You may recognize me from Twitter or other SEO communities around the web like SEOmoz. If not, nice to meet you! I run Evolving SEO in Massachusetts. Just like to thank Tom Critchlow for offering me the opportunity to guest post on the Distilled site. This is my first guest post ever. I hope I fulfill your expectations!

Tom Suggested I Write About This Google Search For “Fave”

It all started on a recent Wednesday. I went down to NYC for the Distilled SEO for E-Commerce Meetup. Tom and I were chatting. He had recently read my post on some weird IMDB results in Bing and thought I’d be gullible (ummm… I mean excited) enough to write about this strange Google search for “fave”:

What we both found strange about it was:

• The page had no page authority
• Google did not suggest a spell correction
• The amount of search volume for “fave” (you’ll have to view the slideshow for that)

I had a lot of questions:

• Did the result have something to do with “fave” being the beginning of a misspelling for “facebook”?
• Was there something else that SHOULD rank for “fave” that wasn’t?
• How COULD something else rank for “fave” if you wanted it to?
• And importantly, what can we LEARN from this and apply in practice?

I love these sorts of indexation mysteries, so naturally I thought it was a fantastic idea, thanks for letting me take the wheel on this one Tom!

Like my post on IMDB, this seemed to present best via slideshow – so get your forward and back buttons ready, and check out what I found! Its a nail biter! I’ve included some extra info and resources below:

Three Step Approach

I break my approach into a three step hierarchy:

1. Does this matter? – If there’s not many searches for “fave” or search intent is too generalized across users, than it doesn’t matter so much we’re getting weird search results.
2. WHY do things rank like this? – Once we know it matters, I want to know what’s causing it to rank like that? There must be something beneath the surface.
3. What can we learn from this? – Finally, it a cause is identified, how can we learn from this? What can we learn about the SEO research/diagnostic process? And are there any takeaways for leveraging the findings to apply to real life situations?

Links To Resources Used

Where Can You Find Me Next…?

Conferences

March 19 – 23 – SES NYC. I will be part of a sponsored session doing a live site review and a case study. Exact day/time TBD.

April 2nd, 2012 – I will be at LinkLove Boston / Atlantic as an attendee. You should come!!!

The ‘Net

I’ll be branching out to YouMoz! Starting with an update to this post Rand did a while back on more weird Google results. Didn’t ask him, so hopefully he won’t mind

Do I got straight into the code? Do I navigate around? Should I go straight into their backlink profile. I JUST DON’T KNOW!

Although these are all important, they may not give you a high-level understanding of the site and a clear picture of how the project should proceed. To tackle the “deer-in-the-headlights” approach I’ve put together 5 simple steps that will help you identify some common and crucial problems that are made. These steps are especially useful when you being speaking with a potential client and they want to know how you can help.

So away we go…

1) Duplicate Content

Why does it matter? Duplicate Content can be any sites downfall. When there is more than one piece of the same content, it can make it difficult for search engines to distinguish between the original source and mere copies. This will then cause the search engines to present less relevant information, and more importantly to the site, lower rankings and traffic metrics.

How to check?

- Copy a snippet of text
- Paste the text in Google (or your search engine of choice )

- Analyze results!
If you want to be sure there isn’t duplicate content on your domain (which you should definitely do!) combine the above method with a site search. By doing this, your query will only return instances of duplicate text from said site.

2) Check Robots.txt

Why does it matter? Utilizing robots.txt can be super helpful when you’re trying to block the search engines from crawling certain pages or folders. However, this function can be just as detrimental if you do not implement it correctly. To be certain that a site isn’t blocking themselves (I’ve seen this..a few times!) or any of their own important content or pages; check the robots.txt file of the website.

How to check? Super easy! Taking Distilled as an example:

3) View the Cache Version of the page

Why does it matter? You’ve heard of cloaking right?! Don’t do it. By showing different content or URLs to Google you’re asking for a world of pain to be brought down upon your rankings and traffic. My advice, don’t wake a sleeping bear (that’s a saying right?!)

This check is also useful because you can see the last time Google crawled a specific page. This is important because you want your site to be crawled as frequently as possible, while a long lapse in time can indicate a possible penalty or crawling issues.

How to check? Again, super easy!

4) Canonicalization

Why does it matter? Just like duplicate content, having multiple versions of the same content across multiple locations can decrease your rankings, spread out your link equity, or more simply devalue your site.

How to check? Easy as pie. Check all the common versions of the site are redirecting to the canonical version:

http://www.example.com

http://example.com

http://www.example.com/index.html

http://example.com/index.html

http://www.example.com/INDEX.html

https://www.example.com/index.html

If the page loads the same content without redirecting then you’ve got some canonical issues, my friend. Don’t get too panicked though, this is a basic mistake that can be easily rectified.

5) Top Page Report

Why does it matter? You can often find missed opportunities by looking at the top pages report in Open Site Explorer(OSE). By doing this, you can see which pages have the most links pointing to them, as well as the response codes. For a sites top pages, you want to ensure that they are in fact the sites “top pages” (AKA most important) that return the right response code. If the page returns anything but a 200 or 301 that page might not be passing valuable link juice.

How to check? Input the site into OSE and analyze the top pages, linking root domains, and HTTP status codes.

Get to it!

So get to it! Hope this helps and remember, it’s as simple as pie. Specifically, pecan pie. mmmm

Julianne Staino is an SEO Analyst at Distilled NYC. She loves learning new SEO techniques and creative approaches to link building.

Follow @JulianneStaino

Julianne Staino

Recorded in crystal clear HD format at our leading industry conferences, we have over 50 hours of some of the brightest SEOs out there (Rand Fishkin, Will Critchlow, Wil Reynolds, Richard Baxter, Michael Gray, Rhea Drysdale and Bob Rains to name but a few) giving away their best tips, tricks and research.

Check out the trailer below for a taster of what these videos contain:

So all you need to do is register your details below and you’ll get 3 videos in the first month, and then we’ll send you a fresh new one every month from then on. Nice!

To that end, today we have announced DistilledU.

(logo possibly in progress)

You can sign up here to know when we launch it to the public. But let me tell you a bit more about DistilledU.

The Vision

DistilledU has been the brainchild of numerous people at Distilled, including Will, Tom, and myself. We love teaching people, as is evidenced by our SEO Events and SEO blog, both internally and externally.

We originally dreamed up DistilledU for our own internal purposes as we grow and train new hires, but we decided to be as transparent and helpful as possible, so therefore we are going to release it to the public.

As many as you know, we partnered with SEOmoz a few years ago. We see DistilledU as being complimentary to SEOmoz’s tools and resources. SEOmoz helps you to do SEO on your site or your client sites, though some SEO knowledge is required to use these tools to their full advantage. DistilledU is being created to help you learn SEO so that you can make better use of the tools and other resources available to you.

What’s in DistilledU?

I’m so glad you asked! The TL;DR (Too Long; Didn’t Read) version is that DistilledU will be a place for everyone to learn something new about SEO, or to learn SEO for the first time.

We’re creating checklists, videos, guides, linking to external resources, and generally trying to provide as much value as possible to everyone enrolled in DistilledU. We’ll also have different levels of achievements for you to attain as you move through the lessons and modules.

What Topics Are There?

Today is the first day that we have worked hardcore on it, and we’ve built out the following modules in their version 1, possibly beta-ready form. They are:

• Keyword Research
• Information Architecture
• Onsite Optimization
• Technical

We’ll be building out more materials, guides, spreadsheets, templates, and more as we go and as you tell us what would be useful!

Here are a few shots of what you will see once we open up the private beta:

A module, this one about Information Architecture. Many will include videos.

Tick off your accomplishments as you work through the modules

Test your new knowledge with flipcards (notecards for the web, essentially)

The Launch Plan

We need a few more weeks to build out more content so that DistilledU is as useful as possible to you. Then we plan to open it in private beta to work out remaining bugs and gather user feedback. Then as soon as we feel DistilledU is war-ready, we’ll launch!

Watch this space for updates. We’ll be in touch

Sign up for DistilledU now!

John Doherty is an SEO Consultant in the Distilled New York City office. John's work time is filled with data consumption and strategic awesomeness, while his free time consists of extreme sports, travel, and bad reality TV.

Follow @dohertyjf

John Doherty

One of the most time consuming parts of content-based outreach is finding and qualifying prospects. Advanced search queries can return a large volume of opportunities, but you need to be experienced in putting together prospecting queries. Even then, you often have to cull through a lot of low quality sites once you get past the first few pages. Another approach is to use curated lists as the starting point for your prospecting. While these lists are more qualified, often they aren’t big enough to support a campaign. When you’re reaching out to bloggers, one way to address this problem is to start with a curated list of blogs and use blogrolls to expand it. Because blogroll inclusion can be a vote of confidence from a trusted person, it can be a great source for finding high value blogs.  That said, in order to find the high value blogs included in blogrolls, you need to deal with a couple of challenges:

1. You can’t tell how relevant or valuable each of the blogs in a blogroll is without reviewing each one (this is particularly the case now that blogrolls have fallen out of favor for many verticals).
2. Blogrolls often include a mix of blogs that are focused on the niche and ones that are completely unrelated.  So until you dig in, you don’t know if the blog was added because it’s a go-to resource for this blogger or because it’s a client’s blog, friend’s blog, etc.

Without a process and tools, trying to take a set of blogrolls and quickly isolate the relevant niche blogs in blogrolls is time-consuming. Fortunately, with a handful of free tools and a bit of ingenuity, you can take a set of blogrolls and quickly identify the ones you should research first.  If you can focus first on the ones that appear promising, you can save a boatload of time.

Here’s the basic approach:

1. Create a seed set of blogs by extracting blog URLs from a public blog directory (e.g., alltop) or a curated lists of “top blogs”
2. Use a blogroll discovery tool to expand the list,
3. Create Excel pivot tables to find blogs that are included in blogrolls most frequently, and focus your research on them first (i.e., use co-citations as a quick-and-dirty proxy for measuring influence).

Let’s dive into an example to show you how this works.

Step 1: Build your seed list

Let’s say I’m trying to find out which blogs are influential in the internet marketing space.  There are lots of sources I could rely on, but for this example, I’ll use TopRank’s Big List of Search Marketing Blogs.

The Big List is a great source for a seed because it’s curated in a thoughtful way and because it’s, well, big. Not all niches will have a definitive source like this, but you can often find good pages on Alltop, or you can merge lists from multiple sources – often times a simple search for “best x blogs” can unearth some good lists, or you can look for badges on influential blogs in the niche and that will lead you back to a comprehensive list (doing this in the Internet marketing space would lead you to The Big List, The Ad Age Power 150, and Invesp’s Blog Rank). Mining general directories like DMOZ can also be very effective.

So we’ve identified a great source for our seed list, but now we need to get this list of URLs into a CSV format.  There are online apps for extracting links, including iWebTool’s Link Extractor and BuzzStream’s link extraction tool (I’m one of the founders of BuzzStream). You can also accomplish this by using a Chrome/Firefox extension like Web Developer to grab the list of URLs on a page.  Since I’m most familiar with BuzzStream’s tool, I’ll use it for this example:

1. In Firefox, highlight the section on the blog that includes the links you want to capture.
2. Right-click and click “View Selection Source”
3. Copy-and-paste the selection into the Link Extraction Tool

This gives us a CSV that includes all the URLs in the Big List and the domains for each of them (385 blogs).  This is our seed list of blogs.

One thing to note is that, depending on your source and the tools you use, you might need to clean up the CSV a bit.  Look for URLs that are clearly improperly formatted, links to share buttons, etc.

Step 2: Use blogrolls to expand the list

Now that we have our seed list, the next step is to expand it by finding the list of blogs that are included in the blogrolls of these list members.  To do this, we’ll use the free Blogroll List Builder tool:

• Paste the list of URLs in your seed list into the blogroll tool and click ‘Go’
• Give the tool some time to run…there’s no feedback mechanism in the tool right now, so you’ll need to check the CSV against the list shown on the page to know when it’s done.  Once it’s finished, click “Download as a CSV file”

So this gives us a list of all the blogs from our original list and all of the blogs included in their blogrolls.  If you’re feeling ambitious, you can feed this new list back into the Blogroll List Builder tool to find 2nd level blogroll members (I’ve found that this improves results significantly).  We’ll skip that for this example.

We’re almost ready to find out which blogs are included the most frequently in blogrolls, but first we need to clean up the data a bit.

Step 3: Clean up the blogroll list

The first thing we’ll do to clean up the data is remove blogroll members that are included multiple times in the same person’s blogroll.  For example, searchenginecollege.com is included in ask-kalena.com’s blogroll 15 different times.  If we leave all of these in the dataset, the pivot table is going to overstate how often this blog is included in blogrolls.  So to prevent this, we’ll create a formula in excel that will help us find the duplicates and then delete these rows:

• In Excel, click Data>Sort.  Sort first by ‘Input URL’ and then by ‘Blogroll Domain’
• In cell E1, put in the header “Multiple Inclusion per Blog’
• In cell E2, add this formula – =IF(C2=C1,E1+1,1) .  What this does is check to see if the domain for the blogroll member is the same as the domain in the previous row. If it is, it adds 1 to the count.
• Paste the formula down to the bottom of your list.
• Select all of the cells in column E that have formulas in them, right-click and select copy, right-click again and select ‘Paste as Values’.  This will replace the formulas in these cells with the actual numbers that the formula returned.  If you don’t do this, you’ll have problems when you start deleting rows

• Scroll through your data to find any cells in column E that don’t have a value of 1 and remove these rows.

The other thing we’ll do to clean up the data is remove any rows where the blogroll member is the same domain as the source blog (i.e., the XYZ blog includes xyz.com in their blogroll).

• In cell F1, put in the header “Same Blogroll Member as Source”
• In cell F2, add this formula – =IF(LEFT(A352,20)=LEFT(B352,20), “match”,”").  This isn’t a perfect formula, but it’ll help you find ones that appear to be the same (i.e., the blogroll URL has the same first 20 letters as the input URL).
• Paste the formula down to the bottom of your list.
• Again, select all of the cells in the column that have formulas in them, right-click and select copy, right-click again and select ‘Paste as Values’.
• Scroll through your data to find any cells in column F that say “match.”  Review them to make sure that they’re an actual match and remove these rows.

Step 4: Create a Pivot Table to Find Blogs in the Most Blogrolls

Now let’s make use of our excel ninja skills and create a pivot table that shows us how many times each of these blogs show up in a blogroll.

To start, click Insert>Pivot Table

The data you created should be selected, and in the section titled “Choose where you want the pivot table report placed,” keep “New Worksheet” selected.

Click OK.  Your screen should look like this:

On the right-hand side, drag ‘Blogroll Domain’ into both the ‘Row Label’ and ‘Values’ box:

In the ‘Values’ section, it should say “Count of Blogroll Domain’.  If it doesn’t click on the dropdown for the item in the ‘Values’ section, select ‘Value Field Settings and change it to ‘Count’.

You’ll now have a table with two columns.  The first column shows the blog and the second column shows how many times it’s included in blogrolls.  To sort this list, simply select any cell in column B, right-click and select Sort, largest to smallest.

So from our original list of 288 blogs, we’ve found 483 more blogs in the first level of blogrolls.  Of these, 65 show up in at least 2 blogrolls.  The ones with the highest count will usually be multi-author, broad category blogs.  To find influencers within these blogs, you can usually find the specific tag or category you care about in the blog, find out who writes about that topic for the publication and search to find out where else they publish.  Beyond the big publications, you’ll start seeing more of the single author blogs that cover a fairly broad category and beyond that, you’ll start seeing more of the “sub-niche” blogs (e.g., people in the SEO community who are heavily focused on link building, outdoor adventure bloggers who are specifically focused on rock climbing, etc.).

That leaves over 400 blogs that are only included in a single blogroll.  Should you ignore these? Definitely not…you’ll miss out on lots of relevant, highly valuable sub-niche bloggers in this list. There are other tools you can use to quickly identify the good sub-niche blogs and to filter out the ones that are obviously unrelated.  But that’s another post entirely…

In a world where Google’s idea of ‘Article Freshness’ is becoming an increasingly important factor in determining a page’s value it is critical that blogs remain current, timely and on topic.  Posting a relevant news item on your blog isn’t just one of the easier posts you’ll ever compose, it will also demonstrate to your users that you have your finger firmly on the pulse of your particular niche.

Now, I know what you’re thinking: “Jacob, who do you think I am, Walter Conkite?  I don’t have any insider information or access to my own private news chopper!  Who would possibly come to my site’s blog for their news?”  Rest assured I’m not suggesting that by posting a few timely news articles your blog could necessarily become the news authority on “cars” or “boots” or “widgets.”  But I’m here to tell you that a well-timed, well-pumped, under-reported news story can nab you a ton of links for a minimal amount of effort.

There are several ways to generate news related to your area of expertise.  The first and most obvious option is to simply skim a highly targeted news aggregator for under-reported stories that may be of interest to your niche of readers. There are often several stories with only one or two sources (more mainstream sources have hundreds) that you could easily cite and reblog for a quick, informative, timely post rife with target keywords.  I’ve received several links from blogs citing my news blogs’ articles along with the original source if they happened to read the story on my page first.  You can also follow the major blogs in your field by grabbing their RSS feeds or letting sites like alltop do that for you.

To that end you don’t have to be first, second or even third with your news.  You can acquire links, retweets, Facebook shares and +1′s by simply being one of the several sites that happen to be reporting on a news story.  Followers of your blog may not have seen the news item and depending on the juiciness of the scoop they’ll be more than willing to pass the knowledge along in the form of a social media share or better yet, a link.  This is where strong relationships with bloggers, news outlets and other influencers will be worth their weight in gold.  You can even push the article to other news sites that haven’t yet reported the news.  Send them a friendly, personal email with your link and you might be surprised at the responses.

Another option for news generation is to rework the mundane title of a news post to speak more directly to your particular area of focus. Imagine a massive news organization conducts an interview with Sean Bean and in that interview he reveals that he will return in Game of Thrones Season 2 (keep dreaming, people).  The news organization’s title is probably something of a snooze-fest such as “Our Exclusive Interview with Sean Bean.”  As a web marketing specialist and Game of Thrones fan you think to yourself “Wouldn’t a more appropriate and link-bait-tastic title for this article be something more like “Sean Spills the Beans on Game of Thrones Season 2″.  Of course you can’t just copy & paste their content on to your blog but it’s perfectly acceptable to quote the most relevant segments, write your own beefy commentary, and properly cite the original article all under your new (better) title.

“Okay, Jacob I’m willing to add a news article or two to my monthly blogging regimen.   I mean, it sounds easy enough and if it can get me links and shares once in a while it’s certainly worth the easy post.  But how do I stay ahead of the curve?”

That’s the easy part my fellow blogger.  In the past I’ve simply kept a few bookmarks to various Google News search queries on my toolbar.  For example a car dealership’s blog may want to stay abreast on the latest car news.  Pull up a Google News search for “Ford” or “Lexus” and skim for interesting tidbits that you wager others may find interesting enough to share.  A good rule of thumb is to ask yourself whether or not you, as an renowned blogger in your field, think that this story would be interesting enough to pass along to a friend.  You can narrow your news search within Google to just blogs to find more obscure items to reblog or spin in whatever direction you see fit.  You should probably be following the latest news in your industry anyway so if you’re already doing so than this part should be particularly easy.  If your blog becomes newsy enough you should even consider submitting your site for consideration in Google News yourself.

To be perfectly clear, I would never recommend simply copying an article you’ve found on another site.  In fact, that’s a rather horrible idea.  But if posting news that has already been reported by another news agency is bad form then why are there 6317 articles on last night’s Giants game?  Or 1647 articles reporting Heidi Klum and Seal’s divorce?  The internet is an echo chamber of information and as long as you properly cite the original source and/or the sources you used to form your site’s take on the news then your white hat will remain firmly in atop your head.

‘Freshness’ is a relatively new term in the SEO world.  Google is now using recent events and hot topics to help determine which sites will be taking the top slots in SERPs.  Adding news articles to your blogging rotation will send signals to the big G that your site is timely, relevant and fresh.

Posting timely news articles isn’t just a simple way to add relevant, fresh content to your blog, it also has the potential for generating links, tweets and shares due to the immediacy of the moment.  You’re also providing a valuable service to your users.  They don’t always catch every news item but you, as an authority in your area of expertise probably will so why not leverage that knowledge by reporting on relevant, on-topic news items?

Breaking News: The offices of Distilled USA have convened at the company’s corporate headquarters in London for what has been affectionately referred to as ‘Distilled-a-thon’ 2012.  The New York and Seattle teams have joined their London colleagues for a week of team building, collaboration, and to continue our never ending quest to perfect ‘The Distilled Way’ of web marketing.  If you happen to be in the London area and want to chat just stumble into the closest pub and there’s a good chance you’ll run in to one of us.

If you’re a talented, smart, web marketer who knows how to ‘get shit done’ and you want to be a part of this world-class team, Distilled New York is expanding and looking for SEO consultants with experience.  Cheers!

Jacob Klein Bringing about the Singularity... one link at a time.

Follow @

Jacob Klein

The original post does a reasonable job of describing the process but I’m going to try to go into more detail for those that aren’t as familiar with Python and all that jazz. And unlike the original post I’ll delve into how to use the Windows Task Scheduler for automation. I’m going to assume that you as a user know how to access the windows command line, but that’s about all that I’ll assume.

Getting the Job Done!

1. Download and configure Python.

Download Python 2.7. The default settings will be fine for installation, but you will want to take them one step further by adding Python to your system path if you haven’t. Access the advanced system settings dialogue (it’s in the Control Panel under System and Security > System > Advanced system settings).

After clicking “Edit…”, you’ll want to add a semicolon to the end of the list of directories that shows up, followed by “C:\Python27\”. If you installed it to a different directory, enter that instead.

2. Get the Google Data Python Library.

This will be the only step that requires you to access the command line directly. Download the latest versionof the Google Data Python Library (2.0.16 as of this post). Extract that archive to a convenient directory, and run the following lines in the Windows command line or PowerShell.

NOTE: If the directory you extracted the archive to is different than mine, be sure to account for that.

If you have problems running the script, make sure the install or test scripts, be sure that you added Python to the system path, as described in the previous section.

3. Download and configure the necessary scripts.

The original scripts are hosted by Google Code, from which you should download all three scripts, but must at least download the downloader.py file. You can get by with just that if you would like to use our Distilled-customized script: get-wmt. As mentioned above, this script has been modified to download a Top Pages CSV as well as the Top Queries CSV, and it also allows you to input multiple domains if you so choose. The downloader.py is used in the background by whichever script you choose to go with, so be sure it resides in the same directory as get-wmt.py or example-simple-download.py.

So when you know whether you want to use our script or Google’s provided script (example-simple-download.py). You’ll need to open your chosen script in a text editor and change the lines highlighted below to reflect your own personal information. The username and password are for the Webmaster Tools account you use, the domains are the sites you want to run the queries for. The domains must, of course, be verified in your WMT account.

This screenshot is of the get-wmt.py file. If you use example-simple-download.py, the lines which need editing are very similar.

4. Run the script!

Navigate to the directory that the script is in and double-click the get-wmt.py file to run it. A command line window should open up while the script is executing. Once it’s closed, you should have new CSV files in the same directory as the script you just ran!

Automating It!

Now we’re going to depart a bit further from the original article, so whip out your Task Scheduler!

You’ll want to create a Basic Task:

This will start the Basic Task Wizard. The first few steps are fairly self-explanatory and deal with setting the schedule for the task. I personally run the script weekly so that my data will overlap. When you select an action, you’ll want to choose “Start a program”.

This will bring you to the “Start a Program” page:

The proper way to set this up is hinted at above. Simply enter the script name in the top box. To tell the Task Scheduler where to find the script, specify its location in the “Start in” area. That’s it! Your script will now run weekly or monthly or daily or at whatever interval you specified.

Have fun with your new data!

Measuring the ROI of SEO can be challenging. Here at Distilled, we’ve found that creating a solid benchmark report that is specifically tailored to the client’s needs helps with this process. However, benchmark reports shouldn’t just be used in agencies, but are also useful in-house, especially when demonstrating the value of your work or when you want to make a business case for a specific change to the site, such as to the site’s information architecture.

KPIs and Building a Benchmark Report

One of the first metrics that we request from our clients is their KPIs or key performance indicators, which are the metrics that they track and consider valuable. For example, most of our e-commerce clients track revenue growth.

After receiving the client’s KPIs, we use that knowledge to build a benchmark report, which report upon the metrics the clients can then use to track and measure the success of our work together. At Distilled, we have two types of clients: playbooks or campaigns.

• Playbooks are short engagements (2-3 months) that provide the client with an audit of their site, as well as some strategies on how to improve upon their SEO.
• Campaigns are long-term engagements (6-12 months) that incorporate everything that is included in a playbook and can also include more hands-on activities, such as linkbuilding, writing content, building link bait, etc…

For playbooks, the benchmark report is provided for the client, so that they can track the success of their efforts after our work with them is complete and they are able to systematically implement our changes.

For campaigns, we provide the benchmark report on a regular basis contingent on the client’s needs. Some of our clients request them on a weekly basis, while others on a monthly basis.

Metrics

The benchmark report should be customized. However, there are some metrics that we continuously use in our benchmark reports. Below are some of the common metrics that I use:

Metric: Number of organic keywords sending traffic to the site each month

This metric is not as accurate as it used to be since Google started encrypting sessions for logged in users. However, it can still provide value. Also, the keywords can be further broken down by a designated threshold (example: the number of keywords sending >500 visits to the site)

Process:

1. Go to traffic sources (located on the left-hand navigation in Google Analytics) and select “Organic”

2. Select “keyword.”

3. Export the data via CSV

4. Open up the keyword count template that is located below. Copy and paste the formula in column G (alter the number based on your criteria- in this specific instance, we want to know which keywords had a search volume that exceeded 100. If you want to know which keywords generated a search volume greater than 500, copy the formula and change the 100 to 500.) The “yes” indicates that the keyword does have a search volume that exceeds 100. The formula in column H tells you the number of keywords that met the criteria in column G. In this example, there are 3 keywords that had search volumes greater than 100.

Keyword Count Template

Metric: Number of landing pages (example: pages receiving at least one visit)

Process:

1.  Go to traffic sources (located on the left-hand navigation in Google Analytics) and select “Organic”

2.  Select “Landing Page” on the top navigation

3. The number that appears on the upper-right hand side shows the results. This represents the number of landing pages that brought traffic to your site.

Metric: Number of unique referrals to the site (example: domains sending at least one visit).

This could also be done on a month-over-month overlay.

Process:

1. Go to traffic sources (located on the left-hand navigation in Google Analytics and select “Referrals”)

2. The default option should be “source”. Then look at the number that appears on the upper-right hand side. This represents the number of domains that brought traffic to your site.

Metric: Breaking up keywords by number of words (For example “pants” is one word and “drawstring pants” are two words)

Process:

 Follow steps 1-3 from instructions for “Number of keywords sending traffic each month.“

4. See keyword template count Excel sheet below. Copy the formula that is in column I. This tells you the number of keywords in each phrase. Sort accordingly.

Keyword Count Template

Other common metrics to track include: keyword rankings and segmenting branded vs non-branded traffic visits to the site.

Conclusion

Measuring the right type of metrics that provides both a sufficient overview and some depth in your benchmark report can be extremely helpful. Recently, one of my clients received a massive spike in traffic and based on their current reporting methods, they weren’t able to determine why that massive spike came about – was it one keyword? Was it from long-tail traffic? Having this type of data in the benchmark report can help you determine the cause for rises/falls in traffic. It can also help you develop their own targeted keyword list. For example, if you kept track of which keywords generated the most amount of traffic and page views/time on site/lowest bounce rates, this could be useful information to have for site redesign/site architecture changes in the future.

It’s January

How many times?!

Now that the holiday season is over we’ll see code freezes lifted and the web team asking, “What’s in store for our site in 2012?”. While grand redesigns and marketing promotions are being bandied about, you should probably make sure to speak up about CRO. Why? Read on.

Panda / User signals

“Make pages primarily for users, not for search engines”. Heard that before? It’s a pretty popular mantra from the Google camp, and there’s isn’t much more of a user-centric practice in which you and your web team can engage than CRO.

In the past, the link between CRO and organic traffic generation was pretty weak. CRO will make your site perform better with the traffic you’re already receiving, but only indirectly might it lead to new traffic through, say, users sharing the fantastic experience they had on your site.

Now, with the Panda algorithm using aggregate user site behavior as a ranking signal, the link between CRO and organic traffic is a bit stronger. Changes made in the name of increasing conversion might actually improve your site’s organic ranking as well. Getting buy-in from both the UX and the SEO sides just got easier.

Happy Panda says Weee!

Only Google knows how exactly the Panda algorithm systematically determines whether a piece of content is high quality. Since their efforts are to build an algorithm that emulates the human response to a page’s quality, the best we can do is build a page that users really actually like. Is that such a bad thing?!

I can’t help but wonder if much of the post-Panda efforts by Pandalized sites would’ve been better spent on CRO, rather than on removing ads, hunting down every ounce of duplicate content, removing entire site sections, or revising site architecture. At least in this case, you know you’re getting the most out of the traffic you still have, and you’re actively improving the user experience.

It can be dead simple

Seriously, the setup process for some of the A/B testing solutions out there is incredibly simple. There are quite a few platforms that only require a single snippet installed in theof your test page. From there, the test can be designed, built, executed, monitored and reported on all within the admin of your testing platform.

Designing your tests can be alarmingly simple, as well. Visual Website Optimizer‘s test design process can be completed without altering a single line of code. Take a look:

Of course, large sites that want to test site-wide template changes might need a more robust testing platform (perhaps an in-house solution), but think of the quick and simple platforms as the gateway drugs.

Your competition is (or will be) doing it

With inbound marketing growing by the quarter, the focus will inevitably shift to conversion rate optimization. Why? Because it’s one of the few inbound marketing methods that can lead to a nearly immediate ROI. No one can say no to fast ROI, can they?

Additionally, this year in SEO was a hectic one. With the ups and downs site owners saw in organic traffic from algorithmic changes, many may find themselves looking to protect the household, so to speak.

The point is, practicing CRO is looking more and more like a solid business decision, regardless of whether or not it comes in the form we know it as web marketers. If your competition is smart, it’s on the menu for 2012.

It’s insanely measurable and quick

One reason for the acceptance and practice of online “outbound marketing” methods like PPC advertising, banner ads, and affiliate programs is their measurability and speed with which we can determine effectiveness. With PPC campaigns, we can develop, launch, monitor, and evaluate all within the space of a week. The “I want it done yesterday” types have reason to be enamored with this marketing channel.

Many of the “free” inbound marketing channels, conversely, often have a lag time that is hard to swallow.

CRO is different. It’s like a hybrid inbound/outbound marketing He(or She)-Man, combining the best of both worlds. It is entirely feasible to run a CRO test, find a winner, and benefit from that test permanently within a week (provided your test results are significant).

Conversions > Traffic

Sometimes we all lose sight of what really matters. Of the marketing channels mentioned in this post, only one lives and dies by what matters most to your business, converting visitors: CRO. It deals with the last frontier of your sales funnel.

Where to Start

Start with a simple question: “Where and how can I test improvements to the user experience on my website?” If some answers aren’t immediately obvious, spend time in your web analytics determining where users are falling off. We’ll talk more about this process in a future post.

Rob Millard posted on this very blog a few weeks ago with some great case studies for CRO. If you’re having trouble getting buy-in, some of the studies detail some pretty substantial increases.

Obviously, you might also look into the various website testing platforms. Here are just a few:

• Google Website Optimizer
• Visual Website Optimizer
• Vertster
• Sitespect

Look out for more CRO posts from Distilled over the next few months!

If you don’t know what Screaming Frog is, (then where have you been!?) It’s a tool that let’s you crawl an entire site on demand. But that’s not all it’s good for…I was recently working on a clients site and I found myself turning to the tool more and more for tasks other than the standard crawl feature. I thought it might be useful to put together a list of other ways to use Screaming Frog.

1 – Verifying Google Webmaster Tools Crawl Errors

I recently wrote a post on how to do a site audit using Webmaster Tools. I’m a big fan of it’s features, find that the GWT doesn’t update the crawl errors frequently enough. This means there are often things like 404’s reported when they’ve actually already been fixed. I’ve been using Screaming Frog to solve this problem. Below is my new process for 404’s and other common errors.

1. Crawl the entire site and put all 404’s into a spreadsheet.
2. Download all 404’s from Google Webmaster tools and put them into the same spreadsheet.
3. Remove duplicates.
4. Copy all URL’s into a text document and save it as 404.txt
5. Using the Screaming Frog list mode, upload the 404.txt document and start the crawl

You can then export all the remaining 404s and fix them.

2 – Checking If a Site Migration was Successful

A site I was working on recently changed their URL structure. For a couple of reasons, some of the URLs were not redirecting correctly. The list in Screaming Frog mode came in useful for checking which URLs were not redirecting correctly.

I got the client to send me a list of all the old URLs and followed the same process as above to find out which of the URLs were returning a 404. I then downloaded all the URLs with the problem and passed them to the developer to fix. This made identifying the problem really easy.

3 – Finding Infinite Links

Sometimes websites that use relative URLs can create never ending chains of links. Again, this recently happened on a client’s site. They were using relative URLs everywhere except some pages on the blog.

This meant that sometimes when they linked to a page on the blog, it was being appended to the existing URL. For example:

http://www.example.com linking to www.example.com/page1

Was creating http://www.example.com/www.example.com/page1

This was causing infinite lists of URLs. This means that search engines could be wasting their time crawling pages that technically don’t exist. Because this wasn’t the case on every page, I had to identify where on the site the issue was. When looking at some of the examples, the cause was using links that didn’t include the “http”. To find where this was happening I used the “custom” feature.

This is under the configuration tab. I asked it to include only pages that included the html :

“<a href=”www.”

This then returned all of the pages that were linking to other pages in this way.

4 – Checking a List of Links

During processes of outreach, you often end up creating a large list of pages you are expecting links from. Going through each one to check that the link actually exists on the page can be a tiring job. To speed up the process, the Screaming Frog list mode lets you check a stream of URLs very quickly. There is already a post on the Screaming Frog blog on how to do this: Auditing Backlinks Using Screaming Frog.

5 – Creating a Sitemap

Screaming Frog makes creating an XML sitemap really easy, but it’s important that you set up the crawl correctly before you start. If you don’t limit the spider to crawl only those pages that you want in your sitemap, you can end up with a bunch or URLs that shouldn’t be in there. An example of this is with wordpress, which I discovered when I crawled my site.

A common problem with WordPress is that it creates pages like http://www.craigbradford.co.uk/about-craig/?replytocom=12 when people leave comments.

I don’t want these pages indexed and definitely not in my sitemap, so I can use the exclude tool (which is under the configuration menu), to ensure anything with this style of URL tail is excluded.

Once you have set up the configuration, let Screaming Frog compete a full crawl of the site. Once complete you have the option to export the sitemap. Under the main navigation go to “Export” then select sitemap. You can then upload it to your site and submit it through webmaster tools.

6 – Check Sitemap for Errors

Duane Forrester from Bing recently said that Bing allow a 1% level of ‘dirt’ in a sitemap. “Dirt” could be anything from 404 errors to redirects.

Screaming Frog can be used to keep your sitemap clean and healthy. If you have the XML file like shown above, you can simply change to list mode and upload the XML file. Screaming frog can then crawl all of the URLs and tell you if there are any errors such as 404s or pages that are redirecting.

7 – Using Screaming Frog for Linkbuilding

When doing outreach, I often find it easier if I first contact a link target through something like Twitter.

Taking this one step further, an easy and innocent way to get on someones radar would be to crawl their site for them, find a blog post that returns a 404 and tell them about it.

Assuming you don’t use an SEO profile, this is a good way to be nice and draw attention to the fact that you read their blog. Now when you actually do contact them for outreach, it’s not out of nowhere and they’ll at least recognise your name and face.

8 – IP and User Agent Redirection

Two features that don’t get much attention are proxy option and the ability to change user agent. Taking them in order, the proxy feature can be useful for clients that are using IP based redirects. To see what’s going on you can use buy a set of international proxies or else try some of the free ones at Hide My Ass. You can them compare the results. To use this, select “Configure” then “Proxy”.

Tick the box for “User Proxy Server” and enter your proxy details. When you crawl the site now, it will be using the international address instead of yours. If you are going to do this, I would recommend paying for private proxies as the free ones can be quite temperamental.

Changing user agent can be useful for checking if websites are treating search engine crawlers such as Googlebot differently to users. it can also pick up if robots.txt is explicitly blocking certain content from individual search engines. To use this feature just select “Configuration” and then “User agent”, it doesn’t get much easier than that.

That’s all folks! I’m sure there are Lot’s of other ways to use Screaming Frog or other crawlers, if you have any tips, please leave a comment and I’ll update the post with any other tips. If you have any questions, you can get me on Twitter @CraigBradford.

Craig Bradford is an SEO Analyst at Distilled London. He loves learning anything and is currently learning to be a code master. SEO interests include site speed and deep diving into competitor analysis.

Follow @CraigBradford

Craig Bradford