April 2009

Tracking your online direct response results
A variety of simple, free methods exist to determine your online campaign’s effectiveness. Here is a cross-section of them By Jay Aber

Just when you thought it was getting easier to track results from your online campaigns, reality strikes. You get conflicting advice. The experts differ. Which tracking methodology is the “gold standard?” Who’s right? Who’s wrong?

There is no “perfect” way to track results from online campaigns, just as there is no perfect way to track the results from traditional direct marketing activities. But that doesn’t mean you should give up on tracking results from your online direct response campaigns – especially since there are a number of simple and free ways to determine your campaigns’ effectiveness.
In this article, I’ll discuss various ways to track online marketing activities and provide the pros and cons of each. To keep things simple, an “ad” refers to any link, banner ad, solo e-mail, text link and search result that is clickable by a recipient or viewer. A conversion or transaction refers to the desired event (sale, lead, opt in etc.) resulting from your marketing campaign.

Tracking by impressions
Definition: Tracking methodology counts how many ads are shown to Web site visitors, searchers or e-mail recipients during your campaign.
Pros:
Very simple methodology that Web publishers and list owners provide automatically to advertisers. Valuable in determining whether a campaign that was purchased on an ad impression basis delivered properly. You can use the impression delivery statistic to calculate more traditional advertising metrics such as reach, frequency and share of voice.
Cons:
Provides no information about response to the ads, or conversion data.

Tracking by clickthroughs
Definition: Tracking methodology counts how many times users clicked on an ad or link.
Pros:
Very simple methodology that Web publishers and list owners provide automatically to advertisers. Clickthroughs are better indicators of interest in the advertised offer and they have greater likelihood of response than impressions, since users who click on an ad are directed to a landing page on the advertiser’s Web site – one step closer to a conversion or sale.

Clickthroughs indicate relative interest in an ad since by definition, a clickthrough only occurs when a user clicks on an ad to find out more about the advertised offer. Ads with a higher clickthrough rate (total clickthroughs/total impressions served) generate more interest in the offer than ads with a lower clickthrough rate.By comparing clickthrough rates, advertisers receive information about the relative appeal of various offers, reaction to creative executions, and differences in Web site audiences.
Cons:
Unfortunately, like ad impressions, clickthroughs to a Web site do not tell a marketer anything about whether the user completed the transaction (purchased, enrolled, sampled, downloaded, entered a contest etc.).

Tracking by Unique URLs
Definition: Clickthroughs on ads are directed to a unique URL or Web site. Since the user has no other way of arriving at that specific URL apart from clicking on an ad, Web analytics packages are able to track activity after users arrive at the unique URL.
Pros:
This approach is relatively simple to track through basic Web site analytics – such as Google Analytics, which is a free Web analytics system. Unique URLs, plus impression and clickthrough tracking provide information about the strength of your offer and some of the conversions generated by your campaign.
Cons:
This system under counts conversions. A visitor who lands on a unique URL but leaves the site before completing the transaction, yet returns at a later time via your home page and completes the transaction will not be attributed to the ad campaign. As most conversions do not happen in the same user session as the initial clickthrough, research shows that tracking by unique URL under counts results by 50 to 80 percent. This is especially true for complex transactions such as those that require a user to spend time researching before purchasing, e.g. buying a trip online.

Post-clickthrough & post-impression tracking
Definition: This approach tracks conversions from people who saw or clicked on your ads, and converted within a 30 day window. It requires a third-party system such as DoubleClick’s DART or Google Conversion Tracking.
Pros:
This method is relatively simple, free (or very low-cost) and does a good job of tracking the full value of an advertising campaign.

It requires that the advertiser place a code on their Web site transaction confirmation or “Thank You” page, which provides the information the ad serving system needs to match conversions over a 30 day period to users who clicked on or saw an ad. The ad server produces reports automatically that indicate the total number of conversions the campaign generates.
Cons:
Ironically, this more accurate approach creates its own challenges. For example, there is a debate over which clickthrough or impression should get credit for generating a conversion. Most systems attribute the last clickthrough or impression viewed prior to conversion to be the one responsible for the transaction. But what happens if that isn’t the case? What happens if the reason someone converted was made on the first clickthrough, and the last one was simply a reminder? Or the purchase decision was the result of a chain of events including display ads, a solo e-mail and several searches? This issue comes into play when an online advertiser wants to reinvest in a campaign for rollout. Which media property or list should receive the incremental budget? Marketers also question whether it makes sense to attribute transactions to ads that are viewed, if there is no clickthrough. The issue really comes to the fore when a campaign is ubiquitous, running across many channels and media. Is it really accurate to attribute a conversion to an ad impression viewed on a Web site if the same message is played across other major media in heavy rotation? Maybe. Maybe not. But there are times when a view through most definitely deserves credit for a conversion, especially in those instances when a campaign is limited to the Internet and more importantly, when the campaign is limited in scope and a customer is new. All things being equal, it is reasonable to assume that the user became a customer due to seeing the ad impression.

So what’s an Internet advertiser to do to accurately track conversions?
My advice is to track everything you can, review the data with a critical, analytical eye and come to conclusions based on your own unique situation. You can never have too much data. The final check I use is the tried and true control and exposed test. See if you can find a period when the campaign was “dark” (there was no Internet marketing activity) and compare transactions generated in that time frame with a similar period when the campaign ran. Where possible, try to keep the comparison on an “apples to apples” basis, excluding one-off or other data that obscures the results.
Finally, direct marketers love using the Internet because of its ability to track results better than any other medium so don’t worry about being 100 percent right. Your conversion tracking will get better over time.

For more than 13 years, Jay Aber, president of The Aber Group, has been advising Canadian marketing executives on how to use the Internet to create successful direct marketing campaigns.

The Aber Group’s team of Internet direct marketing strategists, online media experts, and paid search engine marketing pros develop, execute and optimize online direct response campaigns for a broad range of blue chip clients. For more information, contact jay@abergroup.com or 416-322-2909 x222.