The practices of domaining and subdomaining have steadily grown into increasingly hot topics in the local search marketing space for the past few years. “Domaining” is the practice of buying domains mainly for their potential keyword value. Speculators purchase keyword domains with a view towards selling those properties at considerable markup, and/or using those domains to host relevant affiliate content links and/or contextual ads to derive profit from the traffic, clicks, and purchases resulting from the users who arrive at the sites by typing their URLs directly into browser address fields. Subdomaining is the practice of delivering sections of content on sites under third-level domain names.
There are two separate goals involved in these domain-oriented tactics:
One goal arises from the belief that having valuable keywords present in the domain name may likely translate into greater amounts of referral traffic from search engines—this trend motivates both domaining and subdomaining in the local space. The second goal comes out of an increasing awareness of the online traffic potential that the domaining industry is reporting from the “direct navigation” sector—a goal of grabbing and monetizing that vein of traffic. Both of these goals have affected the local search space considerably, as indicated by public announcements from local companies about their strategies, and demonstrated by observation “in the wild” of their use of specialized URLs to exploit the perceived benefits.
In this article, I’ll be looking at which top local companies have been doing subdomaining, and examining whether this practice is likely to be beneficial. There are quite a number of industry-specific vertical sites who are also doing domaining or subdomaining such as for lawyers, contractors, doctors, restaurants, etc—but, within this article, I’m focusing upon the broader local search and online directory sites.
Subdomaining remains a popular topic in search engine optimization circles. A subdomain is a variant domain name that’s created by adding a third-level domain onto an existing second-level domain. For instance, if “example.com” was a primary domain name of a worldwide company, they might offer up country-specific content on alternate third-level domain names such as “france.example.com” or a shopping section under “shop.example.com”.
Having desirable keywords in domain names and URLs can be helpful for rankings in search engines. In earlier days of SEO, it was generally held that having keywords in the domain name was likely a much stronger factor than having them elsewhere in the URL structure. Since setup of third-level domains doesn’t require additional domain registration fees, many have used the third-level name convention in hopes of saving on registration costs and perhaps infrastructure complexity while still gaining benefit from the apparent importance of having keywords in the domain names.
So, who’s doing this in the local space? It’s not difficult to see who’s trying to leverage this method, since one can run a number of searches for various types of businesses coupled with local parameters, and then see what sites frequently show up on the first pages of the SERPs. Here’s some of the top local companies that are doing subdomaining:
- Citysearch: miami.citysearch.com
- AreaGuides: lasvegasnv.areaguides.net
- Addresses.com: seattle-wa.addresses.com
- Local.com: jacksonville.localconnect.com (apparently only done in their private labelling deals)
- Craig’s List: minneapolis.craigslist.org
- AreaConnect: memphis.areaconnect.com
- Yellowpages LTD: accountants-tax-in-denver-co.yellowpagesltd.com
Is this approach effective or even worthwhile? Some SEOs report that the presence of keywords within domain names regained importance in Google algos during the last year, as perhaps a way of counteracting the effects of trust filters in order to help ensure that sites would be enabled to rank well for their own brand names. But, it is not clear that this benefit would necessarily extend to keywords in subdomains, even if it were true. Google and the other search engines all recommend a conservative approach to using subdomains, stating that there should be sufficiently unique, stand-alone content to justify new subdomain instances.
Does subdomaining even work? In recent months Craigslist pages have been jockeying up in local search results for a wide variety of geographical/biz-category search term combinations in Google for which they historically haven’t held high ranking. For instance, they now appear high for terms like “Dallas Auto Parts“, “Huntsville electronics“, “Nashville Accountants“, and “Houston Jewelry“—combinations for which they really didn’t even appear in the first page of SERPs on Google a year ago.
Oh, sure—they’ve long been popular in San Fran, and they’ve been having growing usage in other major cities, too, so one might suppose that their ranking increase could chalked-up to rising popularity, right? But, many of the city-oriented SERPs they’ve climbed in are not cities where they’re all that popular. For instance, look how high they rank for “Louisville child care“—they’ve never had the kind of user adoption in Louisville that they have in San Francisco, and do you think that people would be more inclined to click on fly-by-night classifieds when seeking childcare, than online directories containing stable businesses? Does that seem likely, even? So, what’s responsible for their relatively abrupt rise in rankings for these types of local searches?
Their sudden rise leads some local SEOs to speculate that perhaps their subdomains, such as dallas.craigslist.org, could be responsible for this sudden improvement. Has Google changed their algorithms to give subdomains greater keyword relevancy weighting? Perhaps, or perhaps not.
Craigslist has been crawlable for quite some time, and it doesn’t appear that they’ve done anything particular to improve their rankings. In an interview earlier this year, Craig states that they focus on good usability, but not on SEO. Still, their sudden improvement in SERPs for cities where they haven’t been historically popular is suspicious.
I’d theorize that this could actually be caused by a geographical bias introduced into Google’s ranking through their quality/usability testing. Craigslist has long held highest popularity in the San Francisco region where Google is located, so if Google has composed a “quality score” based off of usability testing of subjects only in the San Fran area, it could’ve biased their ranking of Craigslist which then got extended out to apply to all geographical SERPs, even where Craigslist has had considerably lower popularity.
But, maybe I don’t need to come up with a convoluted theory involving accidental bias—perhaps it’s an outright bias. Quite a number of people have suggested that perhaps Google has given ranking preference to Wikipedia, and this seems to resemble that effect. Craigslist very nearly is the wiki of local products and services listings, and their simplistic design and dedication to their user community above selfish profiteering are the sorts of things which would endear them to Google employees.
Either way, Craigslist’s recent increase in local popularity could easily be based solely upon factors other than their subdomaining with city names.
From viewing the first SERP for a number of traditional business-plus-locality search terms, one can quickly see that sites with local-flavored subdomains are neck-and-neck with regular domain names and don’t necessarily enjoy any greater ranking benefits:
- "Lawyers in Nashville"
- "Auto Parts in Dallas, Tx"
- "Italian Restaurants 80206"
- "Boston Accountants"
- "33160 Locksmiths"
There are a number of local search directories which rank just fine without the subdomaining:
- Judy’s Book
- Yahoo! Local
Considering how all of these sites may be found ranking strongly for various types of local parameter searches, it appears to me that keyworded subdomains provide no real advantage for SEO purposes. Without coming out and saying it’s worthless for keyword weight, engineers at Google have cautioned heavily about launching many different subdomains if each one doesn’t have sufficient content—they’ve said it might be better to just use a subdirectory (ex: www.example.com/plumbers/) on the primary site URL. Subdomaining can sometimes also introduce extra headaches since it becomes much easier to launch duplicate content under different URLs, and duplication can dilute a site’s overall PageRank.
In the relatively isolated cases where keyword-rich subdomains have risen higher than non-subdomained sites, I’d say it’s due to other ranking factors at play. It’s more about the quality and content of the pages ranked, and the domain name itself doesn’t appear to need any keywords whatsoever in order to rank well. The recent rise of Craigslist was likely not due to subdomains. It happened because Craigslist has great natural search optimization, rising popularity, and possibly preferential treatment from Google.
So, subdomaining likely doesn’t assist in gaining on local keyword rankings. Sites using subdomains for the purpose of managing content or for load balancing have a valid reason to do it, but doing it just for ranking could be construed as being a spammy behavior. People in the local space who are seeking a quick fix via subdomaining as a shortcut to achieving rankings similar to a CitySearch or Craigslist are likely to find themselves sorely mistaken.
(Stay tuned for my next installment in this two-part series: Domaining & Subdomaining in the Local Space – Part 2: Domaining).
Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.