Posted on February 14th, 2011 by John
A few clients sent me yesterday’s great NY Times feature story on JC Penney’s underhanded SEO techniques, so I figured a blog post was in order. The Times uncovered huge numbers of spammy links pointing to JC Penney’s site, which it suggests helped it achieve #1 Google rankings on a number of keywords for months. While probably the most damning, this was only the most recent in a bad few months for Google in the mainstream press and the SEO community:
- In November the Times ran another fascinating front-page business section story on Google search results, this time focusing on an online eyeglass retailer who purposely treats his customers terribly because he’s found that their negative online reviews (with links) help his organic rankings. He goes so far as to threaten customers and post things like this on forums where negative comments appear: “I just wanted to let you guys know that the more replies you people post, the more business and the more hits and sales I get. My goal is NEGATIVE advertisement.” Danny Sullivan, possibly the most respected expert on SEO, told the Times “it’s fair to say this is a failure on Google’s part.”
- Two weeks ago the Washington Post featured a story titled “How you and Google are losing the battle against spam in search results.” Among other things, the articles points out that one measure of Google’s success rate — the percentage of people to visit a site after a search — fell 13% last year. Internet pioneer Tim O’Reilly told the Post “It’s clear that Google is losing some kind of war with the spammers. I think Google has in some ways taken their eye off the ball, and I’d be worried about it if I were them.”
Meanwhile, a number of prominent SEO experts have been offering their own in-depth criticisms, and pleading with Google to stop rewarding spam-like techniques. Like anyone who practices good SEO techniques like rigorous keyword research and on-site content optimization (often called white hat SEO), I have long been frustrated and perplexed by many of the commercial search results I see on Google — and by the ability of companies using bad SEO tractics (black hat SEO) to achieve success — so many years after these tactics became common knowledge in the search community. The recent string of bad buzz is, in my view, mostly good thing. I’m optimistic that it will help:
- Embarrass Google into putting more resources into fighting bad SEO tactics
- Decrease Google’s emphasis on links — the main vehicle for bad SEO
- Increase Google’s use of other, less “game-able” factors. Bruce Clay, for instance, predicts huge growth in the importance of social media “likes” and local search results, which he thinks will increase from 20% of all search results to 50% by the end of 2011 and 70-80% by the end of 2012
- Cause a lot of large and mid-size companies to back off link spam and question their SEO consultants about it — for fear of getting punished like JC Penney. I suspect a lot of online marketing managers in the Fortune 500 got emails from executives today saying something to the effect of “we aren’t doing this stuff are we?”
- Help efforts to establish a credible third-party designation for good SEO consultants
The bad news of course is that the buzz will encourage more bad SEO activity by smaller companies willing to take the risk who didn’t previously realize how effective it is. Bad SEO companies won’t be going out of business anytime soon. But hopefully all this press will help to start a shift in the tide.
Posted on February 8th, 2011 by John
This is part 3 in a series of blog posts capturing our SEO makeover for the Melete Foundation.
On any SEO project, before you jump into core tasks like keyword research, content optimization, and link building, it’s important to learn as much as you can from existing data sources. This could be Google Analytics or another web traffic tracking tool, as well as AdWords or another search advertising platform.
Your goals with “pre-SEO analysis” are, in order of importance, to …
- Identify anything that will help you do smarter keyword research and search optimization
- Create a baseline against which to measure your post-SEO results
- Gain familiarity with the site’s overall metrics and trends
Melete has had Google Analytics running since September of 2010. I spent about 90 minutes analyzing and summarizing their data, focusing on the traffic sources reports. You can read my pre-SEO analysis for Melete here. As I note in the report, to date they’ve had almost no search traffic from the types of target keywords we identified in our kickoff meeting brainstorm. The vast majority has come from brand searches (e.g. “melete foundation”), people searches (e.g. staff names), and generic searches related to songs mentioned in blog post titles (not a good match).
If I had found traffic from our target keywords, I would have tried to break those keywords into buckets (e.g. teacher-related phrases, volunteer-related phrases) and looked at their respective levels of engagement on the site (bounce rate, time on site). I would make a note of which keyword groups perform best and worst on these measures — information that would help us as we prioritize target keywords during keyword research, our next step.
In the absence of this kind of data, this exercise mostly serves to reinforce the need for rigorous SEO on this site, and to act as a baseline going forward.
Up next: keyword research. See posts on all of our Melete SEO makeover steps so far.
Posted on February 3rd, 2011 by John
This is part 2 in a series of blog posts capturing our SEO makeover for the Melete Foundation.
Following our kickoff meeting, the next step in Melete’s SEO Makeover was to optimize their web tracking. Since measurement and analysis are key to the process, it’s critical that you set up and optimize tracking early on. In this post, I’ll share the process I went through to do this.
Fortunately Melete already had Google Analytics (GA), our preferred tracking tool, in place when I started working with them. When they set their site up on WordPress, they added a great Google Analytics plugin from Yoast and have been receiving data since September of last year. Still, there were 2 critical pieces we needed to ensure that our data is accurate and actionable: goals and internal traffic exclusion.
Our aim with SEO isn’t just more website traffic, it’s more high quality traffic that takes desired actions. The “goals” feature in Google Analytics — frequently called “conversion goals” — allows us to identify these actions and integrate them into our analysis. The most common goals we use are 1. a visitor viewing or hitting a key page on the site (e.g. a purchase thank you page) and 2. a visitor spending a minimum amount of time on the site before leaving (e.g. 2 minutes).
With goals in place, we’re able to say that a site had, say, 200 leads or 450 engaged visits last month, which over time is a lot more meaningful than just saying they had 1,000 visits (what most companies focus on). But the really great thing about goals in GA is that you can tie them back to traffic sources. This allows us to be much smarter when we measure SEO results using Google Analytics; we can see which keywords send traffic that converts and which ones don’t, and then focus our future optimization rounds accordingly.
1. Identify potential goals
When I met with Melete we identified two key desired actions on their site:
- somebody clicking the Support (donate) button on their Support Our Work page
- somebody signing up for their newsletter, which can done at the bottom of any page via a Mailchimp widget.
Since their site traffic is light right now, and there aren’t many people taking either of these two actions, I identified two other “soft” or intermediary conversion goals that are likely to get more data:
- somebody spending 1 minute on the site
- somebody spending 3 minutes on the site
I chose 1 and 3 minutes after glancing through the Traffic Sources > Keyword report and looking at keywords that had only 1 visit. This allowed me to get a sense of how long individual visits last (vs. just looking at averages which can be skewed by one really long visit — I’m sure there’s a better way to find this). I saw that not many people last more than a 1 minute on the site.
2. Set up easy goals
The time on site goals were easy to set up. To set up the “1 minute” goal, I just went to the GA Settings (entry) page and clicked Edit.
Toward the bottom of the next page I clicked Add Goal.
Then I made the selections and inputs below.
And then I followed the same process for the “3 minute” goal.
3. Set up tricky goals
One big problem with GA goals is they can’t be used to track clicks on links or submit buttons that take a visitor to another site – what are often called “outbound clicks” or “external links.” And for a lot of small businesses and non-profits that take advantage of 3rd party solutions for newsletters and payment collection, these types of “outbound clicks” are in fact the primary goals on their site. That’s the case with Melete. Their newsletter signup form takes the user to a Mailchimp-hosted page. Their “Support” button takes the user to a PayPal-hosted page.
If you do want to set up events, the best place I can point you is this thread on the Google Help forum which I initiated a few months ago when I was struggling to get them to work for another project. Now I just use the code and process that Joe from Blast Advanced Media was kind enough to provide in that thread. Since both of the outbound clicks we wanted to track for Melete were in effect form submissions (the “Support” button doesn’t seem like a form but in terms of HTML code it is), I used the “onsubmit” variation of the code he provided, which goes inside the <form> tag.
One last note on this: if the outbound action we wanted to track for Melete had just been on clicks on links, rather than clicks on submit-style buttons, all of this would have been much easier. That’s because Yoast has integrated the heavy-lifting described above and allows you to track outbound clicks by just checking a box (see below — found under the GA Settings in your WordPress account). But at least in our experience, this feature didn’t pick up “submit” clicks.
Internal traffic exclusion
Internal traffic is a common cause of inaccurate data in GA. Internal traffic for your site includes you and everyone at your company, as well as vendors and partners who are likely to visit the site often. Especially when you are making changes to your site, you visit it often to test the changes. If you don’t exclude this data, you can end up with a very distorted picture of your site’s performance. (Thought it’s good to wait until after setting up goals and events to add these filters, because you may need to do some testing that you want to see in the reports; e.g. making sure event tracking is capturing your newsletter signups.)
For Melete, as with all of our projects, we use two techniques to exclude internal traffic: 1) cookie-based exclusion, which excludes traffic from individuals’ computers, and 2) IP exclusion, which excludes traffic from geo-locations. Since neither method is perfect, we use both.
Google Help has instructions on both methods here. The IP steps I think are a little out of date based on the way their interface now works, but they’re good enough.
To get the cookie exclusion to work, you need to add a page to your site. Once I did that, I sent an email to the Melete team explaining what I needed them to do: click on the link that exclusion page (to get the cookie) from each computer they use, and send me their IP addresses from each place they work, which they can get at WhatIsMyIP.com.
Once I received their IPs, I went to the same place under Settings that I went to to add goals, and then added Marketade’s and Melete’s IPs.
Setting up goals and internal traffic fitlers are very hard to do in one sitting, since you’re often relying on other people to click things or test things for you. So the best way to approach web tracking optimization is to break it up into smaller steps and try to tackle it over a few days.
Up next: pre-SEO analysis. See posts on all of our Melete SEO makeover steps so far here.
Posted on February 1st, 2011 by John
This is the first of a series of blog posts capturing an “under the hood” look at our SEO Makeover for the Melete Foundation. The first step in our process was a kickoff meeting. I visited Melete’s office in DC’s Dupont Circle and spent about 90 minutes talking with the team: Mija, Guy, Ben, and Heidi. We followed this agenda:
- About Marketade (15 min)
- Marketade’s SEO Process (15 min)
- About Melete / Positioning (30 min)
- Keyword Brainstorm (30 min)
For the About Melete (#3) part of the meeting, I asked questions like:
- What’s your elevator pitch?
- Who are your competitors, broadly speaking? Or what organizations do something similar to Melete?
- What are the things that make Melete different from these other organizations?
- If a donor was deciding between giving to you and one of these other orgs, what would you tell him to convince him to choose Melete?
They had great answers to these questions, and their passion for their mission and work came through. This exercise is critical because SEO is so intertwined with content writing. If you’re going to be rewriting and reorganizing a lot of a site’s content for SEO, our view is that you should improve the overall writing while you’re at it. Part of doing that is getting a good handle on an organization’s positioning. What makes them unique? What are the key benefits of their offering?
For the keyword brainstorm (#4) part of the meeting, I asked the team to throw out words or phrases that people might type into a Google and that are a good fit for Melete. Here were some of the ones they said:
- meaningful international experience
- community based international experience
- teacher development
- teach abroad
- working with teachers abroad
- fostering cross-cultural knowledge
- making world smaller
- citizen diplomacy in action
- educational citizen diplomacy
- local citizens
- summer teacher exchange
- summer cross-cultural exchange
- empowering young women
- middle east peace
The goal here wasn’t to come away with a definite list of keyword targets, but rather to give us a starting point for the later step of keyword research, which often doesn’t have the benefit of a meeting setting.
Up next: web tracking optimization.
See the full series of blog posts capturing our SEO makeover for the Melete Foundation.