SEO News

Posted by randfish

Late last month, Google made an update to its search algorithm that caused our MozCast to spike to an all-time high of more than 113 degrees. Our work as web marketers can be frustrating when we’re aiming for a continuously moving target. In today’s Whiteboard Friday, Rand covers how we can keep our cool and learn from those changes when they happen.

For reference, here’s a still image of this week’s whiteboard.

Video Transcription

Howdy, Moz fans and welcome to another edition of Whiteboard Friday. This week, I want to talk a little bit about how marketers should be reacting when Google makes big, dramatic changes in their rankings and their algorithm. Now, this can be a challenging topic, right?

So we’ve seen, for example in the recent past, MozCast, which is Dr. Pete’s project that monitors several thousand search results and sort of looks at changes in the top ten and what percent of them are churning in and out, and we saw one of the biggest spikes we’ve ever seen, bigger than Panda, bigger than Penguin, just hugely dramatic.

Dr. Pete represents those in MozCast with temperatures. So the average day temperature is 70 degrees. This one was 113 degrees. Very, very hot, meaning a dramatic amount of change. Lots of things in the first page of results on average moving out and being replaced by other things and lots of positions moving around too.

Now, the way I like to approach big algorithm updates is to look at, number one, what happened? What actually changed in the results? Because sometimes a dramatic variety of different things can be happening. So we see through MozCast and through monitoring lots of search results ourselves, for ourselves and for campaigns that we pay attention to, we can see that you’ve sort of got one, two, three, four ordering. That might shift over to be, oh wow, look. Almost everyone who is in the first page of results kind of fell down or fell out of those results, and now it’s number 11, 19, 4, and 16 that are ranking in there. Wow, okay. That was a big algorithmic shakeup. Push a lot of people down, a lot of new people in.

Or it might just be a reordering. So, one, two, three, four went to four, two, six, eight. Well, okay. I mean, two and four are still in the top four. Six and eight are still in the top ten. But we’ve had some bouncing around. So this is a shift, but not nearly as dramatic as the prior one, and actually MozCast temperatures represent that because Dr. Pete looks at sort of where things are shifting to figure that out.

Or, and we also see a lot of this, Google has introduced new types of results. There’s now a carousel at the top. There are now news results going in there. There are other things that are pushing results off of page one that are shaking things up, that are making things dramatically different, that are making essentially organic visibility quite different from how it used to be.

Those different types of results are of a vast variety, and Google rolls them out in tests all the time and then permanently when they like the results of those tests. Now, if you’re observing these patterns in the change of types of results and observing the patterns in what’s rising and falling, this can really help you get to the bottom of, “What should my strategy be? What tactics should I take?”

But the second question that I want to take you to before we get there is: What is Google saying about the update? Sometimes Google is very quiet and they don’t say anything, and sometimes they’ll give some information. Right?

So, for example, Google mentioned with regards to this big update that happened recently that there’s a rolling update going on, meaning you can see spikes in values potentially over a period of time as they roll out the update, and it will be ending on or around July 4th.

Okay. That’s potentially very interesting information. That might tell me, “You know what? Before I do a big, wholesale analysis of how this impacted me, I’m going to wait for this whole thing to roll out. Let me just give it a few more days, wait until the 4th of July and see what actually happens at the end of the shakeout.” Gianluca Fiorelli asked Matt Cutts, he said, “Is this a global update or just U.S. or English results only?” Matt nicely replied, “Well, it’s global.”

So that is also helpful to observe and to know so that people can get this sense of, “Oh, wow. I’m targeting mostly Spanish language search results in Spain or in Mexico, or in South and Latin America. I guess I should be paying attention to whatever is going on with this update.”

Third, I like to ask, “How has this update affected me?” Of course, because I’m a marketer who observes broad trends and runs a software company in the field, I like to see what those broad trends are and know about them. But I also really want to see how it affects me, and as a search marketer, that’s certainly what you should be thinking about, too.

So being able to monitor this through data is really important, and there are three points of data that you can collect from your own analytics. Those are the number of pages that receive one or more visits from Google search, the number of keywords that send one or more visits from Google to your site, and the total amount of Google search traffic that you’re receiving.

Then, if you want to get more granular, you can go down to the keyword level and look at what are individual keywords sending. Of course, remember that because of “not provided” a lot of that won’t be trackable anymore, which is frustrating and challenging.

Then the last thing that you’re going to need in order to see how this has impacted you is ranking position. So I like to collect rank position data in non-personalized, non-geographically biased results. This is not perfect. A lot of people are geographically biased, are searching on mobile phones or devices that are location-enabled, do have Google accounts that are biasing them personally. But this is the best that we’re going to do, those non-personalized, non-geo biased results.

You can achieve that by going outside of your country code. So for example, if I’m in Google US, I’m going to go search “Google.co.uk/search?q=” whatever keyword I’m tracking, “&gl=US”. That will bias me back to the U.S., but taking me to the U.K. and then saying U.S. will make it so that I’m not geo-personalized to just Seattle or just Washington, or just wherever I happen to be on the road where I’m searching.

Using “pws=0” will help remove personalization. This actually removes most of the personalization anyway. If you want, you can also log out or use a browser window that is non-personalized where you’re not logged in. From this, you get the best picture we can really get as search marketers about what’s going on and how the shift has impacted you, and you can see really different things.

I mean, if I see that my rankings haven’t really changed, but the number of pages that are receiving one or more visits from Google has dropped dramatically and that’s affecting my overall total traffic, I can presume, “Hey, you know what? This is probably an indexation problem for me.”

Whatever update Google has been making, the way it’s affected me is that I’ve lost pages that used to be in the search results. I’m no longer performing for them at all, and they weren’t the ones that I was tracking. So probably it means my long tail is where this is impacted, and so that can inform my strategy and my tactics from there.

This is the last question that I like to visit whenever something like this has happened which is: Are there actions that I should be taking? Not just what actions, but are there actions? Sometimes I just kind of go, “Hey, it’s cool. I’m going to let Google do what they’re going to do, and I’m going to do what I’m going to do. I’m not going to worry about them.”

But sometimes there are tactical actions like, “Hey, you know what? I need to bolster some individual keywords. We lost rankings on some keywords that are really important. Let’s see if maybe we should produce new pages of content. Maybe we should update the existing content. Maybe we should redirect the old ones to the new ones. Maybe we should be trying to earn some new links and social signals and shares to that stuff, whatever that might be.”

Or there might be more strategic level SEO types of things like, “Man, Google just introduced this big carousel across all these different types of hotel and travel results. I’m not sure that keyword phrase of city name plus hotels or city name plus places to stay is really going to help me anymore. Maybe I should start to consider whether I need to go earlier on in the keyword search funnel.”

Maybe I need to get in here where people aren’t yet searching for hotels, but they’re searching for destinations or places, or those kinds of things, rather than targeting down here where it looks like Google is kind of dominating the search results themselves. That’s a big strategic kind of shift that you’ll have to make with your content and your website and your keyword targeting strategy.

But being able to ask these questions, all of them, and then getting down to the tactical and strategic can really help make you more reactive in an intelligent, considerate way to the big changes that Google might be making.

All right, everyone. I hope you’ve enjoyed this edition of Whiteboard Friday, and we’ll see you again next week. Take care.

Video transcription by Speechpad.com

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!

Posted by Erica McGillivray

When we sat down in our big MozCon command room — think the Enterprise-D’s swank Observation Lounge — we knew we wanted to bring someone who’d worked for the Obama re-election campaign to MozCon. Why? Because no matter your flavor of politics, the re-election campaign was full of internet marketing brilliance and used big data to connect with its audience in a way that affected the world. This is what we’re all trying to do, right?

When we sent feelers out, Kyle Rush answered our call. He served as deputy director of frontend web development for the Obama for America campaign. Kyle’s currently director of technology at The New Yorker, and he’ll be speaking about how to “Win Through Optimization and Testing” at MozCon, July 8-10. You don’t want to miss this advice about testing and conversion rate optimization. You can follow Kyle on Twitter @kylerush and on his blog.

How did you come to work for the Obama for America campaign?

I’ve always been interested in technology since a young age. In junior high, I would spend all of my free time on the family computer making websites. In my second year of undergrad, I got really into politics and changed my major from advertising to political science. I followed the ’08 campaign closely and really admired how the campaign was able to innovate in the political field because political campaigns are notorious for deploying outdated and unusable technology, if any. After graduation, I worked for Blue State Digital which contributed to the ’08 digital effort. My boss Teddy Goff left Blue State to become the Digital Director at the Obama campaign, and I asked if he needed an engineer. Two weeks later, I moved to Chicago.

What did you learn most about yourself as part of such an intense experience?

A lot of political staffers will tell you that working on a campaign is like building an airplane while you’re flying it, and that’s very much the case. I moved to Chicago, where I had never lived, on a two-week notice. We quickly hired some engineers and built out a team. During the last half of the campaign, we worked seven-day weeks and not less than 10-hour days. We saw all the work we were doing play out on TV every day. There were extreme highs like when we had a 90% chance to win the election a few weeks before the first debate. There were extreme lows like when we lost the first debate badly. We worked 18 months towards one night, and then we won. I think that we all learned that the limits we thought we had don’t actually exist and that we can go as high and far as we want to.

What was your favorite win for the campaign?

The obvious favorite win is election day, which was one of the happiest days of my life. Aside from that, though, I would say the day we tested our “sequential” donate page. At that point, we had already optimized all the low-hanging fruit, and it was time to put some serious investment in a variation to try and beat our control donate page. We put in a lot of thinking, time, and effort, and we ended up beating the control by 5%. That was a huge win, because it taught us that even though you might think your page is optimized, there is always more you can do.

What win surprised you the most?

I don’t know if it qualifies as a win, but the Democratic National Convention was by far the most surprising thing that happened during the campaign. The convention went perfectly, but the amount of traffic we received to the website during the convention was unreal. We blew through our entire test queue for donate pages and were just coming up with tests to do on the fly while the traffic was pouring in.

Okay, for those us who are political nerds, did you get to meet President Obama? What’s he like?

We did get to meet the President! He came to the office several times during the campaign to talk and meet everyone at headquarters. For having the stature of the President of the United States, he is very human. The day after the election, he came and hugged everyone at headquarters. He is a great guy.

Switching gears from your past to your present, what’s inspired you lately?

Web performance is always really inspiring to me because deep down it’s what I care about on my engineering side. Specifically, I really like to read about the way that other engineers make web apps feel like native apps on mobile devices.

We’re all a bunch of data geeks. What are some of your favorite metrics to dig into?

I’m a performance engineer at heart so I really like to look at metrics around page speed. My favorite metric is time-to-paint, which is the amount of time it takes for the browser to do its first render of a web page. This differs from the pageload metric in that the browser often paints the web page before all of the assets have loaded. For this reason, time-to-paint is a more valuable metric to me.

As an engineer obsessed with UX, I also like to focus on metrics that quantify user frustration. On the Obama campaign, I spent a lot of time measuring any form error from a validation error on a certain field to the number of validation errors that occurred for each for submission. On our donate forms, we found that the people most commonly had two errors on their form submission which were the employer/occupation fields. Then we wrote a script that measured if people were entering any data in them at all. Turns out people weren’t entering any data, and we soon came up with solution to fix this problem. Once we solved the problem, our error rates plummeted.

MozCon attendees love to engage on social media with speakers. What’s your favorite social media network?

Instagram is definitely my favorite social medium. It’s beautifully simple, and it doesn’t require a lot of attention, unlike Twitter. You can go a week without opening it and not feel like you missed a lot.

Finally, for some pure fun, what music have you been listening to lately?

I like pretty much all music, but I really like upbeat/dance-y music. Lately, I’ve been listening to Kanye’s new album Yeezus non-stop. Typically though, I listen to house music at work and while working out. Over the past few days, I’ve been listening to songs like “Boy Oh Boy” by Diplo, “Play Hard” by David Guetta, and “Alive” by Krewella.

Thanks, Kyle, for sharing with us. Can’t wait for your talk at MozCon. Follow Kyle on Twitter @kylerush and on his blog. We’ll see the rest of you there!

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!

Posted by ChristopherFielden

This post was originally in YouMoz, and was promoted to the main blog because it provides great value and interest to our community. The author’s views are entirely his or her own and may not reflect the views of Moz, Inc.

Creating timeless content is something all SEOs should aspire to do. Why? When placed in front of the right audience, amazing content is highly likely to generate ongoing interest, engagement, links, and traffic, leading to increased sales/conversions and brand awareness. These results tend to make all but the most difficult client quite happy.

Image by Dominic Alves

In 2012, I decided to undertake an evergreen content experiment. I created a piece of content that I planned to update regularly over an extended period of time. I was in this for the long haul — I wanted to keep working on this content for at least a year. The aim was to see if putting ongoing effort into one page on a website would prove more efficient than spending time planning and creating multiple pieces of fresh content.

Common content performance patterns

Many creative content campaigns produce spikes of interest when they’re released and then dwindle in popularity. If you’re nodding your head in agreement, this might look familiar:

Creative campaign referral traffic spike, taken from Google Analytics

When shared, you see a brief spike in traffic, and then visits decline. This example is taken from the release of a well-received infographic that saw a lot of visits when it was shared on Reddit.

A spike isn’t always a bad thing. At the last count, this example generated over 35 decent quality links (ranging from DA 30 to DA 82) and thousands of social shares. This is a good result, but I wanted to try and create something that saw continued growth in traffic, engagement, and links over time rather than a spike.

Evergreen experiment

So I could share the results openly without contravening any client confidentiality agreements, I decided to conduct the test on my personal website. I write fiction, and I originally created my website to showcase my short stories. I launched the site in October 2011.

Image by Rose Craft

I’m not famous. No one knows who I am. No one found my writing, because no one was searching for it. Aside from friends and family, few people read my stories. Boo hoo.

In an attempt to gain an audience, I decided to try and make my website useful to the short story-writing community (people who write short stories also like to read them). I’d spent a LOT of time researching short story competitions to enter. I’d found a few decent resources, like Booktrust, that listed some writing competitions, but none of the lists or calendars were exhaustive or kept very up to date, and many of them didn’t list the full range of details I was interested in (closing dates, prize money, word count limits, genres, publishing opportunities etc).

So I decided to create an extensive short story competition list on my website.

Research

I was fairly certain, given the amount of competition lists in print magazines and the amount of writing websites I’d found, that there would be an audience for this type of content. To be certain, I did some keyword research.

There was an audience. Further research showed there was a large amount of long-tail keyword opportunities.

So I created the page, initially listing details of approximately 50 writing contents. The list went live during April 2012.

Page content

The page format is fairly simple. I started out with two tables, one listing regular writing competitions (monthly, quarterly, triannual and biannual) and another listing annual contests. Over time, I’ve added more tables so the resource is as easy to use as possible.

At the top of the page I openly invite users to contact me to have writing competitions listed. I also invite users to let me know if any of my details are incorrect, out of date, or if they find any broken links.

Use of outbound links

Again, to make the resource easy for writers to use, I’ve linked to all the competitions I’ve listed. I’ve read all sorts of discussions regarding outbound links and whether it’s best for them to be follow or no-follow, as well as discussions about how many links you should have on a page alongside concerns about the quality of the sites you link to and whether that has any impact on SEO.

As there doesn’t seem to be a definitive right or wrong way to do this, I decided to ignore all these concerns and just link to the most useful page on the different competition websites for the user. The only exception is when I link to a competition website that updates its URLs each time it updates the competition details. In this instance I link to the homepage to avoid excessive administration and maintenance of the page.

All links are followed.

Page maintenance

Image by Abhisek Sarda

From the day the page went live, I decided that I was going to display the date the content was last updated prominently at the top of the page. I wanted users and search engines to be able to see that the page was cared for and updated regularly.

I’ve read many arguments against using dates. This is usually because time constraints mean webmasters can’t update content regularly and the date often has the opposite effect, showing how out of date the content has become. But as I knew I’d be updating the page regularly, this wasn’t a concern.

I update the page at least twice a month, sometimes as frequently as twice a week, depending on how much time I have available.

On average, one competition contacts me a week, asking to be added to the list.

I respond to the vast majority of comments, either privately via email or as a comment, depending on what seems most appropriate given the subject matter.

Technical notes

My website is pretty basic. From a technical standpoint, I have ensured that the menu structures and URLs made sense and that my authorship has been setup correctly. Aside from that, all I’ve done is generate content. I’ve purposely kept the amount of pages on the site low, only adding new pages when I have to. At the time of writing, the site has 36 pages.

No linkbuilding

While undertaking this experiment I haven’t done any active link building at all. Any links the website has gained have been natural. Likewise, I haven’t undertaken any outreach. I have only engaged with writers and competition administrators that have approached me directly.

I did this to see how well the page could perform naturally, with internet users initially finding the content via organic search. Over time, this has led to natural interaction through comments, social sharing and links (and the unavoidable plethora of spam comments in my inbox). But I haven’t actively pushed the content. The results have come from natural content discovery and users outreaching to me.

Results

Traffic

This first graph shows the growth in traffic to the entire site from all mediums since launch in October 2011:

Traffic from all mediums to entire site from October 2011 to May 2013

Below is a breakdown of the figures from the different mediums:

The second graph shows the visits from all mediums to the short story competition page from its launch in 2012:

Traffic from all mediums to short story competition page from April 2012 to May 2013

Since its launch, the short story competition list has accounted for 67% of all the visits landing on my website (total entries to all pages are 77,374 — page entries to the competition page are 51,861). Full details of growth in visits to the page from all mediums can be seen below:

Visits have increased substantially since the competition list was launched. The dip we can see in April and May seems to be due to seasonality. The page still ranks well for a wide variety of long-tail phrases, and the New Year and autumn are seasonal peaks in writing-relating searches — admittedly, this is a generalisation, but as the site only launched in 2011 I don’t have a great deal of data to work with.

If patterns follow those of last year, I’d expect to see a rise in traffic in September.

Amount of search terms

10,728 search phrases have been used to find the page through organic search.

Most popular search terms used to find the short story competition page

Given that ‘(not provided)’ accounts for 30% of these searches, it’s safe to assume that the figure is actually substantially higher, so there is a lot of long-tail search involved here.

The large word count of the page copy contributes to this. At the time of writing this post, there were 11,632 words of copy on the page, of which user comments account for 3,463. At the time of writing this post, there are 66 comments on the page, some of them replies from me.

Social shares

The total amount of social shares to date is 127:

Details generated using Shared Count

I find that writers will often share the page on Facebook and Twitter, as will administrators of the competitions I list, if they run social profiles. Since the beginning of 2013, I have seen the share counts rising more rapidly, which I would expect given the large increases in traffic the page has seen when compared to last year.

Links

You can see details of the links that have been attracted below:

Data taken from Majestic SEO

Results from Moz’s Open Site Explorer

The volume of links isn’t huge. But this project is aimed at slow growth, and I haven’t actively asked anyone for a link. I want links to be entirely natural, only coming from those who think the content is worth linking to of their own volition. The only exception I can think of is me writing about the experiment.

As the resource becomes more widely recognised, I would expect the amount of links to increase accordingly. Recently, I have received my first university (.ac.uk) link, and started to receive correspondence from university lecturers who are involved with creative writing courses, asking about writing opportunities for their students (which led to me adding the ‘Writing Competitions for Young Writers & Children’ table to the page). This bodes really well for the future, as relevant university website links are likely to help the site’s performance greatly. And this kind of natural link building should make my backlink profile Penguin-proof long into the future.

I guess the key point here is that it’s taken almost a year of developing this content to start gaining links of this quality. Now that a handful of lecturers have found the site and started using it, it’ll be very interesting to see how the link results fare over the next twelve months.

Hmm, I feel another blog post coming on in the not so distant future…

There are a couple of other points to bear in mind:

1. I’ve done this work in my spare time, around work and other commitments. If you had the time to focus fully on projects of this nature you could probably generate these types of results far more quickly.
2. The links generated have been entirely natural as I haven’t actively asked anyone for a link.

Point 2 proves that detailed, focused content can work in its own right. You don’t have to outreach and link build to see some level of success.

Does this type of content help conversions?

Due to the growth in traffic to my website, I have increased my audience and engagement with my site. I’m beginning to be recognised as a thought leader (and a brand, I guess) in my niche area. Users have started to approach me with all manner of queries. I also receive frequent requests to proofread other writers’ work. If I had more time, this is a paid service I could consider offering in the future. So producing the content has revealed business areas I could expand into.

Ultimately, all the extra traffic has led to a rise in the number of people buying the book I sell through Amazon and Lulu. I now sell a few a week, compared to one every couple of months.

So, in answer to the title above, ‘Yes.’ I am getting what I wanted — a wider audience for my writing.

Amount of referrals other sites receive

Below you can see the amount of referral visits my page generates to other websites:

Referral traffic received from my competition page between January 2013 and May 2013

One of the writing competitions I list was kind enough to share this data with me. They were first listed on my site in January 2013.

A breakdown of figures can be seen below:

The highly relevant traffic I can offer writing websites makes being listed appealing to most competitions. From speaking with the administrators of the competition in the example above, I know that the traffic also converts well into competition entries, so they are very happy with the results related to me listing them.

This means that when I receive enquiries I can be confident in the value my list offers.

Summary

So far, this experiment has proved that investing time in creating content that is updated regularly can bring excellent results. In 2013, the page attracts between 6,000 and 9,000 visits a month, 22% of which return to the page time and again.

All you need to emulate this is some vision and common sense:

• Find something your target audience wants
• Give it to them
• Keep the content fresh with regular updates and improvements
• Listen to user suggestions and make changes accordingly
• Listen to user suggestions about other resources they might find useful, and create them

That’s a content strategy that is likely to keep me busy for the next few months and generate excellent results.

Keep it simple

One of the more common mistakes I’ve seen SEOs make is developing content no one is interested in. You might end up creating something sexy based on an amazing concept, but will it actually gain you the result you or your client wants to see? Sometimes the more mundane ideas, like generating a useful list, can work far more effectively. It might not be sexy, it might not look awesome, but it is useful and can appeal to a community.

Keep it simple.

I believe you can learn more from those three words than you’d like to believe. Looking forward to hearing your thoughts in the comments.

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!