in the whole index, I’d be grateful thanks.

When is an EMD not an EMD?

When it gets triggered for queries that aren’t the domain name of course.

Looking through my data, I’m seeing lots of scalpel like precision drops for exact matches, but guess what – in circumstances where the same site ranked for other terms, those appear to be doing just fine.

That’s not on Google sorry, it’s a really crappy solution to a problem that never really existed, unless you were an SEO complaining about the strength they used to hold.

I for one am perfectly happy with exact matches remaining powerful – at the very least you know they are going to be relevant.

27 thoughts on “Why EMD’s should stay”

  1. I think I broadly agree with you. In terms of demoting EMDs which were used for low quality content and spam, then I’m not against that in any way – in addition to making search results better for consumers, it should, in theory, allow brands and legitimate efforts to perform better.

    But in some ways it also hinders those efforts where good quality content and brand building has been undertaken, and an EMD has been used for this, or to supplement it. If I’m working on a small business that needs to support itself, and there was an opportunity to use an EMD to boost awareness and search results for a specific area of their business with unique content and a focused offering, is that a bad thing?

  2. Hey Martin

    You have a point in that a site should not be punished just because it was an exact match and really, the EMDs that are just cruddy doorway pages should not be up on high on the back of that any way.

    Likewise, the slow, steady upturn in the return of informational sites is great, to a degree, for informational searches, but if people are searching for a service, the search should give them what they want.

    This is where the conspiracy theorist in me wants to scream “they are doing this to promote adwords” but in reality, it would, seem they are just doing this as the algorithm seemingly can’t do it’s job in this instance so EMDs are another ‘special case’ and as with everything else, there will be casualties, innocent, casualties.

    As ever, it will be interesting to see how this pans out and if we do see an improvement or if the bigger boys come off like the mob as seems to be par for the course.


  3. I think you’re right in that just having an EMD is not a reason to discount a site’s score all other things being equal. Personally I’d prefer if having an EMD gave no impact (positive or negative) and is was just removed as a ranking signal.

    The other thing to consider is EMD’s on some occasions were already damaging themselves. For example I’d never buy from a site like buyredwidgets.com as it doesn’t sound like an established brand. I’d much rather pick something like thewidgetcompany.com as it seems more reliable. So even when EMD’s outrank other sites it’s still no guarantee of a sale lost to a competitor or a reason for a mass cull of EMD’s.

    The other factor is this is further evidence that building a brand helps, because Google will still be keeping the EMD benefit given to brands as there’s no way they would nuke them. So if you have an EMD and a strong brand (called the same as the EMD) then you’ll most likely survive this with no issues.

  4. From my own observations many EMD increased in rankings after Penguin, and from a selfish point of view, I don’t mind if they drop back down.

    Seen any evidence of ranking dips for exact match page name URLs?



  5. I think (single word) EMDs can also be quite a good quality signal. Or at least a good spam filter. As some spammer, not investing in good content, would also not raise the money to buy an awesome EMD like diapers.com etc.

    Usally people playing around with EMDs spend some more effort than the usual mortgage-loan-247.com spammer. BUT: most of the EMDs containing more than one keyword are still crap.

  6. I’m with Mike Essex on the EMD thing – if EMDs received no ranking benefit, and were ranked purely on their own merits (IBL / Relevenacy / Social etc) then I think that would be pretty cool. Then again, I don’t really own any EMDs…

  7. I’m quite happy to see EMDs completely eviscerated. They’re a lazy, inauthentic way of ranking. Not to mention that it’s a closed shop. Only one person can have the EMD for any term – that makes it a race, not anything involving skill or quality.

    Two thumbs up from me.

  8. Hi Martin,

    personally I am closer to the idea Mike expressed here above: EMD should not have had a too extreme high positive impact for being an EMD, neither they now should they been so punished for that same reason.

    I must admit I have not really dug into the details of the EMD Update (not having sites fighting in Google.com for queries in English), but from what I’ve read the penalization was presenting a wide range of grades: sites that lost few positions and others literally disappeared from the SERPs.

    From what you say, it seems that the update penalize “a la” Penguin, being the keyword in the domain name the equivalent of the over-optimized anchor text in the toxic links. So… I can only imagine the combo of Penguin and EMD Update.

    Said that, even though I applause the idea of acting against the EMD supremacy, at the same time I am (again) surprised by the lack of sophistication Google shown. As you wrote, Google can perfectly cross data and see if an EMD site is really the online representation of a real business and valuable for the users and EMD built just for gaining traffic and then monetizing it without offering a real value, as in many affiliates or done-for-adsense sites.
    Or, simply, Google could have discounted a big part of the EMD positive ranking value, indistinctly, so that only those ones with a real authoritative link profile (now that they were depured by Penguin) could stand.

  9. My ideal is that an EMD shouldn’t rank just because it’s an EMD, but at the same time, it shouldn’t carry a penalty because of it either.

    It should be treated as any other domain. Remove the EMD advantage for sure, but don’t swing it the other way, that way, the sites are ranked based purely on the other ranking factors we all look for.

  10. Hi Martin,

    I think the issue Google is trying to deal with goes beyond a simple of case of whacking EDMs down the ranking. We’ve seen an influx of exact match domains an pure generic keywords, some of which do not offer value to users. I’ve come across domains like iphone4price.co.uk, iphone4.com etc with really thin pages and I think that is what Google is going after. Google want websites to have brands and lets face it, you would struggle to build a decent brand on iphone4price.co.uk and probably won’t be able to get enough decent links to give it weight. It therefore painful to see such websites ranking ahead of the big mobile phone providers and decent review sites.
    If you have an exact match domain, have good number of links and have good content, I think you should be fine.

  11. I’m not sure that sites are actually getting “punished” because they have EMDs. My suspicion is that factor has been reduced in weight in the algo, not that a penalty was applied. If the EMD was 50% of what was making a site rank, then of course a reduction in that factor’s weight will drop traffic and rankings. If you’ve used that EMD to build a big audience who is engaged, shares your content, links to you, etc. and to get incoming links with good anchor text, and the EMD itself is 2% of what makes your site rank, you probably won’t see much of a decline.

    It depends on the SERP though. If there is only a tiny bit of difference (according to the algo) to the top 10 sites a tiny downgrade might drop you several spots. That’s not different than any algo change.

  12. Hi Martin,

    Interesting post – thanks for sharing your thoughts.

    I agree with you that EMDs are typically a reliable signal of relevance – but there shouldn’t be anything stopping a branded site from achieving the same level of relevance.

    It seems a tad early to suggest that Google has declared war on EMDs in general. If we set the graph min to 0 instead of 3.1, the change doesn’t appear very drastic to my eye. There are also still plenty of EMDs ranking.

    Google specifically mentioned “low quality” EMDs. I’d be curious to know how G defined, judged, and quantified that – separately from Panda. Perhaps some specific examples could help us understand the problem you see.

  13. When I’m searching I am more than happy clicking on EMD’s – if I’m trying to find something online and an EMD delivers the “right” snippet in the results then I’ll click on it – in fact I’ll twice as likely click on it! Why? ‘Coz I’m trying to make the task easy as possible.

    Unless the brand is a well known (global/UK) brand to me then it won’t make any difference to me whether it’s an EMD or not. I don’t choose an unknown site just because it sounds like a “brand” in the SERPs…

  14. Counter to what I said earlier, it astounds me that this applies to page names. I agree with Google that exact match domains should be a thing of the past, but why take value away from correctly-named pages? Bizarre.

  15. Its a disappointingly simple cull – an arbitrary excision. Hitting EMDs is just as daft as hitting SRDs (Semantically Related Domains). Taking out sites with EMDs is far too simple a solution to a tougher problem. Id like to believe there are more metrics involved in the decision, like time on site etc. Please Google, please tell us you did more than just drop domains based on exact KW matches.

  16. It does feel like google has ignored the problem for years (IE crappy adsense sites ranking on top) and all of the sudden nuked it all down. I liked the way they were dealing with it by reducing their power over time if you look at the seomoz data. Anyway I hope this didn’t break any lives :/

  17. Google took an discrimitive action against EMDS not because they are low quality but because of who they are. If I specialize in selling orages why call my site fruits.com where everybody agrees that oranges.com is better? What rationale are google using?
    Like Regan would say “if it is good tax it, if it continues to do good tax it again and when its dead subsidize it.” I may be a third rate thinker but I believe Google is simply telling you if you want to use EMD and remain on top, then advertise. If not we will pull you down.

  18. What a great article on many points that I personally agree with.
    I have built many sites with EMD’s over the years and do so to provide relevance and quality within an exact match niche. I always provide quality content, products and service so why shouldn’t I rank well for my efforts? Google should do more to analyse the sites as opposed to dismissing an EMD when quite often, its what my customers were looking for.
    —->EDIT BY MARTIN MACDONALD: Really? You’re dropping anchor text in your name field on this comment?? REALLY? ROFL.

  19. EMD should stay, because lots of webmaster are doing seo by just taking exact match domain. Those website are just ranking with the keyword. Website of exact keyword match domain with high quality backlinks are not get effected. Only low quality EMD websites are got effected..

  20. > Only low quality EMD websites are got effected..

    My EMD had BBC backlinks and still got nuked.

    Strangely Adwords is less fussy, giving me 10 out of 10 on the quality score

  21. I couldn’t agree that only low quality domains got affected by this. On our local market in a few niches we have seen many samples of high quality, trusted websites that suffer from EMD penalty. No duplicate content, no link networks, no spammy and automated backlinks.

  22. What an amazing post. What I like about it is you’ve taken a very systematic view of it all. I have the same views, but came from a different angle.

    You my friend, have just earned a link from my blog post about EMDs. I’ll credit you as a source of great EMD information – though, i’m not sure Google will like it because my name is michael charalambous, as is my domain, ooops I mustn’t rank!

  23. EMDS continue to dominate local searches for companies in and around Glasgow. They are often low quality websites that only have a few domains linking to them, yet its enough for them to stay in the top spot.

Comments are closed.