There is an old adage in Pay per Click lore: keep your clicks close, but your conversions closer. To do this you need to know how to gather useful and accurate conversion data. This was bellowed at me over and over through a megaphone by our AdWords Guru while I completed the leopard-crawl segment of our basic training. Man, those landmines were tough to spot.
When she finally took her combat boot off the back of my head, I managed to throw in a few questions along with some hand-grenades. Impressively, I managed to fend her off long enough to hear the answers. This is what I remember learning before lapsing into sweet unconsciousness:
- While click data is useful for sites that nominate traffic as their primary goal, conversion data is usually considered the Holy Grail of PPC. Therefore, setting up AdWords and Analytics goals properly is of paramount importance (especially if clients are charged with implementing the conversion codes). Unfortunately, there are a wide variety of situations where it is not as simple as just pasting code onto the right pages.
- It is also worthwhile remembering that Analytics and AdWords conversion data will not be 100% identical, so unless there is a big discrepancy in figures, things should be OK.
- Remember, Analytics goals are not like AdWords goals. Every AdWords goal needs its own piece of code generated by AdWords. Each piece of code is then put on the page that signifies a conversion has occurred. (Usually this is a thank-you page of sorts that can only be accessed by performing a certain action - like submitting your information through filling in a form). Analytics code goes on all pages, and goals are differentiated by user-inputted URLs. This makes URL tracking of prime importance.
- If a thank you page has dynamic URLs (those that change according to what data you submitted), then you need to select the Head Match or Regular Expression options in Analytics.
- Head Match: If your URL is always the same for this step of your funnel, but is followed by unique session or user-selected data or identifiers, use the Head Match filter and leave out the unique values. Thus, www.website.com/transactions-done=?Purchase=0&National=1 and www.website.com/transactions-done=?Purchase=1&National=0 can both be recordable if you enter in just www.website.com/transactions-done and set to “Head Match”.
- Regular Expression Match: Uses regular expressions to match your URLs. This is useful when the stem, trailing parameters, or both can vary between users. For example, if a user comes from one of many subdomains and your URLs use session identifiers, use regular expressions to define the constant element of your URL.
- A different problem to dynamic URLs is when URLs are the same for multiple pages or links. This will make it difficult to put an Analytics funnel in place that can be helpful since Analytics only differentiates pages through user-inputted URLs. This usually happens when on-click events form part of the goal process (e.g. checking the “Licence Agreement” box is a step, but would not take you to another URL (which can then be recorded as a funnel progression).
- To get around this, you can put a piece of code on the desired page that will give the URL a unique name for the purpose of Analytics. This code will most likely have to be inserted within an on-click event (or anchor text link) within the page’s HTML code. The function is called “_trackPageview” and should be applied to each step that takes place on the same page.
Gathering the right conversion data makes the difference in running a successful campaign. (Image by Proptology under CC)
Of course, these are just the basic (if important) steps in obtaining accurate conversion data for relatively uncomplicated websites. There are many more tweaks, code changes and modifications out there to help with more complex commercial tracking and so forth. But until next time, comrades.
Did you enjoy this post? Why not check out:
