 |
|
|
|
Posts Tagged ‘search engine marketing’
Jan 13
2009
You are working on your site’s SEO by publishing press releases and you wish to track traffic to your site from those press releases. You are not adding source campaign parameters (and therefore no campaign parameters at all) to your links because you are not sure which sites will pick up your press release.
Sounds familiar, doesn’t it?
In Google Analytics, you noticed that links from press releases are tracked as:
- Source = cnn.com, bbc.co.uk, or domain.com
- Medium = referral
- Campaign = (not set)
I am sure you are not satisfied with this basic level of tracking because it does not tell you much, especially if you wish to track across different campaigns and mediums.
The following example makes more sense and will help you evaluate and analyze your campaigns.
- Source = cnn.com
- Medium = press_release
- Campaign = hurricane_katrina
To overcome this challenge of tagging links from unknown sources, I came up with the following trick.
Algorithm:
- Add a parameter on all links to your site that are in the press release. (example: http://www.mysite.com/myfile.html#id=1)
- On the target page (http://www.mysite.com/myfile.html), check the value of the “id” parameter.
- If the “id” parameter equals “1″, replace the parameter in the URL with the following utm parameters (utm_source, utm_medium and utm_campaign) before the call to pageTracker.
- If the “id” parameter does not equal 1, call the pageTracker function normally.
View the entire code segment.
Let us explore the code, section by section:
var parameter = get_parameter('id');
function get_parameter(name)
{
name = name.replace(/[\[]/,"\\\[").replace(/[\]]/,"\\\]");
var regexS = "[\\?&#]"+name+"=([^&#]*)";
var regex = new RegExp(regexS);
var results = regex.exec(window.location.href);
if( results == null )
return "";
else
return results[1];
}
This portion of the code will return the value of the “id” parameter from the URL and assign it to the “parameter” variable.
if (parameter == '1')
{
window.location.hash = "utm_source="+srcPage+"
&utm_medium=press_release&utm_campaign=hurricane_katrina";
}
If the page url contains the “id” parameter and its value is 1, then the url will be updated with the utm parameters.
Link URL
http://www.mysite.com/myfile.html#id=1
New URL
http://www.mysite.com/myfile.html#utm_source=www.cnn.com&utm_medium=press_release&utm_campaign=hurricane_katrina
* Notice that we did not use window.location.href function because this function will re load the page with the new URL, which is not what we want to happen. We just want to update the URL, without affecting the visitor experience, in order for the Google Analytics tracking code to attribute the visit in a certain way.
How to get the value of the utm_source (referral site)?
var srcPage = getDomain (document.referrer);
function getDomain (thestring)
{
var urlpattern = new RegExp("(http|ftp|https)://(.*?)/.*$");
var parsedurl = thestring.match(urlpattern);
return parsedurl[2];
}
This portion of the code is responsible of assigning the URL of the referral site to the variable scrPage. The “getDomain” function parses only the domain name (www.domain.com) out of the long url string (http://www.doman.com/file.html?parameter=abc)
The last piece of code that needs to be added is the setAllowAnchor command, which allows the # sign to be used as a query string delimiter instead of the question mark (?).
We used # in the press release link instead of ? for SEO reasons, but you could use ? in the original link and still use the above method.
pageTracker._setAllowAnchor(true);
Alright, now it is time to use our friend Advanced Segments to track our press release visitors, measure their engagement, and analyze their behavior.
Now we can really analyze! 
Technorati Tags: google analytics, search engine marketing, search engine optimization, web analytics
Tags: google analytics, search engine marketing, search engine optimization, web analytics
Posted in web analytics | 24 Comments »
Jan 02
2009
Questions I always hear with regards to SEO and marketing optimization:
- How does Google Analytics improve my search engine optimization (SEO)?
- How can I get more out of my SEO?
- What is the real effect of ranking on search engines for my business?
I am hoping the case study below will shed some light on these questions. If you apply a similar analysis you can help your customer, manager, or whoever is delaying your SEO effort. But instead of answering “the importance of ranking” question, maybe something more quantifiable and measurable might get your decision makers’ attention! How about “how does ranking on a specific keyword, or lack thereof, impact the bottom line?” Answering such questions will help us make the most of our marketing spend during these tough economic times and help us do a more effective job in marketing and campaign optimization.
I’ll use real time data and analysis but won’t mention the name of the website for privacy reasons. Here are the details:
- Website type: eCommerce
- The website used to rank near the top of Google on two competitive keywords until August 2008.
- Historically, these two keywords have driven traffic to the site. For a specific time period, these two keywords drove 5,684 visits and led to 46 conversions for a 0.81% conversion rate.
If you have your Google Analytics eCommerce features properly configured and working, the above data is easily accessible under the Traffic Sources -> Search Engines -> Non-paid report. The inline filter was used to get data for just these two keywords.
- We then start examining the time period when the ranking for these two keywords took a big hit.
- Next, using the date comparison function in Google Analytics, we compared the traffic generated by these two keywords for this year with poor ranking versus last year with better ranking. Here is the result:
The table on the left is for one of their keywords and the table on the right is for the other. As you can see, a significant drop in visits in 2008, 4471 to be exact. So the negative impact of the drop in SEO ranking was less opportunities to make sales on their primary keywords!
- The 4471 visits might be a small percentage of the overall website traffic but when you put a dollar sign next to it, we typically react to it more quickly. Take the 4471 visits and multiply it by your average conversion rate for these keywords, which is 0.81%, and then multiply by the average order value, which is $846.
- 4471 x 0.0081 = 36 lost sales
- 36 x $846 = $30,456 of lost revenue!
Now one can argue that this number is not accurate because of many factors BUT the findings are very actionable! If I were to present this analysis to my boss or client, I would add 2-3 other scenarios:
- Scenario 1, with a higher conversion rate of 1.62% after improving the design and usability of the site, the lost revenue would be $60,852. (ouch!)
- Scenario 2, with a lower average order value and the existing conversion rate of 0.81%, our lost revenue would have been $15,228.
- You could create a table to show the range. The main point is that there was between $15K and $60K of lost revenue. In tough economic times, wouldn’t you rather have that revenue?
Depending on your company size, marketing budget, and other factors, the $30K might be a significant number or it might be a rounding error. But at the end of the day, $30K of lost revenue is $30K of lost revenue, especially in times like these where cutting cost and marketing optimization is more important than ever. By doing similar analysis, you can find other lost sales opportunities, monetize them, and get some corrective actions underway.
I think you can take this analysis to a business owner or marketing manager, and I am pretty sure they would get the SEO effort prioritized.
Technorati Tags: google analytics, search engine marketing, search engine optimization, web analytics
Tags: google analytics, search engine marketing, search engine optimization, web analytics
Posted in search engine optimization | 5 Comments »
Dec 12
2008
In September of this year, Yahoo! announced the formation of a Digital Advisory Council to drive discussions with advertisers. Our agency was invited and we welcomed the invitation to listen to what Yahoo had to say and provide some feedback. Our office being literally less than ten minutes away made the decision a little bit easier!
During the meeting we were told that we are under an NDA so I won’t go into the specifics. But here are some comments and thoughts:
- We applaud Yahoo’s efforts to be open, reach out to their clients (advertisers and agencies), and solicit their input. That spirit was felt throughout the council meeting, from the Senior VP that hosted the event to managers and other Yahoo staff. They really wanted to be transparent and sought input from us. Good job, Yahoo!
- Despite the poor picture that might have been portrayed in the news about Yahoo and search, we were reminded that Yahoo! is still a profitable company and has some very high traffic sites, including Yahoo! Sports (we were told it has more traffic than ESPN) and Yahoo! News (gets more traffic than CNN). I didn’t know that and I would say it is pretty impressive!
- Various folks from Yahoo! presented about new features and innovations (sorry, can’t blog about it yet
 ) but expect to hear from Yahoo! on areas they are strong in, including Display (banner) Advertisement.
- There was a talk about Traffic Quality and the favorite discussion topic of click fraud. Without spilling their beans, we got a better picture of how Yahoo! handles click fraud and also learned that you can get reports from within your Yahoo! Search marketing account about credits your accounts is getting for false clicks.
- Lunch was really good too, including the Israeli Couscous which was very tasty! To be fair, I have to say that the food in Mountain View during the GAAC Summit was really good too.
So in summary, I am really glad to see Yahoo! opening up and creating additional forums to communicate with advertisers and agencies.
While we are on the subject of Yahoo! Search Marketing, there is another topic that we have been wanting to blog about but didn’t get to until now. Yahoo! has a mechanism of optimizing your accounts and campaigns on their own! Yes, that means without your
knowledge. Maybe this is not so recent but many clients I speak with are still not aware of this mechanism, so it is important that we share it. The following was drafted by a couple of folks in the office:
We hope you are aware of Yahoo’s decision to optimize PPC accounts by creating optimized campaigns on their own and running them without checking with the client. As strange as this may sound, it is very true! For a regular advertiser who doesn’t tag his campaigns and review them in any analytics tool, this may not be that bad. For those who do tag their campaigns and review all of their traffic sources and conversions via an analytics tool, this could be trouble.
Yahoo does the whole nine yards when it comes to campaign creation (naming convention, ad group, keywords, ad copy) but they miss a very crucial step. They don’t tag the campaign at all with tracking parameters so a web analytics tool can recognize it as PPC traffic. This means traffic from this new campaign will be recorded as organic traffic and will dilute the quality of your data. The number of visits attributed to Yahoo cpc and Yahoo organic will be wrong, and conversion rates for these two sources will be subject to speculation.
There are two ways that you can deal with this issue:
- If you would like to try these optimized campaigns and see how they will perform while capturing the data properly in your analytics account, you need to go to your Ads under the AdGroup they created and make sure that you tag the destination URLs properly. If you are using Google Analytics, here is a good post from our friends at PPC Hero that will help you.
- Otherwise, once you see a new optimized campaign in your account, pause or delete it and make sure that you opt-out from this service. You can opt-out by submitting a request to Yahoo customer support.
If you have too many campaigns in your own or your client’s accounts, the easiest way to spot the Yahoo optimized campaigns is by their naming convention. See the below snapshot for an example.
My request to Yahoo! is that they back out of this practice and allow user more control before they turn campaigns on and spend someone else’s money!
Technorati Tags: google analytics, pay per click, search engine marketing, web analytics, yahoo search marketing
Tags: google analytics, pay per click, search engine marketing, web analytics, yahoo search marketing
Posted in pay per click | No Comments »
Dec 01
2008
Here at E-Nor, we recently completed a project for the The Learning Community (TLC), which happens to be a collection of links to informational websites, articles, and videos based on different subjects that affect child development. Their mission – to provide that “children’s manual” parents never seem to receive with the baby!
The project initially started in December of 2007 as a conversion of their original pure HTML site to Dotnetnuke (open-source Content Management System), but ended up turning into a significant redesign and restructuring project.
The services we provided were:
- Basic online marketing consultation
- Creating a new, brighter aesthetic look.
- Implementing a different site-structure based on our understanding of their users flow. (We also took some tips from their more successful and professional competitors, such as the commercial magazine www.parenting.com)
- Improving their SEO (Search Engine Optimization) by cleaning up their meta tags and recommending some content enhancements. Their site is now on the first page of Google for some keywords when previously it was nowhere to be found.
- Helping promote their videos on Youtube.
- Cleaning up their Google Analytics setup to properly track where visitors are coming from, which external sites they are going to, and which PDFs they are downloading.
Oh yah, I forgot to mention the project was done PRO BONO. We had a great working relationship with the client, in that any work we recommended, if they could find volunteers to implement, they did, which saved us time. Any technical implementation we could throw in, we did, and they practically understood that since the work was pro bono, it would take priority accordingly with respect to our other projects.
E-Nor encourages our clients, partners, and blog readers to support non-profits. Though altruism may already be inline with your corporate and personal values, a year’s worth of pro-bono work may scare even the most giving of companies and people. However, here are some benefits you may not have considered (in no particular order):
- Necessity is the mother of innovation. The nature of non-profits is that their revenue is limited yet they provide great services to the community. Thus, they may require strong functions for their site. You’ll be forced to learn valuable work-arounds when their budget may not cover high end modules or spending, giving you great ideas for options when you need to close a sale with those paying clients who are a little tighter with their money.
- Practice makes perfect. Just like anything you do in life, the more you do it the better you will get. You can chalk this pro bono run as practice. For us, TLC being a year project, it strengthened and even expanded our research on techniques, functions, modules, etc, that we can now apply to all our sites!
- A non-profit “word-of-mouth” could still lead to profit. We know that as technology evolves, so does marketing. If Google has taught us anything, free services and products actually go a long way in branding and exposure, and could result in lucrative opportunities in the future. Non-profits do have friends that could end up being your paying clients with the right referral. And because your existing relationship had no financial motives, the trust and rapport has already been built.
- Had a bad day? They’ll pick you up. For all those clients who didn’t see the extra work you did for them and who complained instead of showing gratitude, you could expect the opposite for your non-profit pro bono clients. They can’t help but see the void you filled for them and be grateful for it.
- Testimonials. Along the same lines as the above, a testimonial will virtually be an everyday occurrence if you’re doing your job right.
- Pat yourself on the back. You did a good deed! Because of the site you provided for a parenting non-profit or a domestic violence shelter, a lost parent now has a little bit of direction or a helpless victim is a little bit safer.
- Experimentation (with the permission of the client). Since your client doesn’t have dollars riding on this project, they are more likely to allow you to do light experimentation on it, within reason of course. Not only are they more comfortable since no hard earned grant money is at risk, but that also means there’s a more flexible timeline for you to play with the site. For example, if you see a new module you wanted to try or your organization is new to analytics and you need a site to try it on, especially if the end result could possibly benefit your client, non-profit free sites may in fact welcome experimentation. Don’t forget to back up, though!
Technorati Tags: dotnetnuke, search engine marketing, search engine optimization, web design
Tags: dotnetnuke, search engine marketing, search engine optimization, web design
Posted in general | 1 Comment »
Nov 17
2008
| Back in November of 2006, our friends at GA Experts posted a method to get detailed PPC keyword data from Google Analytics. This method involved creating a seperate profile and using some intimidating but powerful filters. I really liked and referenced this method because it isolated the changes to a single profile and didn’t involve any updates to the GA tracking code. Obviously I would prefer if this feature came built-in to GA but at least now you could get to some very useful data with very little work. |
 |
As Google Analytics was updated over time, the way the filters were using Custom Field 1 was no longer possible and so the profile broke. The smart folks over at ROI Revolution came up with another way to get to the exact PPC keyword data using the User Defined field and adding 2 lines to the GA tracking code. This was a great solution and probably suitable for a lot of people. However, I preferred not having to update the tracking code (this can be very difficult on some client sites), saving the User Defined field for other uses, and also being able to isolate the keyword extraction and concatenation to one profile.
I still see some blogs referring to the original method that doesn’t work anymore so one day I tried playing with the original filters approach and after 2 months of observing the data, it appears to be working correctly. Please follow these screenshots & instructions to give it a try for yourself.
Filter 1 – look at only paid traffic data.
- From your account, click on Filter Manager.
- Click on Add Filter.
- Give it a name such as Data Filter – PPC.
- Filter type is custom filter.
- Select on the Include option.
- The filter field should be Campaign Medium.
- The filter pattern can be cpc|ppc
- Click on Save Changes.
Filter 2 – extract the actual keyword that was typed and concatenate it to the paid keyword that was matched.
- Assuming you are still in the Filter Manager, click on Add Filter.
- Give it a name such as Data Filter – PPC Keywords.
- Filter type is custom filter.
- Select the Advanced option.
- Field A -> Extract A should be set to Referral and the value needs to be (\?|&)(q|p|query)=([^&]*)
- Field B -> Extract B should be set to Campaign Term and the value needs to be (.*)
- Output -> Constructor should be set to Campaign Term and the value needs to be $B1 – ($A3)
- Field A required, Field B required, and Override Output Field all need to be set to Yes.
Please click on the screenshot to view the full version.
Finally create a new profile with a name such as PPC Keywords Data and all your usual settings (default page, search settings, goals, etc). Make sure you apply these two new filters to it.
This method seems to work for Google, Yahoo, and MSN paid traffic. Below is a screenshot with sample results. Let me know if it works for you!
Technorati Tags: google analytics, msn adcenter, pay per click, search engine marketing, web analytics, yahoo search marketing
Tags: google analytics, msn adcenter, pay per click, search engine marketing, web analytics, yahoo search marketing
Posted in web analytics | 34 Comments »
|
|
 |
|
|
|
|
|
|
