The Exact Match Domain Playbook: A Guide and Best Practices for EMDs

Exact match domains have always been the source of a lot of contention among SEOs. For quite some time, EMD’s have offered a competitive advantage for SEO’s who understood how to use them. In the early days of search when relevance algorithms were rather weak, many folks used “double dashed” domains because they were cheap to buy, and easy to rank. $6 to rank for a 3 word phrase. Sold. However, when you see best-online-seo-company.biz in your search result, you start to question the weighting of relevance factors. This is, in large part, how EMD’s got a bad rep to start with.

Despite the debate and obvious abuse, EMD’s represent what SEO’s do best – Seize Every Opportunity!

I’ve always been a pretty big fan of EMD’s, and I agree with Elliot Silver that  EMD domains can be brands. High value keyword domains have been a commodity since the internet became a commercial marketplace. 

Matt Cutts (esteemed Google engineer) has made this comment in the past (about 2 years ago):

“We have looked at the rankings and weights that we give to keyword domains and some people have complained that we’re giving a little too much weight for keywords in domains. And so we have been thinking about adjusting that mix a little bit and sort of turning the knob down within the algorithm so that given two different domains, it wouldn’t necessarily help you as much to have a domain with a bunch of keywords in it.”

Types of EMD’s

I think it’s important firstly to qualify the different types of EMD. The major factors in identifying quality domains, to me, includes:

• TLD extension (.com/.net/.org/.other)
• Number of keywords
• Dashed or non-dashed
• Domains with “stop words” only qualify as “partial match domains”

Let’s start with dashed domains. It has been proven statistically that domains with more than a single dash are very likely to be spam. Multiple dashes in a domain was an early spammer trick because of the low barrier to entry with cost. This rules these out.  Don’t bother with a double dash domain. 

It is very common to see domain names that include a single hyphen, but when two, three, or more hyphens are present, this is often an indication that these domain names are associated with companies that are attempting to trick search engines into ranking their web pages more highly.

From this amazing patent post by Bill Slawski regarding EMD’s and detecting commercial queries.

Even though there is plenty of evidence that a single dash domain can rank just fine, I would avoid this technique as well. Many of the single dash EMD’s that rank are old existing domains. It is my opinion that a single dash EMD still really provides very little advantage over a non-keyword domain with all things considered. Skip the dashed domains as well.

The next question is how many keywords in a URL is too many. I would answer 2-3 for a .net/.org and 3-4 for a .com. BestBusinessCreditCards.com may be long, but I think it is still effective and too the point. Four words is pushing it, but I think you can still make a valid argument for a 4 word .com domain in certain spaces where most all the domains are taken, and there are some very niche commercial products worth targeting specifically. While .net/.org domains are still very credible, there are more of them available, so I think you have to reduce by a word. 4 words max for .com, 3 words max for .net/.org.  3 word .com or 2 word .net/.org is the best idea.

.net/.org keyword domains have proven to be very effective as a tool for bootstrapping a website. I think this is valid strategy. Instead of flickr.com, start with onlinephotos.net or even onlinephotogallery.org. I understand the value of a brand, but I think there’s also value in embracing “bootstrapper traffic.” There’s definitely a lot of value to a startup in some highly relevant long tail traffic from your targeted keyword phrase set.  .net/.org domains are generally priced at about 10% of the value of a .com domain. This can be of great value in competitive verticals where most of the domains have been registered for many years. 

Find the BEST two-word .net/.org within your category, and buy it in the aftermarket if it is available. For buying your EMD – see the advice below. This can be great for your mainsite, microsite, or just to keep your competitor from getting it. At worst, think of it as a defensive strategy for your most important phrases. Just don’t think you’re going to dominate the SERPS spending less than three or four times what you paid for the domain in the first place. A crappy microsite  that costs half of what you paid for the domain will get you a one way ticket to Nowhereville these days.

Stop words in keyword domains

Domains that include stopwords don’t truly qualify as an EMD, but can be mildly effective. It probably wouldn’t be my first choice, but if you can get theDetroitRoofer.com for $6, it will probably be a decent bet you’ll have some decent signals at your advantage in ranking for your targeted term for the relatively reasonable future. There is the potential for some brand confusion here though if someone owns detroitroofer.com

The most significant benefit of an exact match domain is that it makes it much more easy to develop targeted keyword anchor text from authority sites. Anchor text as an SEO tool is in decline, but it has always been a very significant factor, and will likely remain this way to some extent. It’s much easier to get someone to link to your site with the domain name, than it is to tell them “link to me with these keywords.” This is probably the major competitive advantage over non-EMD domains. 

Offsite optimization is more than just links these days with the increasing importance of social mentions. Smart money speculation says it will be easier to get keyword rich social mentions for an EMD than for other types of domains as well.

So with all the talk of EMD’s, what the people really want to know is: what should we do? For those of you in this camp, let me offer you my best practices with keyword domain names. Unfortunately, I can make no guarantees to the amount of time these will hold true in the ever shifting tides of SEO change, but this is where I think we’re at as of the time of posting:

EMD and domain best practices

1. Always be willing to spend 10-15% of your overall budget on the BEST domain name you can get. It will make a big difference in both the short and long run. Dive into the aftermarket, and send some emails.
2. Skip the second level TLD’s – .mobi / .travel / .info isn’t worth it.
3. No more than one dash in your domain (better to just skip dash domains altogether)
4. 3-4 words max for .com EMD’s
5. 2-3 words max for .net/.org EMD’s
6. Best to build a Brand site on a keyword domain so you get both brand mentions and generic intent keywords (see Toys.com owned by ToysRus.com and associates)
7. Geo-local EMD’s are great to own, and offer lower barriers to entry
8. You’re going to have to focus some efforts on “de-optimization”

As the proud owner of MiamiFishing.com (no, I’m not a retired fisherman, but thanks for asking) and other exact match domains, I can say that there are both pros and cons to EMD’s. I saw a few sites of my own pay the price for “over optimization” during Penguin. It’s hard to always know how aggressive to be, and how far G is going to turn the “filter knobs,” In a time where disavowing, delinking, and de-optimization seem to be the valid strategies, it’s safe to say you should probably take a more conservative approach to your organic ranking strategy.

SEO factors aside, there’s something valuable about having your domain name “say on the box” exactly what you do when you put it on a hat, t-shirt, or sign. There’s a lot of implied credibility in a .com EMD (and even to some extent .net and .org).

After years of being an SEO, it’s sometimes difficult to maintain a TAGFEE mentality and put my own site up on the chopping block for public criticism, but it’s a site I’m also very proud of, and I think really stands up to the other websites in the vertical in delivering value to our users. Please be gentle.  I do believe the Moz has great tenants, but it can be very frightening to put your site up in the crosshairs for people to take aim and fire at, especially when you haven’t accomplished everything you’d like to do with it sometimes. Being optimized or optimal means getting the most you can from the resources at your disposal, and sometimes this isn’t always enough to create the perfect website (I have others that aren’t nearly as pretty).

EMD’s do have their advantages, but they have some disadvantages as well.

Pros of an EMD

• Great for a startup to gather some relevant longtail traffic
• Easier to get targeted anchor text
• Easier to get social mentions with keywords
• Can dominate a single niche (IE: “Category Killer”)
• Good for targeting variations in the long tail keyword phrase set
• Brand mentions and keyword mentions become one in the same
• They can be very effective for generic commercial intent queries
• They can be very effective in local search
• Great way to build startup “bootstrapper” traction
• Can be an effective strategy with a well built microsite to target a single niche.
• Some businesses have very limited keyword sets – this is a decent approach in these areas.

Cons of an EMD

• Limits future brand expansion
• Can create “brand confusion”
• You don’t get the same “credit” for brand mentions.
• Your brand can come off as “generic”
• It can be harder to claim social media profiles
• It can be more difficult to associate mentions with your brand
• Hatorade on your site quality if you outrank competitors
• More chance of “over-optimization” (seriously, does anyone else hate this phrase as much as I do?)
• There are a limited amount of them
• They can be very expensive
• The effectiveness of the advantages are slowly being neutralized

EMD’s and Brand Confusion

One of the main problems facing EMD’s is the brand confusion that can come with a keyword domain. It’s HARD to own a very sought after generic commercial intent keyword. Google really doesn’t want someone to own a keyword, and for good reason.

Keywords are the new brand. Someone in every vertical is trying to own their generic commercial keywords. Think about the big brands Staples and Office Max; do they really DESERVE to rank better than a well built OfficeChairs.com or OfficeFurnitureOnline.com ?

Generic commercial intent keywords are hard to come by; there’s really not a ton of them around, and they are VERY sought after when you start looking at the search demand curve. It doesn’t make sense to for a SE to allow only one advertiser own the keyword when several can compete to drive prices to a point of maximum profit for G and diminishing returns for advertisers. There will always be competition to be the brand associated with the generic commercial intent keyword. Logic follows that value in the associated domains should stay pretty strong as well.  

This is probably beyond the scope of this post, so I may leave this discussion of “branding” keyword domains for another day, but it is at the crux of the EMD debate. I’ll leave the solutions to the commenters ;). We all know that G is expecting much more out of a website to allow it to remain on their first page these days.

Think there’s a lot of keywords with generic commercial intent? Consider the main ones in each of these categories where G makes the majority of their ad revenues.  The list might not be as long as you think. I’m willing to bet most consultants and agencies here in the Moz community have at least a client or two in each of these major verticals.

So what was the “solution” to the EMD relevance “problem?”

Google engineers have always attempted to “level the playing field” for webmasters. They do a great job in many cases, and provide lots of fantastic tools these days with Google Webmaster Tools. Unfortunately, I don’t really think EMD’s are inherently a bad thing. They were just too large of a competitive advantage for some competitive niches where it was difficult to get targeted keyword anchor text. It’s still going to remain difficult to get targeted anchor text in these niches (though it’s now much less valuable to do so). EMD’s became a goldrush landgrab for optimizers and domainers when they saw the advantages they provide, and the tactics got used and abused and started to create some relevance problems.

As with all landgrabs, people got greedy. Speculators starting creating sites that gave EMD’s a pretty bad rap.Competitors start reporting these websites go Google as S.P.A.M (sites positioned above mine), and users start to complain that the SERPs suck. Speculators started putting up 1 page garbage microsites and ranking for large 2 and 3 word phrases with 3 crappy directory links and a page of outsourced content. The EMD’s started to look like those old double dashed sites, even though the barriers to entry for top search rankings were a bit higher. Those barriers continue to get raised.

You can’t cry about your rankings when you didn’t deserve them in the first place, and honestly you never deserve rankings. You earn rankings, and often lose them. It’s part of the love, joy, and pain that is SEO.  As John Andrews says in “You’re Free to Go Home”  “That’s alot like SEO. You win, you get traffic. You don’t win, you don’t get traffic. It doesn’t matter how you play.”

The real issue with EMDs

The main issue originally posed by suffering relevance was not EMD’s, but the amount of influence that keyword anchor text wielded over the search relevance algorithm. EMD’s just benefitted disproportionately from advantages with targeted anchor text. Anchor text carried too much influence without that added benefit. It’s a whole lot easier to get a link that says “Real Estate” when you’re RealEstate.com than it is to get one when you’re  Zillow.com. The same can be said right down to Buy-my-crappy-spyware-cleaner-software.com.

It was much more important to fix the overall issues associated with the anchor text relevancy problems, than it was to fix the EMD “problem,” and that’s why we saw the anchor text issues being remedied first with Panda and Penguin (which fixed a slew of other issues as well), before directly fixing EMD issues. There is a lot of potential collateral damage that can occur when making the decision of if a keyword domain has enough “brand signals” or “quality factors” to be near the top of the search results for a phrase, so I imagine it’s a pretty difficult search relevance area to tackle. The simple fact is many EMD’s ARE good valuable sites that deliver a quality experience to their end users. Can you really take a way all their advantage that they were wise enough to gain from paying top dollar for a great domain?

As with most important signals, optimizers found a way to take full advantage of benefits that inbound keyword anchor text provided. As with the rest of the history of SEO, we’ve seen a major shift in the importance of anchor text that has sent a lot of SEO’s reeling. If you didn’t see the writing on the wall, you either didn’t pay attention, or didn’t care. Either way, SEO’s who ignored the impending anchor text over-optimization warning bells are now paying the price, and trying to fix mistakes.

Panda and Penguin cured most of the major EMD relevance issues by forcing EMD websites to earn their rankings through achieving acceptable engagement metrics. Think of Panda as a beast that eats sites who don’t give their users what they want. If you don’t hold up the the “relative engagement metrics” within your SERPs, your site gets eaten.

If I were to play “if I were a search relevance engineer” (one of my favorite games), I think would just set the barriers to entry higher for EMD’s to rank in the short and medium tail keyphrases. I would also validate with user metrics the fact that they deserve to be there. Long ago (in 2005), Google introduced the “sandbox” (or trustbox) The “trustbox” made new websites “guilty until proven innocent” with regards to their page authority unless they demonstrated sufficient signals to be let into the index. 

The principles and ideas associated with the trustbox are still very much in effect today. Value to your users creates trust and credibility verifying engagement metrics like high time on site, multiple page views, low bounce rate, repeat visits, and new websites are let into the index more quickly, but the barriers to entry for commercial intent high dollar short and medium tale queries are much higher. Essentially, your user engagement metrics must validate your rankings. 

Yes, that was an “Eminememe”, and as Eminem says: “you get one shot, never miss your chance to blow.” When you get your “audition phase” in the top of the search results, your site needs to perform well against other sites in that keyphrase set. Make sure you pass your “audition” instead of puking on your visitors sweater and telling them it’s value. Positive engagement metrics during your audition phase is equivalent to the importance of quality score in you PPC campaigns; it can really have an effect on the outcome of your webpage’s success.

Positive engagement metrics

• High time on site
• Multiple page view
• Repeat visits
• Low bounce Rates

Not every industry requires 10 minute time on site, and 50% repeat visitors, but some do. These metrics reflect brands and brand signals, which is what G has repeatedly mentioned as their priority for providing quality and relevant sites to users in the search results.   

What are the solutions to my exact match problem?

It’s obviously a bit troubling times for EMD owners. No one likes to be at the center of an SEO witch hunt. It’s all fine and good to do spam reports, until it hits your site, or targets your niche or competitive advantage. One of the best competitive advantages has always been the ability to stay under the radar and keep your mouth shut (though I sometimes fail to fail at what I preach).

The solutions are the same as to many of the problems with Panda and Penguin. It’s a tough time to be a site owner, and admit that you were “over-optimized” and start back peddling a bit, but it’s G’s world – we just play in it. How many times has Google said it? Focus on the user. You may have always scoffed at doing “what’s good for the user,” but with engagement metrics that suggestion has turned into a requirement. We’ll continue to focus on both how to make our websites better for users and Google with more actionable execution taking advantage of how user’s interact with our sites via search engines. 

There’s still plenty of advantages to EMD’s, and we should continue to see instances of their success, but it’s hard to build a generic commercial intent keyword brand. You gotta have the chops to back it up!

We all know ranking for generic commercial intent phrases is valuable, or we wouldn’t be targeting them.  In order to stand up to the scrutiny, you’re going to have choose your favorite EMD’s, and let those other pipe-dream microsites die their slow painful death. It’s important to know when to pull the plug on a losing web property. Any good web entrepreneur has plenty of failures on their resume.

A few things to consider for solving problems with EMD sites:

1. Disavow all public knowledge of SEO
2. De-optimize
3. De-link
4. Prioritize your SEO efforts – you can’t win the battle on all fronts anymore
5. Focus on quality of quantity (with site indexation)
6. Redesign and Rebrand (maybe it’s time to get a mascot for your .org)
7. Innovate ways to improve user engagement metrics
8. Develop a social presence and improve your social mentions
9. Diversify your backlink profile
10. Diversify your anchor text
11. Okay – I’m (kind of) kidding on rule #1 – #3

You know who hates on good EMD’s most? The people who don’t own them. You know why? Because they’ve always carried an advantage with them. While this advantage is diminishing, there is still a tactical advantage in spending some money up front for a great exact match domain name that describes exactly what you do and acknowledges the generic commercial intent of your visitor.

 EMD’s will always receive lots of hatorade because the majority of people don’t own them. Toolman at webmasterworld said it best: S.P.A.M = Sites Positioned Above Mine. There’s plenty of SEO’s who could make Silky Johnson look like Tony Robbins. Don’t participate in the hate, and don’t feed the trolls.

Very few people are going to come out of the woodwork, and “extol the virtues” of an exact match domain, and put their website under the ever scrutinous eyes of search engineers, and a community that often prefers to focus on failure instead of offering opinion for improvement. As usual, I enjoy being the exception to the rule, and figured I’d pitch in my two cents. 

How to find and buy an EMD (and avoid being a hater)

1. Type in whois.sc/yourkeyword.com/.net./org (this will redirect you to domaintools whois search for the targeted phrase)
2. Identify if the domain is owned by a domainer or owner and do some further research
3. If there is no established website – Write an email and ask if the domain is for sale.
4. If you get a response – offer approximately 40% of the asking price, or propose one high enough to not offend the seller.
5. Meet in the middle if .com is worth it.  if .net/.org offer 2-10% of your .com price
6. If there is an established site, check the other metrics, and be prepared to pay much more.
7. After EMD “death” be prepared to pay more for domains in the aftermarket after their “rebirth”

For more on domaining, check out the domainer myths.

Take my opinion on EMD’s with a grain of salt. No, I didn’t test my theories like Pete. This is just my experience. I have bought a fair share of them thinking they were a great buy for future projects, or just to invest in and sell in the aftermarket at a later date. We’ve been warned of the “death of EMD’s” for a long time. I just hope EMD’s continue to suffer the same type of death that SEO constantly battles with: one that is curable with creativity, innovation, and execution.

Resources: