This post was originally written by Ken Shafer, past SEO Manager for Apartments.com.
If you’ve been following search engine optimization at all in the past year, you’ll have undoubtedly heard the term “real-time search.” One aspect of real-time search refers to search engines using the social media pipeline, including channels like Facebook and Twitter, to provide up-to-the-minute results. There is a growing contingent of bloggers out there who believe that real-time is the “next big thing” for SEO. I decided to run a few experiments to help illustrate how this all works.
So how DOES it work? If we take an extremely newsworthy topic that is “trending” right now, you might see Google employing real-time results right there on the first page. Below, I’ve performed a search for “Prop 8.” Google has taken a portion of its search engine results page (SERP) and inserted a constantly updating box displaying the latest blog posts and tweets.
I tried to bring up the same box with “alex rodriguez hits 600” but was unable to do so. This should tell us something about how newsworthy the topic must be or the amount of “trending” that must be occurring in order for Google to display this in their main results page.
Unfortunately for us, it is less likely our multifamily news or our latest special rental offer is going to make it to this level of newsworthiness in the eyes of Google. So does that mean that real-time search is out of our reach? I’m not quite ready to give up yet.
If we type in “apartments in Chicago” into Google, we get the standard links we’re all used to seeing. No real-time magic here. However, if you click on the “Show Search Tools” button on the left of the screen, we see that it opens up some options. Clicking on “Latest” gives us Google’s real-time results. I’ll be the first to admit that this is not the ideal user experience and I don’t have any data that tells me how many people actually click on that “latest” button. However, I can envision two scenarios that make this exercise worthwhile.
- Search engines could start incorporating real-time results on the main page for more general search queries.
- Users could eventually become more familiar with using the “latest” button.
I want to point out something very interesting about this real-time results page. There is no “next page” link or button. That means that any results displayed here are only visible as long as there are no other results to push them off the page. This should have an impact on when and how often items are posted/tweeted to the stream.
Let’s get on with the experiments. These are by no means comprehensive or completely scientific. I just wanted to try a few things out, share the results and see what conclusions, if any, we can come to.
Experiment #1 – News
Here, I took a news story and tweeted it while preceding the link with a keyword-rich teaser.
Twitter account: Apartmentscom – Lots of followers (1000+) and perceived authority
Twitter Post: Amenities are important to choosy renters of apartments in Chicago: http://bit.ly/dvNQOT
Search Query: “apartments in Chicago”
Result:
A couple of things have happened here. First, notice the “apartments in chicago” phrase has been highlighted. Choosing keywords here is just as important as traditional search optimization. Also, notice how Google has expanded the bit.ly URL. This could not only be another place to include keywords, but also a way to catch a user’s eye and send them directly to a desired destination.
Experiment #2 – News Facebook Post:
I posted the same news story as above to our Apartments.com Facebook page:
Search Query: “apartments in Chicago”
Result: A few minutes after the Twitter post appeared…
The Facebook post almost immediately appeared in the SERP, above the previous tweet. Once again, the keywords are highlighted and the URL’s title is on full display.
Experiment #3 – Special offer
Twitter account: FindAnApt – no followers, first tweet
Twitter Post: FindAnApt New special offer on these apartments in Chicago: http://bit.ly/cpKaAZ
Search Query: “apartments in Chicago”
The important aspect highlighted here is that this account had ZERO followers. This tells me that Google isn’t using this as a ranking factor.
Experiment #4 – News
Twitter account: FindAnApt – no followers, 2nd tweet
Twitter Post: Apartments in Chicago need amenities according to renters: http://bit.ly/dvNQOT
I was able to recreate the results for Dallas Apartments as well:
However, when I tried posting another tweet for an apartment in Seattle, I was unable to get a result to appear. What did appear was this:
It’s a foursquare post to someone’s Twitter account. See a reason for encouraging your tenants to use foursquare now? I do. But that is another story…
Perhaps, I used too many similar, potentially spam-sounding posts too frequently and Google decided not to post include the third one. It’s possible but I can’t confirm it either way.
Conclusions
Based on our informal tests, here’s what we can infer:
- Google is only displaying real-time results on the first page if the topic is trending or highly newsworthy
- The real-time-only results page’s listings are transitory, meaning that one must be mindful of when or how often to post items that target a particular keyword phase.
- Choosing the right keywords to include in the post/tweet is an important ranking factor
- Number of followers on Twitter doesn’t seem to be a major qualifier to getting listed in the real-time results
- It’s possible that frequency and content could be an indicator of “spam” or “bot” tweets and could hinder inclusion in the results.
What might be the most important conclusion is that none of the URLs that made it into the real-time search stream received a click. Granted, only two out of three (Chicago and Dallas) actually made it into the stream. But given the transitory nature of these results I believe it’s illustrative of the type of mindfulness that must go in to creating a successful real-time search campaign.
I hope this little exercise has been helpful. I know I learned a few things along the way. I’m eager to hear about your experiences with real-time search. What’s working and what’s not working for you?
[...] Experimenting with real-time search [...]