Wednesday, July 21st, 2010

Facebook Censorship, When Social Networks Block the Sharing of Links (or Worse)

Facebook Censoring LinksSocial networking is booming and millions of people have signed up for Facebook, Twitter, LinkedIn, Foursquare, etc. Using those services, people are continually posting updates, tweeting, liking, checking in, and {enter new buzz term here.} It makes sense. Social networks provide an incredible way to connect with family, friends, colleagues, and others in your industry.

When used correctly, social media can definitely give you a warm and fuzzy feeling inside. But when you take a deeper look, are the social networks really that nice? Who is really in control of your accounts, what are the social networks monitoring, and can you potentially lose the accounts that you’ve worked so hard to build?

-What can happen to your digital equity? -

Facebook Censored a Link I Was Trying to Share
If you follow me on Twitter, then you know that I like to share links to valuable articles and posts. For example, I often tweet about the latest news in Search, Social Media, Analytics, Technology, etc. A few weeks ago, one such link came across Tweetdeck. It was about Facebook apparently shutting down a page that had built up 47,000 fans. The problem had to do with the page in question using a vanity URL with the word Realtor in it (SocialRealtors), which is trademarked by The National Association of Realtors. Jonathan Rivera, the person that had set up the page, obviously worked his tail off building up a following (47K strong), and in the blink of an eye, the page was shut down. That’s disturbing enough, but what happened next is equally as disturbing to me.

Note: The page was finally reinstated and now uses a vanity URL that doesn’t cause any problems trademark-wise. That said, it’s still shocking to know that a social network could crush everything you’ve built up in mere seconds.

On July 15th, I tweeted the article on Twitter and then headed to Facebook to share the link with my friends. I don’t share all of the links on Facebook that I tweet, but this one seemed important. I think too many people frolic around Facebook without understanding the privacy issues involved, the targeting capabilities of Facebook Ads, the ownership of what’s posted, etc. So, I opened up Facebook like I typically do, and started to post my update. I clicked “Attach: Link” and then pasted the link to the post I just read on techsavvyagent.com. When I clicked “Apply”, an error message popped up saying that “Something went wrong. We’re working on getting it fixed as soon as we can.”

Facebook Censorship

OK, I was cool with that. Technology can be glitchy sometimes. So, I waited a few minutes and then tried again. The same message popped up. Huh? So, I took another URL unrelated to the Facebook page that was removed and it went through just fine. Then I immediately tried to share the original article again (about Facebook shutting down the page), and once again, the error message popped up.

Facebook, are you serious? Are you censoring the links I share and trying to make it look like there’s an error? I’d respect Facebook more if it just said, “That link has been flagged as inappropriate based on an internal investigation of the page in question.” At least that would be partly true. To be clear, I totally understand that in certain situations, Facebook should take action. For example, if someone is breaking the law, attacking another user, or someone’s updates are completely offensive. But the idea that Facebook might be censoring links because it doesn’t want certain articles to be shared is ridiculous. By the way, I also asked my wife to try and post the link, and she couldn’t either. I wanted to make sure this was more than an isolated issue.

For almost an entire week, I could not post the link. The error message was coming up as recently as yesterday morning. Today, however, I could post the link. It’s the first time since July 15th that I’ve been able to share that post on Facebook. Why did this happen, how come I can now post the link, why did it take almost exactly one week? I have no idea, but I’m still not happy that Facebook was stopping the link from being shared…

How Widespread Is Facebook Censoring?
So of course the next question that came to mind was, “How widespread is the censoring of links or other activities on Facebook?” And to take it even further, is Facebook using “error messages” as a tactic for making users believe that they aren’t being censored? Unfortunately, I don’t have answers to these questions (yet). A quick search on Google for Facebook censorship yields many articles about Facebook censoring links or groups or updates, so this is nothing new for the social network with close to 500 million members.

I think it’s important to address this topic now. As technology advances, we’ll be dealing with more and more connectivity (on both a technology and a personal level.) Layered on top of that connectivity are serious privacy issues, censorship issues, and a growing concern about “what crosses the line?” To me, Facebook is crossing the line if they are censoring links like the one I tried to share.

Digital Equity and Account Ownership
I wrote a post in 2009 that was titled, Lawyers, Guns, and Twitter – Who Owns Your Twitter Account, which addressed the sticky topic of Twitter account ownership. The post seemed to strike a nerve with Twitter users and executives, since there wasn’t a clear answer. One of the comments was by Steve Plunkett, someone I greatly respect in the SEO community. You can read his entire comment on the post itself, but his point was that none of us “own” our social media accounts. The social networks do. I think his point never meant more than the second I realized that Facebook was censoring my links. Steve is right and the situation I experienced supports his view. Not only was Facebook inhibiting me from sharing a link, but that link was about another person whose page of 47K fans was shut down without any notice. So, you had two examples of how a social network had the ultimate control.

This is part of the reason that I explain to new clients that although social media can be incredibly powerful, they should first invest in their own site, blogs, etc. They shouldn’t simply rely on their social media accounts without having a solid strategy for building up power in their own assets. You own your domain name, you own your website, and you own your content and blog posts. In addition, if you do the right things, your assets can build up search power, traffic, subscribers, customers, etc. (AKA digital equity). For the most part, nobody can take that away from you. Of course, Google can strip your SEO power if they believe you broke the rules, but you get the picture. You have much greater control than you do on social networks (where the socnets are clearly in control…more than you think).

Think about this for a second. Imagine you built up 4500 fans on Facebook and 8000 followers on Twitter only to wake up one morning and find out that both social networks shut down your accounts. You would be left with nothing. Don’t let this happen to you. Make sure you also focus on your own assets (your sites, blogs, content, etc.) Stay in control of as many assets as you can, and build your digital equity. Social media will definitely be part of the equation, but don’t let it be the only part. You can ask Jonathan Rivera about how he would do things differently after his page was shut down (even though it has been reinstated).

Final Thoughts About Social Censorship
As you can probably tell, I believe this topic is extremely important. What happened on Facebook that day disturbed me on several levels. First, Facebook shut down a page with 47,000 fans, with no warning. Second, Facebook was censoring my links about that happening. Third, Facebook showed an error message versus telling the truth about what it was doing. I don’t know about you, but I don’t like social networks telling me what I can share, while also having the ability to strip away hard-earned digital equity in a flash. That’s part of the reason I’ve been blogging for years.

Think about that the next time you feel all warm and fuzzy inside when sharing the photos of your kids…

GG

Tuesday, July 13th, 2010

How to Use The New Day Parts Report in Google Analytics to Improve Your AdWords ROI

Day Parting in AdWords Via Ad SchedulingWhen managing campaigns, search marketers are always looking for ways to improve performance. Paid Search is a multi-faceted online marketing channel, and some of the factors impacting your campaigns are not always easy to determine. For example, is your keyword list sufficient, can you create better ads, should you geotarget your campaigns, etc? But what about the time of day or day of week that you are running your ads? Could your ad scheduling (or lack thereof) be impacting your ROI more than the previous factors mentioned? These are all good questions and the answers lie in the campaign analysis you perform. The good news is that you can find the answers to these questions, but you just have to look in the right places.

The New AdWords Reporting in Google Analytics
As I explained in my last post about Matched Search Queries, Google Analytics released a major update to its AdWords reporting in June. The update was a big improvement and provides a number of new reports and ways to dimension your reports. One of the new improvements is the ability to run Day Parts reporting right within Google Analytics. This enables you to analyze your AdWords performance by time of day and day of week. Based on your analysis, you might choose to utilize the Ad Scheduling feature in AdWords to only show your ads during select times of the day (or days of the week). In addition, you can adjust your bids by time of day. Day Parting is a powerful feature in AdWords and can definitely help improve your performance in paid search.

Based on the importance of analyzing AdWords data by time of day, I wrote a blog post explaining where you can find the Day Parts report, how to run it, and then how to export and analyze the data in Excel. But, you are going to have to visit Search Engine Journal to read my post.

So head over to Search Engine Journal now to learn more about Day Parts reporting:
Time is On Your Side, Using Day Parts Reporting to Increase Your AdWords ROI

GG

Wednesday, July 7th, 2010

How To Use Matched Search Query Reports in Google Analytics to Improve Your AdWords Campaigns

Matched Search Query Reporting in Google AnalyticsLast month, Google Analytics released a new version of AdWords reporting to all accounts. This was much-needed and Google did a great job with adding the new reporting functionality. There are some great additions to the reporting that can really save you a lot of time, while also helping you enhance your current AdWords campaigns. One of the new enhancements is the ability to dimension your AdWords reports by a number of options, including Matched Search Query. Viewing matched search queries essentially lets you see the actual keyword that someone entered in Google that triggered your ad. To clarify, it’s the query that someone entered that mapped to one of the keywords you are bidding on.

For some people new to paid search, this can be an eye-opening report to view. You basically get to see the actual keywords that are triggering your ads and generating click-through. A quick example would be if you are bidding on the keyword Callaway golf clubs and you see a Matched Search Query of where can I buy used Callaway golf clubs. Based on what you find in the Matched Search Query reporting, you might choose to make several modifications to your campaigns and ad groups. I’ll explain more about those modifications soon, but viewing this data can definitely help you enhance your campaigns. Let’s start by finding the reports in Google Analytics (via the new reporting that was released last month).

Side Note: Prior to the new AdWords reporting, you could always run a Search Query Performance Report in AdWords to view your keyword data. That said, there are many people that didn’t even know the report was available (since it was buried in the reporting tab within an AdWords account). You can also find this keyword information directly through the AdWords interface by clicking the “See Search Terms” button within a specific ad group. But, you won’t be able to analyze the various site performance metrics that are available in Google Analytics (including conversion metrics.)

Finding the Matched Search Query Report
When accessing your reporting in Google Analytics, you can click the Traffic Sources tab on the left side of the interface and then the AdWords tab (which has a “beta” label right now).

AdWords New Reporting in Google Analytics

Once you click the AdWords tab, you can click the Campaigns link to view a list of your AdWords campaigns. After you click a campaign, you will see the ad groups that are part of that campaign. Clicking through an ad group will show you the keywords that are part of that ad group. These are the keywords you are bidding on, and not the raw queries that are triggering your ads. There’s a big difference. You will see two dimension dropdowns in this view. The first dropdown will have “Keyword” selected already. The second dropdown is where you can further dimension your reporting. If you click that second dropdown, you will see “Matched Search Query” in the list. If you select that option, then Google Analytics will show you all of the raw queries that matched the keywords you are bidding on.

Using the dimension dropdowns in Google Analytics to find matched search queries

At this point, you should probably take some time and view the list of raw queries for the ad group at hand. My guess is that you will be surprised by some of the keywords that led to click-throughs… You might also be interested in reviewing the variations of keywords that led to click-through (the long tail keywords that are triggering your ads). Both topics will be covered below and can definitely help you enhance your campaigns.

Finding Negatives Via Matched Search Query
Many experienced search marketers already know the power of negative keywords. A thorough and detailed negative keyword list can greatly improve the performance of your campaigns. If you are unfamiliar with negatives, a negative keyword will tell Google not to show your ad if that keyword is part of the query. This is extremely useful, especially in a Quality Score-driven world.

For example, let’s review the query I used above (where can I buy used Callaway golf clubs). If you remember, the keyword you were bidding on was actually Callaway golf clubs, and not the longer tail keyword that triggered your ad. Now, let’s say you don’t sell used Callaway golf clubs. You only sell brand new sets. In this situation, you can use the negative keyword used to keep Google from showing your ad when someone uses that word in their query. For example, the following queries would not trigger your ad, “buy used Callaway golf clubs”, “where can I find used Callaway golf clubs”, etc. Since the word used is part of the query, Google should not show your ads.

Quality Score and Wasted Clicks
Finding negative keywords helps you on several levels. First and foremost, it saves you money. If you don’t sell used Callaway golf clubs, then people searching for used Callaway golf clubs will never buy from you. Every click that comes through from people searching for used golf clubs will be wasting your budget. So, using a thorough list of negative keywords saves you money.

In addition to saving wasted clicks, using negative keywords can help your Quality Score. If you’re not familiar with the concept, Google actually determines a Quality Score for each of your keywords. Your Quality Score is directly tied to the performance of your keywords (with a heavy focus on click through rate). There are several factors involved with determining Quality Score, including how relevant your keywords are to your ad text and the user’s query, the quality of your landing page, etc. Each keyword in your ad group is given a Quality Score from 1-10. In general, a higher quality score will yield ads that trigger in a higher position and at a lower cost per click. The lower your Quality Score and your ads will show up in lower positions and will cost you more per click.

Quality Score in Google AdWords

As I explained above, Quality Score is heavily impacted by click through rate. So, if your ads are showing up for queries that aren’t targeted (like if you don’t sell used Callaway golf clubs), then you run the risk of negatively impacting your Quality Score. And, if you negatively impact your Quality Score, you might have to pay more per click. So, using negative keywords to make sure your ads don’t show up for untargeted searches is very important. One way to find negative keywords is by analyzing matched search queries. As you scan the list of matched search queries, you will probably come across several words that can be used as negatives. It’s somewhat time-consuming to do this, but can pay off huge dividends in paid search.

For example, if your Callaway ads say “New Calloway Golf Clubs”, then people searching for used clubs will probably never click through to your site. This will lower your click through rate, which will lower your Quality Score. If you lower your Quality Score, then it could impact your campaign performance by increasing your cost per click (CPC) and lowering your return on investment (ROI). You want to avoid this situation, if possible.

I recommend running the report the way I explained above and then exporting that report as a csv for Excel file. Then you can work in Excel to analyze the data, pick and choose your negative keywords, and create a final list that can be uploaded to your AdWords account. I recommend starting with your core campaigns and ad groups, with the ultimate goal of completing this process for all of your active ad groups.

Quick Tip: Make sure you have the right timeframe selected in Google Analytics in order to view a good number of matched search queries. For example, if you only select one or two weeks, you might not have enough data in your matched search query reporting. If you haven’t run this report yet, then definitely try and view a few months of data, if possible.

Enhance Your Keyword List With Longer Tail Keywords
You can also use Matched Search Query reporting to find new keywords for your ad groups. For example, you might find dozens of quality keywords (or more) matching a single, broader keyword that you are bidding on. If you continually analyze matched search queries, then you can find longer tail keywords that you can add to your campaigns. These new keywords may very well be less expensive than the broader keywords you were originally targeting. It’s not uncommon to add numerous keywords to your campaigns on a regular basis by checking matched search queries.

As a simple example, let’s say you were bidding on a broad keyword like LCD HDTV. When you run matched search queries, you might find a long list of keywords that matched your broad keyword. You might find manufacturer names, model numbers, colors, features, etc. Depending on your specific business, you might want to include these keywords in your ads groups. You might even find ideas for creating new ad groups. In paid search, you typically want to create tightly-focused ad groups with highly-relevant ads. Your ad text could then address the specific queries that people are entering versus being more generic. Remember, Quality Score is heavily impacted by click through rate. High quality ads that are extremely relevant to searches help gain higher click-through. This can increase your Quality Score, which can lower your cost per click.

Similar to what I listed above in the section about negatives, I recommend exporting your matched search queries as a csv for Excel file. Then you can work in Excel to analyze and break down the reports. Once you have a new list of keywords, you can upload them to your AdWords account. Adding these new keywords can help you target less expensive, longer tail queries, which can help boost the ROI of your campaigns.

New AdWords Reporting = New Ways to Enhance Your Campaigns
As you can see, Matched Search Query reporting in Google Analytics can be a powerful way to enhance your campaigns. It’s a relatively simple way to help you find the right negatives for your ad groups, while also helping you find more targeted keywords to run. This can help increase your Quality Score, lower your CPC, and increase your ROI in paid search. And that’s a pretty darn good combination. So go ahead and run the reports. You won’t be disappointed.

GG

Tuesday, June 22nd, 2010

The Anatomy of a Twitter Profile Review

Twitter Account ProfilesWhen other users receive a notification that you followed them on Twitter, there’s a good chance that they will click through that email to check out your profile. When they hit your profile on Twitter, you have about 10-15 seconds to impress them enough that they will follow you back. Given that you only have a few seconds, are you providing the right information in your profile, is there enough information, is it accurate, and can they get a feel for who you are and what you do? These are all good questions that you should be thinking about.

When I’m helping clients with Twitter marketing, I often get questions about what to tweet, Twitter etiquette, who to follow, how to engage users, etc. However, it’s usually not long before the topic of Twitter profiles come up. For example, what should be included in a solid profile, what should the bio include, what should your avatar look like, should you link to a website, blog, etc? Although profiles on Twitter are relatively simple, they are definitely important. Your profile is often your first impression on Twitter, so it’s important that you put your best foot forward. And that’s especially true when you are first starting out on the microblogging service. In my opinion, you can really shoot yourself in the foot if you don’t understand what other people are looking for on your profile.

The Anatomy of a Twitter Profile Review
Based on what I explained above, I thought it would be a good idea to provide the anatomy of a Twitter profile review. I’ll explain the core elements of your profile that other users are reviewing, provide best practices for fleshing out these elements, and explain how your profile can impact your follower count. If you are new to Twitter, my hope is that you can leave this post and make sure your profile is helping you versus hurting you. Without further ado, let’s dig into Twitter profiles. Note, I’ll simply use screenshots from my own profile in order to highlight each profile element.

Your Name, Twitter Handle and Avatar (Image or Photo)
Twitter Handles and AvatarsWhen you receive email notifications about new followers, the first thing you’ll see is the person’s name and Twitter handle. It might look like “Glenn Gabe (glenngabe) is now following your tweets on Twitter”. A Twitter handle is essentially your username (in this case glenngabe, all as one word). I highly recommend using a handle that ties directly with your name, brand, or company name. I would not get too creative here… If your first impression with someone is corny, weird, ambiguous, or offensive, then you might kill your chances of a follow back almost immediately. For example, “Money Man Jimmy is now following your tweets on Twitter” (with a handle of @moneymanjim) won’t necessarily be a great move for you. Since there are many spammers on Twitter, I probably wouldn’t even click through to a profile like that. Don’t make the mistake of alienating Twitter users before they even reach your profile.

In addition, you can view their avatar in the notification email (the image that someone uses to represent their account). This too can be a telling story. Is the avatar a headshot of the person, a logo, a funny image, cartoon, celebrity photo, is it blank, or is it something even worse like a risky image? Be careful with the avatar you choose. Again, this is your first impression. Twitter is so much about personal engagement that I almost always recommend using an actual photo, if possible. Remember, you’ll be engaging other people on Twitter, so you typically want to know that an actual person is on the other end.

If you’re a company that’s setting up a company profile, then you can use your official logo (although I’m not a big fan of following logos). If you’re a recognizable brand, then it’s a different story. But most companies aren’t recognizable brands… Also, if the person managing your account is your social media manager, then I have no problem letting that person use their own photo, as long as their bio explains what they do (more on that soon).

My recommendation: Use a descriptive handle (something close to your name if you can get it), use an actual photo of yourself as your avatar, and use your real name as the name on the account. If you do that, other users will know that Glenn Gabe (@glenngabe) is now following them (and they can check out an actual photo). You’ll start off the right way, and will give yourself a greater chance of gaining new followers.

Your Bio and Location
Twitter Profile BiosWhen you meet someone for the first time at a meeting or conference, what do you say to them? Do you crack a joke during that first exchange, do you say nothing, or do you succinctly introduce yourself by providing your name and what you do? For most of us, it’s the latter. I believe that it’s extremely important to list your full name in your profile, as well as a thorough bio. “Thorough” is a relative term given you only have 160 characters to play with, but you can still get enough in your bio to clearly state who you are and what you do.

When reviewing someone’s profile, I look to see if the person has their full name listed and if they have created a bio. I typically won’t follow someone back if they just have a partial name listed, along with a thin bio. Let’s face it, if you are looking for a lot of privacy, then Twitter isn’t the place for you. If you check my bio, you can see that I provide my full name, my company name, the areas of online marketing that I focus on, and a link to my blog. More about profile links later. You can quickly get a feel for who I am and what I do. If you leave some of this information out, you risk not being taken seriously. That means you might limit the amount of people that will follow you back. And that’s not good, especially when you’re new on Twitter.

Your Followers to Following Ratio
Twitter Follower to Following RatioWhen you’re just starting out on Twitter, it’s hard not to have follower-envy or a complex about the number of followers you have. Typically, nobody knows who you are, nobody cares, and nobody is chomping at the bit to follow you. You’ll be caught in the black hole of Twitter for a while. During the process of growing your Twitter account, almost everyone ends up following more people than they have followers (and that’s for a while).

So it’s natural that one of the first things you do when someone follows you on Twitter is to check how many followers they have. That’s fine, but what about the number of people they are following? Does the ratio matter? It absolutely does… There’s a natural follower to following ratio that you’ll often end up seeing. For example, 800 followers while following 750 is a normal ratio to see. Then, there are times you see an abnormal ratio, like someone who is following many more people than they have followers. This can be a red flag and could tell you that someone is quickly following a lot of people in order to build up their own follower count. On the flip side, if someone barely follows other people, are they really a good person to follow? It might signal that they don’t engage many users. For example, following only 250 people, but having over 4000 followers. Both scenarios could signal a problem…

A normal follower to following ratio will be closer to 1:1 (for most users). For example, someone might have 1500 followers and they are following 1350. That ratio is much stronger and looks more legitimate than someone following 7500 people with only 1500 followers. As people become more popular on Twitter, they might have more followers than people they follow, but the ratio will often still look normal. For example, someone might have 8000 followers, but they are still following 5000 people. As you get used to Twitter, you can tell pretty easily what’s legitimate and what’s not. If you see someone following the world without many followers, then you might want to explore their profile before deciding what to do. Also, after I check someone’s follower count and follower to following ratio, I often check the number of tweets they have (to see if that number is in line with the number of followers they have). More on that below.

How Many Tweets Do You Have?
Number of Tweets on TwitterOne profile element that I often look at is the pure number of tweets that a user has. The reason is simple. It should make sense, given the number of followers they have (and is hard to game). For example, if someone has over 2000 followers and only 25 tweets, something isn’t right. That’s unless they are famous. Unfortunately, some people game Twitter followers in order to inflate their follower count. Checking their pure number of tweets is a great way to identify people trying to game the system. Most people that have built up a decent amount of followers have tweeted often. For example, I have almost 5000 followers and I have over 13,000 tweets. That makes sense and you can tell I’m on Twitter often. If I had 100 tweets and 5000 followers, then you should probably run from my profile, and not walk. :)

Your Profile Link – The Link to Your Website, Blog, or Profile
Twitter Profile LinkOn Twitter, you have the ability to provide a link in your profile. I believe it’s important to include a link to build more credibility. For example, many people link to their website or blog. You can also point to your LinkedIn profile, Facebook profile, etc. On Twitter, it’s not unusual for someone to check out your profile and then click through the link you provide to see what you’re all about. I see many visits from Twitter to my website. Again, think about meeting someone for the first time. If you didn’t tell someone where you were from and what you do (with some backing), you would have a hard time building up credibility. I highly recommend providing a link and making the destination something that helps you build credibility.

Shortened URL’s in Your Profile:
Some people use shortened URL’s in their profile, such as using a bit.ly link. I’m not a big fan of doing this, since you don’t know where the link is taking you. There are a lot of scammers and spammers on Twitter and links can potentially take you to malware, scams, etc. Therefore, I recommend providing a full link to your blog or website (something people can clearly see). If you go to my profile, you’ll see a raw link to my blog (using my domain name). Again, you want to make it easy for people to learn about you. Don’t make them think too much. You could end up losing potential followers.

Are You Listed?
Twitter ListsLists on Twitter enable you to organize certain users into categories. Typically, you only list someone if you think highly of them. For example, if you were a PR executive, you might list several thought leaders in a list titled “PR-Leaders”. So, the number of times that users have been listed is a metric that some users look at to see what other people on Twitter think of them. To me, that number doesn’t have to be incredibly high, but should make sense given the other elements I’ve listed in this post. If you see that someone has been listed even 5 or 10 times, that might be fine. If you see a very low “listed count”, then you should understand the context before making any judgments. For example, are they brand new on Twitter? If so, then it makes sense that they haven’t been listed much. If they have 2000 followers and haven’t been listed once, then that might be a red flag. You will naturally be listed over time if you are doing the right things on Twitter, but that’s for another post.

What Are Your Last 20 Tweets?
This is where the rubber meets the road. Similar to reviewing the blog posts of a blogger, checking the tweets of a Twitter user can speak volumes for the quality of that user. When you visit someone’s profile, you get to see their last twenty tweets. You might find a great list of valuable tweets to outstanding content, several retweets, conversations, etc. That’s all good to see. But, you might also find random thoughts, profanity, a lack of valuable content, no retweets, no conversations, affiliate links, etc. This is why it’s important to consistently and continually provide high quality content via your Twitter stream. Twitter is a “show me” medium. If you can show other users how valuable your tweets are, you’ll give yourself a much better shot of gaining new followers. So, when someone checks out your latest tweets, do you want them to see low value updates or great content? Think about this the next time you’re about to tweet something. :)

Last 20 Tweets on Twitter

When Is the Last Time You Tweeted?
When you check the last twenty tweets from a user, you can also check the last time they tweeted. That’s important, since you might see that their last tweet was over a month ago (or worse). In a perfect world, you would see that someone last tweeted less than 24 hours ago. Even better, you would see some tweets from just a few minutes or hours ago. You don’t want to end up following a lot of people that don’t post updates! That’s the antithesis of a good Twitter user and a waste of time for you… This is also why I recommend that you tweet daily. Not doing so can result in losing potential followers. I never like seeing that a user hasn’t tweeted in days (or longer).

Last Time You Tweeted on Twitter

Summary – Do You Need To Tweak Your Profile Now?
If you’re new to Twitter, you might be surprised to learn that a Twitter profile review is not so simple! As you spend more time on Twitter, you’ll understand more about the importance of each element listed above. When you break it down, Twitter users don’t want to get gamed, they don’t want to follow spammers, and they don’t want to bog down their accounts by following zombie users . That’s why you should spend some time and make sure you are providing the right information in order to make a great first impression on Twitter. Remember, you will have potential new followers checking out your profile on a regular basis. You just need to make sure you can convert them to followers. Unfortunately, many people are not doing a great job at this. My hope is that you can.

GG

Tuesday, June 8th, 2010

Understanding Average Time on Page and Average Time on Site in Google Analytics [EXPERIMENT]

Avg Time on Page and Avg Time on SiteIf I showed you the statistics for two pages and the first had an Average Time on Page of 1:30 and the second had an Average Time on Page of 4:45, you would probably think that visitors were more engaged with Page B, right? But, what if I told you that the actual Time on Page for Page A was 8:30. Is that possible? If it was possible, you might start to lose confidence in that metric…

But that’s exactly what could happen based on the current way that Time on Page and Time on Site are calculated in most analytics packages. The reason is because the metrics are based on page jumps. For example, if you visit Page A for 2:00 and then visit Page B, Google Analytics can determine how long you spent on Page A by calculating the difference between the timestamps for each page. For this example, Page A would have a Time on Page of 2:00. However, calculating Time on Page via page jumps is inherently flawed since if you exit from Page B, Google Analytics has no way to determine how long you stayed on that page. Your time on page is 0:00, whether you spent 10 seconds or 10 minutes on the page. Needless to say, this isn’t a good thing if you are trying to accurately measure site performance.

I’ve helped many companies with web analytics and I find there’s a lot of confusion with time-based metrics like Time on Page and Time on Site. Unfortunately, inaccurate reporting can lead to a skewed analysis of site performance. Then to make matters worse, some marketers are making decisions based on inaccurate reporting and analysis. I can’t tell you how many times over the past few years I have explained how Time on Page and Time on Site are calculated, only to see a completely deflated look from the marketer I’m explaining it to. Context is critical in web analytics.

So, to help get everyone on the same page, I decided to run an experiment. My goal was to isolate visits to specific pages and directories, so I could drill into the reporting to identify how Time on Page and Time on Site were being reported. The results will demonstrate why you need to take both Time on Page and Time on Site with a grain of salt.

If you are excited to see the results, then you’ll have to visit my post on Search Engine Journal to learn more. The post went live this morning. :)

Tick Tock: The Limitations of Time on Page and Time on Site in Google Analytics.

GG

Wednesday, June 2nd, 2010

How To Track Elapsed Time (or Time to Complete) in Google Analytics Using TimeTracker [TUTORIAL]

Tracking elapsed time in Google Analytics via Time TrackerOver the past few years, conversion optimization has become an incredibly important service that I provide for my clients. Sometimes in order to improve conversion, there are times that I need to analyze a multi-step process that occurs on one page. A good example of this would be an elaborate form that visitors need to fill out (without linking to additional pages). Or, I might be tracking a multi-step process that spans several pages, but want to know more about each specific step. If you run into situations like these, then it might not be sufficient to simply track how many people converted. You might want to analyze how long it’s taking for those visitors to complete a certain process (in order to identify obstacles along the way). This is when using tracking functionality like TimeTracker in Google Analytics can come in very handy.

What is TimeTracker in Google Analytics?
TimeTracker extends event tracking in Google Analytics and enables you to track elapsed time, or Time to Complete. I’ve found that many people aren’t familiar with TimeTracker, don’t use it much, or don’t know how to properly set it up. So, I’ve decided to write this tutorial to walk you step by step through the process of setting it up. Then I’ll quickly show you how you can check the event tracking reporting for TimeTracker in Google Analytics.

For our purposes, TimeTracker will enable us to trigger the start of a process (like clicking a button, entering text, or selecting a checkbox). We will start a timer at that point. Then it enables you to trigger the end of that process (again by some action taken by the user). At the end of the process, we will stop the timer, record the elapsed time, and then communicate with Google Analytics to track the result as an event (via Event Tracking).

Let’s get started. Open up your html editor of choice and get ready to use TimeTracker.

1. Define What You Are Going to Track
For this tutorial, I’m going to keep the process and form simple so you can clearly understand how to use TimeTracker. I’ll show you how to start the timer with the first click of a form element and then how to stop the timer and send the data to Google Analytics when users click the submit button. Again, this will be a basic setup so you can easily follow along. You will probably want to write some custom logic for your own projects, based on the specific process you are going to track.

2. Adding The Form Elements
We’ll start by adding a simple html form to your webpage. In your form, add a group of radio buttons to enable users to select their age. Since this is the first form element, we’ll start the timer via TimeTracker when someone clicks a radio button. After adding the radio buttons, you can add several additional form elements, based on what you need to track. I won’t cover how to add the additional form elements, since they don’t impact TimeTracker. Basically, they can be standard html form elements (text boxes, dropdowns, checkboxes, etc.) At the end of our form, we’ll add a submit button for users to complete the process. The submit button will first stop the timer and then send the data to Google Analytics via Event Tracking. We’ll complete this via a custom JavaScript function that we’ll write later in the tutorial. Don’t worry, it’s a simple JavaScript function. :)

Radio Buttons Will Start Our Timer:
Time Tracker Form Elements

First, add your form, the radio buttons, and your submit button: We will add the TimeTracker code later in the tutorial.

1
<form name="frmTrackTime">
2
    	<input name="rb1" value="18-24" type="radio">18-24<br />
3
        <input name="rb1" value="25-40" type="radio">25-40<br />
4
 
5
        <!—add more form elements here-->
6
 
7
         <input type="submit" id="btnTrack" value="Submit!" />
8
</form>

3. Download the JavaScript (or link to it from your code)
The TimeTracker JavaScript code can be found on the Google Code page for extending event tracking. You can either link to that code directly from your page or you can include the code in your own JavaScript file. I included the code in my own file by copying and pasting the JavaScript code into a new text file and saving it locally as “time-tracker.js”. Note, you will also need your typical Google Analytics snippet included in the page.

Click the TimeTracker() Title to Access the JavaScript Code:
Time Tracker Google Code

4. Add the JavaScript code to your webpage
In order to add the code to your webpage, simply include the following line of code in the head of your html document.

1
<script type="text/javascript" src="http://yourdomain.com/time-tracker.js"></script>

Note, make sure you enter the actual location of the JavaScript file on your server and that you reference the correct file name. Replace “yourdomain.com” with your own domain.

5. Create Your TimeTracker Object
You will need to create a TimeTracker object in your JavaScript code, which we will end up calling from our form elements in order to start and stop the timer. The code will also send the data to Google Analytics (via Event Tracking). Add the following code below the code you entered earlier (where you referenced the TimeTracker JavaScript file).

1
2
<script type="text/javascript">  
3
     var timeTracker = new TimeTracker();     
4
</script>  
5

6. Customize the JavaScript (Histogram for Time Intervals)
There is an optional array you can specify when you create the TimeTrakcer object. It’s called setHistogramBuckets() and it enables you to set the time intervals that get passed to Google Analytics (when the event is sent and tracked via Google Analytics). For example, you can specify the time intervals (the buckets of time) that Google Analytics will drop users into (based on how long they spent completing the process you are tracking). If you leave this empty, TimeTrakcer has default time intervals (which might not fit well, given your specific form or process).

The default buckets will start at 100 milliseconds and go up to 5000 milliseconds. In case you are wondering, there are 1000 milliseconds in one second. I highly recommend setting the time intervals, based on the specific process you are tracking. For example, if you think the process could take up to three minutes, then you would set up intervals leading up to three minutes. Three minutes is 180 seconds, or 180,000 milliseconds. For our purposes, we’ll set six intervals starting at one second and that go up to 25 seconds. If it takes someone more than 25 seconds, you’ll see it recorded as 25000+ in your reporting (so they won’t be left out). Here is the code you should add right below the code you added above (where you created your TimeTracker object). Note, add this code within the script tags you already coded earlier.

Specify your own time intervals via the setHistogramBuckets() function:

1
timeTracker._setHistogramBuckets([1000, 2000, 5000, 10000, 15000, 25000]);

The final block of code should look like this:

1
<script type="text/javascript" src="http://yourdomain.com/time-tracker.js"></script>
2
<script type="text/javascript">  
3
     var timeTracker = new TimeTracker();     
4
     timeTracker._setHistogramBuckets([1000, 2000, 5000, 10000, 15000, 25000]);
5
</script>  

7. Add JavaScript to Start Your Timer
Now that you have your form elements in place and you’ve included the necessary JavaScript code, you’re ready to start your timer. In the first form element (the radio button), add a call to TimeTracker to start the timer.

1
<input name="rb1" value="18-24" type="radio" onclick="timeTracker._recordStartTime();”>18-24<br /> 

Note, you should also add this code to the other radio button that is part of the group. This is because we don’t know which button the user will click (based on their age). We want to start the timer if they click either radio button.

8. Add JavaScript to Stop The Timer and Send The Data
Similar to what we completed above, you are now going to add code to the submit button to stop the timer. Then you’re going to send that data to Google Analytics via event tracking. In order to streamline this process, I created a simple JavaScript function that both stops the timer and then sends the data to Google Analytics as an event. I named this function completeTracker() and included it in the head of my html document. Let’s first add the code to the submit button.

1
<input type="submit" id="btnTrack" value="Submit Answers" onclick="completeTracker();" />
2

Now here is the code for the JavaScript function called completeTracker(). Add this code to the head of your html document:

1
function completeTracker() {
2
	timeTracker._recordEndTime();
3
	timeTracker._track(pageTracker, undefined, 'Lead From Page X');
4
}

The code explained:
The first line simply stops recording the time. The second line sends the data to Google Analytics via event tracking. The first parameter is your pageTracker object, which is created via your standard Google Analytics code snippet. The second parameter is optional and will enable you to customize the category of the event in your reporting. Categories are essentially the top-level name of an event in your reporting. The third parameter is also optional and will enable you to customize the label that is reported in event tracking. I added a label called “Lead From Page X”, so you could identify where the event was triggered. You would obviously want to replace X with the actual page name that contained your form. You can read my post about Event Tracking in Google Analytics to learn more about the various parameters involved with tracking events.

9. Upload and Test
When you upload your files, make sure you include both the html file and the external JavaScript file. Also, make sure you are referencing the external JavaScript file correctly or your code will not work. For example, did you upload the JavaScript file to the same directory, a code-only directory, etc? Make any necessary changes in your html code if the JavaScript file isn’t placed in the same directory as the webpage containing your form. If you are all set, then click away. I recommend going through the form numerous times from separate browsers and separate systems. Remember, the first radio button triggers the timer and the submit button stops the timer and then passes the data to Google Analytics via event tracking.

10. Wait and Check Reporting
You will probably need to wait a few hours before you can view the reporting in Google Analytics. After which, you can access the Content tab and then click “Event Tracking”. If you click the Categories tab, you should see a TimeTracker category. If you click that category, you should see the various times listed. These are the time intervals that we set using setHistogramBuckets(). If you click each time interval, you will see the specific form or page that triggered the event. For our purposes, all the events were triggered via one form and page. However, that might not be the case if you have multiple forms running on your site.

TimeTracker Events in Google Analytics:
Time Tracker Reporting in Google Analytics

Congratulations! You have just successfully tracked elapsed time (or Time to Complete) in Google Analytics. My hope is that you’re thinking of many more ways to use this functionality in your own projects. I recommend brainstorming several ideas for using TimeTracker to see the potential impact on conversion. I would start small and then increase the complexity of each project as you get more comfortable. As you can see, it’s relatively easy to set up and can provide insight into how long it takes visitors to complete certain processes on your site.

By the way, it just took you 10:42 to complete this tutorial. Just kidding. :)

GG

Monday, May 24th, 2010

The Strategy to Execution Gap™ (SEG) and Its Effect on SEO

The Strategy to Execution Gap (SEG)Over the past 15 years, I’ve had the opportunity to work on some incredible projects with some extremely talented people.  Whenever I was about to launch a new initiative, I found it was helpful to look at the various challenges and obstacles to success (in order to minimize them as much as possible).  When dealing with online marketing projects, there are several variables that can inhibit your progress, including technology, process, and people.  All three categories of obstacles can throw a wrench into effectively completing tasks, which can then lead to missed deadlines and a slower path to success.  That said, there are also times that a path has been cleared and you can execute very quickly.  And in online marketing, efficient execution is critical.  I’m a firm believer that you can build the best strategy in the world, but unless you can execute at a rapid pace (while maintaining high quality), you’re dead in the water.  The outstanding strategy you created won’t be worth the paper it’s written on.

The Strategy to Execution Gap™ (SEG) and SEO
About 4 years ago, I created a metric to demonstrate how delays and obstacles can impact online marketing initiatives.  The metric is called the Strategy to Execution Gap™, or SEG.  The SEG is a metric that can help you identify how effective your team is for a given initiative or set of initiatives.  If you’ve read some of my previous posts, you already know I am analytics nut.  I love using data to back what I recommend (and I’m a big believer that opinion gets you nowhere while data is hard to ignore).  This is part of the reason that the SEG can be so valuable.

The SEG provides a percentage, which can show you how effectively your team is executing projects.  The lower the number, the more efficient your team is.  The higher the number, and your team is not executing at an effective level. It can be used for any online marketing initiative, but I’ll focus on SEO for this post.  The Strategy to Execution Gap™ fits SEO very well, since there are typically a number of projects that need to be completed during an engagement (and each usually has a weighted priority).  I’ve written extensively about SEO technical audits and remediation plans in the past, and they help build an SEO roadmap, or the series of projects that need to be completed during the year.  For example, in SEO it’s extremely important that technical barriers are removed before you move into projects like content optimization.  A tangible example would be if you have a massive canonicalization problem or a serious domain strategy issue.  If you do, then you better tackle those projects first before you simply optimize content on the site.  If the search engines cannot effectively crawl and index your content, you can forget about content optimization…  Also, SEO takes time before you see success.  Typically, you need to build up SEO power over time before you see a big change in rankings and organic search traffic.  Even if you fix all of the technical problems on your site, you still need to tackle several additional projects like content optimization, linkbuilding, etc.  That’s why efficient execution is critically important.  A delay in completing projects can impact months of SEO performance.

Based on what I’ve explained above, the Strategy to Execution Gap™ is a perfect fit for SEO initiatives.  Presenting the metric before you get started and then giving periodic updates on a team’s SEG percentage can get key stakeholders involved and on-board with what you are trying to accomplish.  And, they might even help you out by removing obstacles from your SEO path.  More about that soon.

The Strategy to Execution Gap (SEG) Formula
When you map out a series of projects as part of your SEO roadmap, they become the foundation for your Strategy to Execution Gap™.  Based on those projects, and the priority you give each project, you can explain to your team how the SEG will be calculated (and what the score means to their success).  Below, I’ll first present the formula and then present an example so you can see how it works.

The Strategy to Execution Gap = (Sum of Weighted Projects – Sum of Weighted Projects Completed  / Sum of Weighted Projects) * 100

Some notes about the SEG formula:

  1. Each project should be weighted from 0-10 and will be based on your analysis of the current situation.
  2. You can use all of the projects that are part of your roadmap, or just a subset for a given time period.  i.e. You can do this quarterly, semi-annually, or for the entire year.
  3. The SEG will end up being a percentage.  The lower your percentage, the better the score (meaning the gap to execution is low).  A higher percentage means your team is not executing at a high level (or there is a larger gap between strategy and execution).

An example:
Let’s say you had the following projects to complete (the weighted priority score is next to each project).  I’ll keep the example simple, but keep in mind that might have several more projects to tackle as part of an actual SEO roadmap.

Domain Strategy 8
Fix Canonicalization Problems 8
Refine Internal Linking Structure 6
XML Sitemaps 5
Keyword Research 4
Content Optimization 4
Video SEO 3
Linkbuilding 7

Total Weighted Sum: 45

Let’s say you are a few months into the initiative and three items have been completed (domain strategy, internal linking structure, and keyword research).  At your next meeting, you want to give the team an update on how things are progressing.  The SEG for this project so far is 60%, meaning there is a relatively high strategy to execution gap (which is not great).  The formula looks like this: (45-18/45) * 100 = 60%

At this point, you could discuss the various obstacles to execution to see why your SEG isn’t better.  Depending on the size of your team and the number of departments involved, you might be surprised to hear some of the obstacles popping up.

To show you how the SEG can be impacted (especially by the weighting involved), let’s say your team completed two additional projects during this time period.  For example, maybe canonicalization and xml sitemaps were taken care of (both with relatively high weighted scores).  If that was the case, your SEG drops to 31% (almost in half) showing a lower strategy to execution gap (which is good).  I find the SEG is a great metric for quickly showing an efficiency percentage, while also sparking conversation about potential obstacles involved.

How to Close the Gap (Decreasing the SEG)
Let’s face it, execution in online marketing is  everything.  Decreasing your SEG can have a massive impact on the success of your initiative (which can have an impact on targeted traffic, conversion, and revenue).   So, how do you close the gap?  As explained earlier, there are a number of obstacles that can inhibit your SEO projects, including technology, developers, egos, other initiatives, designers, branding, executives, legal, PR, etc.  My recommendation is to meet with key stakeholders before your initiative begins.  Explain what you are trying to accomplish and its potential impact on the bottom line.  Take everyone through the SEG and tie percentages to success (and revenue).  If everyone understands the big picture, you might be able to clear more paths, which can lower your SEG, and increase your chance of success.  Whether this approach will be successful for you depends on a number of factors, including the details of the specific initiative, the people involved, and culture of the organization.  One thing I’ve learned over the past 15 years is that data and metrics always help make a case.  Opinions and finger pointing rarely work.  The SEG can be one more metric that can help your cause.

Your Next Steps with the Strategy to Execution Gap™ (SEG)
The good news is that you can get moving with the SEG right now.  Whether you’re just starting an initiative or if you’re in the middle of one now, start to think about the specific tasks involved.  Then weight each project and explain the SEG to your team.  After everyone on your team understands how the SEG works, then expand your communication to key stakeholders.  As projects are completed, you can start to include the Strategy to Execution Gap™ in your presentations and progress reports.  Make the SEG a quick visual that represents how effective your team is.  Remember, using data and not opinion might end up winning over more people.

And who knows, maybe after the SEG has been used a few times to track efficiency, others in your organization might adopt the metric for their own initiatives.  Wouldn’t it be great if you were the person that introduced and sparked the SEG metric throughout your organization?  Just make sure you reference this post after you’re famous.  :)

GG

Tuesday, May 11th, 2010

Using Simple and Effective Keyword Charts for New SEO Clients

Paid Versus Organic KeywordsWhen I begin helping new SEO clients, I find it’s extremely important to first help them understand their current state before moving to more complex projects.  As part of the initial projects I tackle, a top-level keyword analysis can help clients understand which keywords are driving traffic to their site and how valuable those keywords are.  New SEO clients typically know that they already rank for some keywords, but they often don’t know the exact breakdown.  Understanding this breakdown is important, since you typically need to map out a series of projects that will be part of an SEO roadmap.  The projects could impact the next six months to a year of work, so you want to make sure everyone is on the same page and clearly understands the current situation.

Simple Charts, Effective Messaging
After completing a top-level keyword analysis, I find that some simple yet effective charts can help clients clearly understand their current status.  In addition, these keyword charts can help gain buy-in from key stakeholders in the organization and set the stage for future SEO efforts.  For example, the charts can demonstrate the ratio of paid to organic keywords, branded to non-branded keywords, converting to non-converting keywords, etc. Once you quickly build these charts, then you can dive into what the data actually means and how you can improve SEO results.

So, are you interested in learning more about the simple keyword charts I mentioned above?  I’ve got you covered, but you’re going to have to hop over to Search Engine Journal to read my latest post titled Simple Yet Effective Charts for Helping New SEO Clients Understand Their Current Keywords.  My post covers various scenarios you might come across when analyzing keywords for new SEO clients, examples of some of the charts, and what this can mean for clients.

As usual, if you have any questions, feel free to post them here or on Search Engine Journal.

Read my latest post on Search Engine Journal:
Simple Yet Effective Charts For Helping New SEO Clients Understand Their Current Keywords

GG

Tuesday, May 4th, 2010

How to Set Up Remarketing in Google AdWords [Tutorial] – Behavioral Targeting on The Content Network

Remarketing in Google AdWords

This week I decided to write about Remarketing in AdWords, which seems to be a confusing topic for some marketers.  Over the past month or so, I’ve received a lot of questions about both the concept and execution of Remarketing.  Let’s start with a quick introduction.  Remarketing is a new feature in AdWords that enables marketers to target people browsing the web (across Google’s Content Network) that have already visited their websites.  You can target these people with custom messaging, since you already know they visited your site (or specific sections or pages on your site).  It’s a form of behavioral targeting and can be a powerful way to “re-market” to specific visitor segments.

For example, if you run an e-commerce website that sells sneakers, you could target people on Google’s Content Network that visited specific sections of your website, such as the section containing running sneakers or cross trainers.  To clarify, you could provide specific advertising just for that visitor segment (and your advertising would run on other websites, not on Google.com).  Essentially, this is your second chance to get those visitors back to your site to act, so your messaging might be different than the first time you are in front of them.  It’s also worth mentioning that remarketing can potentially creep some people out, depending on the execution.  For example, an ad that says, “We know you visited {enter your website here}. Come back or we’ll tell everyone!” will probably creep prospective customers out.  Just because it’s possible doesn’t mean you should do it.  :)

Setting Up A Remarketing Campaign, A Step by Step Tutorial

Although Google has provided several articles in the AdWords Help Center about remarketing, I still don’t think there was a great example of walking advertisers through the setup (step by step).  Since it can definitely be a little confusing while you are setting up your first remarketing campaign, I’m going to provide the necessary steps below (with screenshots) of how to set up a remarketing campaign.  After the tutorial, I’ll provide some key takeaways, so you can be off and running.

Without further ado, let’s set up a remarketing campaign:

  1. Develop Your Remarketing Strategy:
    I don’t recommend jumping in and creating a campaign until you clearly understand what you are trying to achieve.  For example, who are you trying to target, what do you want to say to them, and what is the ultimate goal?  In its simplest form, remarketing enables you to target people that visited your site in the past (or visited certain pages or sections of your site).  Then you can “remarket” to them while they are browsing the web (across Google’s Content Network).  You can target them with either text or display advertising.  For this tutorial, we’ll use a fictitious e-commerce website that sells sneakers.  Let’s say that we just launched a new trail running sneaker, specifically designed for running outdoors.  We are driving prospective customers to a landing page for the new sneakers via a number of channels (paid search, email marketing, etc).  To keep this simple, our goal is to remarket to anyone that visited that landing page to get them back to the site.
  2. Let’s first add an audience by creating a remarketing list.  A remarketing list is essentially the visitor segment you want to target on the content network.  For our example, this includes any person that visited our landing page for our new trail running sneakers.  When you create a new list, you will receive a new snippet of code that needs to be placed on the landing page.  This snippet will “cookie” visitors that are viewing the landing page (in order for you to remarket to them later on).  Note, if you target a section of your site versus just a page, then you need to add the snippet to every page in that section.  The same goes for creating a remarketing list that contains all site visitors.  For that case, you would need to add the snippet to all pages on your site.
  3. In your account, click the Audiences tab and then click the “Add Audiences” button.  If you don’t see the audiences tab, then click the arrow located on the right hand-side of your tabs.  You’ll be able to “show” the audiences tab by selecting a checkbox.
  4. Finding the Audiences Tab in AdWords:
    Audiences Tab in Google AdWords

  5. Once you click the “Add Audiences” button from within the audiences tab, you will see two sections, one that says “Select An Ad Group” and then one below it that says “Add Audiences”.  Don’t select an ad group at this point (you haven’t created the campaign and ad group you will use for remarketing yet) and click the link for “Create and Manage Lists” to create a new audience (this link is in the “Add Audiences” section near the bottom of the page).
  6. Creating a new remarketing list:
    Create a New Remarketing List

  7. You will be presented with a new page.  Once you are on this page, click the “New Audience” button and then click “Remarketing List”.  Enter a name for your audience, a description, the membership duration (cookie length), and then select the radio button for “Create a new remarketing tag”.  The membership duration is how long you want the cookie to last for users that visited our landing page.  You can keep 30 days for our example.  Click “Save” when you’re done.
  8. Adding Details For Your New Remarketing List:
    Adding details for a new remarketing list.

  9. When you click the save button, you will see your new audience displayed in a table.  Chances are you only have one remarketing list (since we have only created one in this tutorial.)  If you have created others, they will also be listed.  There should be a column that says, “Tag/Rules” which contains a link that triggers your new snippet of remarketing code.  When you click the link, you can copy the code that needs to be added to the landing page.  You can place the code snippet before the closing body tag in the landing page </body>.  Again, we are only adding it to one page, but if you wanted to target visitors of a section of your site, you would need to add the snippet of code to all pages within that section.
  10. Triggering Your New Remarketing Code Snippet:
    New remarketing snippet of code in AdWords

  11. Congratulations, you have created your first audience (based on the remarketing list you just created).  Now you need a campaign and ad group that targets that new audience.  The new ad group will contain the ads that will be displayed to your visitor segment (the people that viewed our landing page and are now browsing the web after leaving our site).  Go back to your campaigns tab to view all of your online campaigns (you can click the “all online campaigns” link in the upper left-hand corner of the page you are currently viewing.)
  12. Create a new campaign and append “Remarketing” to the name of the campaign.  It’s not necessary to do this, but will help you quickly identify the remarketing campaign in the future.  Make sure the campaign only targets the content network.
  13. Selecting Content Network Only Under Campaign Settings:
    Selecting Content Network in Campaign Settings

  14. Create a new ad group and give it a descriptive name like “Trail Sneakers Remarket”.  Do not add any keywords to this ad group (since they aren’t necessary).  The ad group will only contain ads that will be presented to the visitor segment you are targeting.  This will be possible once you “connect” your remarketing list to your new ad group.  Set a proper content bid in your ad group settings (based on your knowledge of your current campaigns and ad groups).
  15. Now create your ads that will bring visitors back to your site! Be creative and provide messaging with the understanding that these people were already on the landing page for our new sneakers.  This is your second chance to get them to convert, so you should tailor your message knowing this.  For example, you probably don’t want to provide the same core message that you would present to a new prospective customer (one that never saw the sneakers before).  Maybe you could provide additional selling points for the new trail sneakers, or maybe you could provide a special offer, discount, or other incentive to get them to come back.  As always, you can test various ads to see which ones perform the best.This is also where you can go overboard and creep people out (like I mentioned earlier).  For example, “You liked our sneakers enough to visit once, come back or else!”  If you creep people out, you’ll hear from them alright (just not for the reasons you want!)  :)
  16. What About Visitors That Converted Already?
    You might be wondering if your remarketing ads will show up for visitors that did convert already. For our example, your ads will still show up. You would need to set up a “custom combination” if you want to exclude visitors that converted. I cover custom combinations later in this post, but the topic is bit more complex so I plan to cover that functionality in a later post.

  17. Once your ad group and ads are created, it’s time to connect this ad group with the remarketing list we created earlier.  To do that, make sure the new ad group is active by clicking on the new ad group in the remarketing campaign (as you normally would to edit settings, ads, keywords, etc.)  Click the “Audiences” tab after selecting your new ad group and then click “Add Audience”.  You should see the audience we created earlier in the list.  Click the “add” link located to the right of the audience list.  Then click “Save” to finalize the connection.
  18. Finding The Audiences Tab And Add Audiences Button In Your New Ad Group:
    Adding an audience from within an ad group

    Connecting Your New Ad Group With A Remarketing List:
    Connecting a remarketing list to an ad group.

  19. You’re done! You have successfully set up a remarketing campaign.  You can check back to see how many people are being added to the remarketing list by visiting either the ad group you connected the list to or by visiting the Audiences tab at the account level.  You will see the number of users added to the list in the “number of users” column for each audience.

Now that you know how to technically set up a remarketing campaign, let’s run through some key takeaways.  Depending on the site at hand and the strategy you develop, you will probably end up testing various scenarios to see how remarketing works for your business.

Remarketing Key Takeaways:

  1. When using remarketing, you can target all site visitors, specific sections of your site, or even specific pages.  If you are new to remarketing, you might want to start by remarketing to any person that has already visited your website.  Once you get the hang of it, then you can develop more complex remarketing strategies.
  2. You can also remarket to visitors based on actions they take on your site.  For example, visitors that did not convert on your site.  You could remarket to people that added items to their shopping cart, but never purchased the item.  How about people that just signed up to learn more about a specific product or service?  Some of these scenarios require that you set up “custom combinations”, which enables you to combine remarketing lists using Boolean operations (AND, OR, or NOT).  I can probably write an entire blog post about custom combinations, so I’ll save that for another day.  That said, you can read more about custom combinations in the AdWords Help Center.
  3. In addition to focusing on specific areas of your site or specific actions visitors take, you can also remarket based on channel or campaign.  For example, you could remarket to visitors from your email marketing campaigns.  Or you could remarket to paid search visitors, or just paid search visitors from specific campaigns.  As I explained above, I would start with more general remarketing campaigns and then get more granular and complex once you get the hang of it.
  4. Don’t forget to take advantage of the visitor intelligence aspect of remarketing.  Your ads will be presented to people based on something they did on your site in the past.  You can provide tailored messaging based on that intelligence.  But as I explained earlier, be careful not to cross the line and creep people out.  And that’s very easy to do by the way.

It’s Time to “Remarket”

I hope you now have a better understanding of what remarketing is and how to set up a campaign in AdWords.  As you can see by what we’ve covered in the post, remarketing can be a powerful way to provide targeted ads to people that were already on your site (via both text and display advertising).  Just be careful with your messaging… I’ve already seen some companies use remarketing and creep people out from a privacy standpoint.  There’s definitely a fine line between effective marketing and turning off prospective customers.  Now go ahead and join the behavioral targeting revolution.

GG

Sunday, April 25th, 2010

A Baker’s Dozen: A Quick Update on Kati’s Kupcakes, The Winner of The Search a Small Business Holiday Giveaway [PODCAST]

Kati's Kupcakes New CMS and Website DesignIf you’re a frequent reader of my blog, then you probably remember the Search a Small Business Holiday Giveaway I launched this past December.  The purpose of the contest was to give an ultra-small business in New Jersey a free online marketing audit, which would produce plan for enhancing the company’s digital strategies.

I launched the contest because it didn’t seem fair that many small businesses lack the resources or budgets to tackle online marketing the right way.  Many SMB’s move at light speed to keep their businesses running and simply don’t have time to keep up on the latest strategies and tactics.  I opened the contest to any small business in NJ with less than ten employees (what I categorized as an ultra-small business).

The winner of the contest was Kati Angelini and she owns Kati’s Kupcakes, a gourmet cupcake business in Moorestown, NJ.  Kati’s Kupcakes creates specialty cupcakes, cupcake towers, wedding cupcakes, etc.  When I first spoke to Kati after announcing the winner, I could tell right away that she was passionate about her business, and was eager to learn more about SEO, Local Search, SEM, Web Analytics, etc.  It also became apparent to me that Kati represented the classic case of someone that had serious skills and was extremely talented, but lacked the exposure she needed to rapidly grow her business.  So, after my initial conference call with her, I got started on analyzing her business, her site, her current marketing efforts, etc.

What I Found and How I Helped (And We’re Not Done Yet…)
During my audit, I began to analyze Kati’s website, her presence in Local Search, her rankings in Natural Search, her tracking capabilities, etc.  It didn’t take long to understand the path we needed to go down.  In a nutshell, the content management system (CMS) Kati was using to run her website was wreaking havoc on her efforts.  The website had a rudimentary design, it was riddled with technical problems, her content wasn’t optimized (at all), she had no presence in local search, and lacked the ability to quickly add new content or blog posts.  For a local small business, the combination I listed above was causing serious problems for gaining exposure, rankings, quality visitors, and customers.  When you boil it down, Kati needed to show up when people were searching for her services.  At the time, she simply wasn’t.

The Marketing Spot with Jay Ehret
When I launched the holiday giveaway, Jay Ehret from The Marketing Spot reached out to me to learn more about the contest.  In case you don’t know Jay, he focuses heavily on small business marketing and branding and he also runs a podcast called Power to the Small Business.  Jay learned about the contest via Twitter (since we have been following each other for some time now).  Jay recently invited me to be on his podcast to speak about the contest, the changes we implemented for Kati, and the results from our initial efforts.  Both Kati and I were on the call and we had a great conversation with Jay about the project.  The podcast launched today and you can listen to it now by visiting the blog post (listed below) or by downloading the podcast via iTunes.

Listen to the Podcast About Kati’s Kupcakes:
Local SEO Clinic: Improving Traffic, Improving Business

Kati’s Kupcakes: Highlights From Our Initial Efforts
Although we are only a few months in, there have been some great improvements marketing-wise.  Kati still has a long way to go, but our initial efforts are definitely paying off. Again, you can listen to the podcast to learn more (it’s 30 minutes long).  Here are some highlights from the project and I’m eager to keep helping Kati grow her business.  By the way, you can hear Kati’s view of the projects on the podcast (including how they have impacted her business.)

  • I performed a thorough SEO technical audit, which revealed a number of technical issues impacting her performance in Natural Search.  There were also a number of content optimization problems on the site (to say the least).  The deck was 30 slides long…
  • We moved Kati’s website from a problematic content management system (CMS) to Wordpress, which helped fix a number of technical problems inhibiting her efforts (usability, SEO, blogging, domain strategy, etc.)
  • Kati’s indexation (the number of pages indexed by the search engines) has increased from 12 unoptimized pages to 136 pages.
  • I developed a local search strategy for Kati’s Kupcakes in order for the business to show up for local searches.  This included making a number of changes and additions across several websites and databases.
  • I started training Kati on best practices for Search Engine Optimization (SEO).  Kati is the person that would be creating new pages, new blog posts, etc. and needed to understand how to properly optimize those pages on her site.  This took the form of multiple one hour training sessions (and the training is on-going).
  • I explained the power of the long tail of SEO, as well as how to target it.  The long tail is incredibly powerful and businesses should avoid the long tail at their own peril. A bit dramatic?  Sure, but definitely true.  :)
  • Kati’s original website ranked for almost no target keywords.  Since March 1st, 622 keywords have led visitors to Kati’s website, 561 of those keywords were unbranded terms (not someone simply searching for Kati’s Kupcakes), and 128 of those keywords included a local qualifier.  For example, using a town, zipcode, or state along with target keywords.  Down the line, I believe that Kati’s website can rank for thousands of keywords, but 622 is a great start (and a huge improvement.)
  • We added Google Analytics for tracking site performance, which enables Kati to view granular reporting across traffic sources, campaigns, keywords, etc.  I am going to help Kati develop an analytics strategy, including identifying conversions and events to track, so she can quickly and efficiently identify which efforts are paying off for her business.


A Good Start, But There’s Still A Lot To Do:

Although we have made a lot of changes and Kati’s business is seeing a nice impact, we still have some things to do marketing-wise.  For example, there are several content optimization projects we need to perform on the site.  I also want to help Kati learn more about paid search and understand how it can supplement organic search.  I think Kati can have a stronger integration with Facebook and Twitter, and possibly start to use location-based services like FourSquare or Gowalla.  And I want to keep training Kati on SEO best practices so she feels comfortable with making changes that will help her organic search efforts.  In addition, I mentioned that I’ll be helping Kati develop an analytics strategy.  That’s probably our next major step.

So, if you are interested in hearing more about the contest and the projects we tackled, head over to the Marketing Spot and listen to the podcast.  Again, it’s about 30 minutes long, but contains some good nuggets of information for small businesses. Heck, put in on your iPod and listen to it during your next workout.  :)

GG