Google Adwords data is included in your Google Analytics data and includes metrics like click through rate (CTR) and cost per click (CPC). Google recently released AdWords API v201101, which allows you to more efficiently run reports, as well as implement campaign experiments and other recently released advertising features at scale.

But how do you track success after PPC visitors are on your site? What metrics should you investigate to ensure that you are getting high quality traffic and capitalizing on opportunities to convert?

For these types of performance metrics, you need web analytics! Here are 4 important metrics you should review on a daily basis to evaluate PPC campaign performance.

1. Conversions

It pretty much goes without saying that conversions are the best metric to determine how a PPC campaign is performing. You should have your web analytics set up to record both online conversions (newsletter subscriptions, content downloads) and offline conversions (phone calls, offline campaigns).

To track online conversions, configure your web analytics to record a conversion every time someone arrives on a specific URL. In the case of web forms, this URL would be something like a thank you or confirmation page. For downloads, you might need to add a piece of tracking code that will register the download as a pageview.

To track offline conversions, see one way of doing this in our previous post on Measuring Success of Offline Campaigns in Google Analytics. There are plenty of other solutions out there for tracking other offline conversions (e.g. by telephone) which will integrate directly into your web analytics program so you only have to access one dashboard.

2. Bounce Rate

A bounce is when someone lands on a site and leaves without viewing any other pages. Your bounce rate will vary for each campaign. A high bounce rate may be an indication that your content is not relevant or engaging to visitors.

3. Pages Per Visit

The interesting data comes from a very low or a very high number of page views. Very low could mean that visitors are not finding content useful or interesting, and have resigned to go back to search results to find a more relevant page. A high number of page views could mean either you are producing interesting and engaging content (look at time spent on page for engagement), or that the visitor cannot find the content they are looking for.

In both cases, review the relevance of the page content to traffic-producing keywords, and make sure the information people appear to be seeking is on the landing page, or a click away.

4. Average Time on Site

It goes without saying that a higher time on site is better than a low one. Extremely low (0-1 second) — There is no way to read a page’s content in this amount of time. If there are a lot of visitors spending less than a second on the site, it may be the result of one of two things:

• Invalid clicks – Check with your PPC platform to ensure you are not being charged for these.
  Slow site load time – May cause people to get frustrated and hit the back button before ever arriving on the landing page.
• Low (less than 15 seconds) — Generally, those visitors who spent 10 seconds or less on a site quickly decided that they were in the wrong place. This may be because at a first glance they didn’t find any relevant information, see their keywords anywhere on the page, or were confused by the landing page’s layout.

Look at these 4 web analytics metrics and you will have a better idea of your PPC performance. Once you have gathered enough data to draw conclusions about which parts of your campaign work well and which don’t work so well, you can start implementing small changes and tracking the different outcomes. By taking this methodical approach you should be able to optimize your campaigns to get the best ROI.

This is a guest post written by Kirsty Lee from Bime Analytics, an SAAS business intelligence and data visualization service.

What Is A Tracking Parameter?

A tracking parameter is information appended to a URL that provides additional data about the visitor in your preferred web analytics tool. For example, here’s a tagged landing page for a Yahoo Search Marketing campaign.

Original URL:
www.xyz.com

Tagged URL:
www.xyz.com/?utm_source=yahoo&utm_medium=cpc&utm_term=blue%2Bwidget &utm_content=warranty&utm_campaign=widget%2Bcolors

What does this string of data mean?

Campaign Source (utm_source):

In the example the visitor source is from Yahoo. This parameter is supposed to distinguish different sources from the same channel. For paid search it might be Google, Yahoo & Adcenter. A newsletter might separate sources by segment (b2b vs consumer) or websites where you’ve placed the same banner ad.

Campaign Medium (utm_medium):

In the example the visitor arrived from a Yahoo paid search ad (cpc). This parameter designates channels, such as email, banner, blog, affiliates, etc.

Campaign Term (utm_term):

In the example Blue Widget is the keyword being bid on in Yahoo Search Marketing, not to be confused with the visitors actual search query.

Campaign Content (utm_content):

In the example, warranty is a reference to the specific text ad the visitor clicked to arrive at the site. It could also be set to adgroup if you’re not looking for that level of granularity in your pay per click campaign. For an email, you could use this parameter to distinguish the performance of individual links in the message.

Campaign Name (utm_campaign):

In the example, Widget Colors is the campaign that contains the warranty text ad and keyword blue widget. A campaign is the only parameter that can be common among different sources and mediums. A sale on blue widgets might be promoted in an email, a dedicated pay per click campaign, and by your affiliates. You’d then be able to view the aggregate performance of the campaign under the Google Analytics Traffic sources tab.

While I’ve listed all the campaign variables here as a demonstration, you don’t have to use them all when tagging your links. At a minimum you can use Source and Medium to track your campaigns. To prevent typing errors, and quickly build out tagged links for your campaigns you can make use of Google URL Builder.

Issues With Tracking Parameters

Capitalization

When naming sources and mediums always use the same letter casing. If you tag two newsletters with mediums “email” and “Email” these will show up as two separate mediums in analytics. Choose a letter case and stick with it for sources and mediums.

Length

You probably noticed that adding tracking tags to a URL makes it really long, and ugly to look at. This poses a potential click through problem for plain text emails, or messages on Twitter with the 140 character limit. A way around this is to use one of the many URL shortening services.

Duplicate Content

Google views the original URL, and tagged URL as separate pieces of content. So when you use tracking parameters extensively, and people start linking to both URLS it dilutes the power of your link equity and search engine rankings. To avoid this issue change your query parameter from a question mark (?) to a hash tag (#). However, this will require modifying your Google Analytics tracking tag to pick up the hash tags.

301 Redirects

A 301 redirect of your landing page will strip all your tracking parameters. To prevent this from occurring the URL should point to the final destination page after the redirect. Or the server will have to be modified to pass tracking parameters.

Photo credit The Cosmic Cat

When someone goes through the checkout process on our site the url’s look something like:

1) /cart
2) /session/new?return_to=%2Forders%2Fnew
3) /orders/new
4) /orders/3575/payment
5) /orders/3575/confirmation

Notice the dynamic url’s in numbers 3 and 4? After doing a little research I stumbled upon an informative post by Lunametrics, the comments section especially has a wealth of information about setting up goals with dynamic url’s. However, the examples provided didn’t fit my exact situation so I had write the following regular expressions.

1) ^/cart
2) ^/session/new
3) ^/orders/(?=new)
4) ^/orders/b0*[1-9][0-9]{0,8}b/(?=payment)
5) ^/orders/b0*[1-9][0-9]{0,8}b/(?=confirmation)

This will work for a range of numbers from 1 – 1,000,000,000. When we reach 1 billion orders I plan on putting a big flashing neon colored sign on the website announcing the fact. Just like McDonald’s does on the Golden Arch.

Be sure you’ve set Match Type to Regular Expressions in the goal information settings.

Hopefully this helps anyone who finds themselves in a similar situation.