Questions and issues over tracking and reporting online ads have persisted among agencies, media vendors, ad servers, Webmasters, and clients since I started my career. Terms and procedures around tracking have not been standardized, resulting in lots of confusion.
Last time, I listed five common tracking and reporting issues and offered sensible approaches to deal with the first two. In this column, let’s examine ways to address the latter three tracking issues.
The five most common tracking and reporting issues are:
- Why online reports aren’t 100 percent accurate.
- Why view-through conversions are not conversions.
- How double counting occurs in different systems.
- Why tracking from different systems don’t always match.
- How to compare first-click conversions versus last-click conversions.
How Double Counting Occurs
Double counting between different systems is a frequent problem when tracking actions known as conversions, such as when a Web site visitor makes a purchase, downloads a white paper, or takes some other action.
Let’s say I run a banner ad program and use persistent cookies via my ad server to track anyone who clicks on a banner ad, comes back within 90 days, and converts. Here’s what happens: My ad server counts the click as a conversion from the banner (the first visit and how one met the site); my second tracking system, let’s say Google Analytics, attributes the conversion to the source of that particular user session, such as a keyword search. Now the ad server counts the conversion from my banner ad campaign and Google Analytics counts it from a natural search on Google.
When you add all your conversions from your different tracking systems, you’ll end up double-counting conversions. That’s because the conversion counted from the banner ad campaign is actually the same conversion counted by Google Analytics.
There are a few ways to handle this. For starters, never use your tracking systems to report on total monthly conversions. Tracking systems are for giving you directional information on where to spend your money or where your quality traffic is coming from, not for accurately totaling how many leads or sales you made. Actual conversion tracking totals should come from your lead or sales database. Get the figure right from the most accurate source.
Second, utilize an integrated tracking solution that can accommodate all the sources of your clicks and duplications based on your settings. If you have the means or budget, you can integrate your totals with most major ad servers and high-end analytic applications, like Omniture. The key is to decide on whether you want to count conversion against the first source or last point of entry. As a media person, I prefer the first source with persistent cookies, because many people discover you from a paid ad and come back later to convert after a natural search or direct hit.
Myth: Tracking Systems Match 100 Percent
Every once in a while, a client will look at results from a tracking system they’re running in parallel to our tracking systems and say, “Hey, our tracking results are different. Tracking from different systems should match.” This statement is just not true; different tracking system hardly ever match.
Online real- or near-time tracking, which is directional at best, can accurately help you make fact-based decisions about where to spend your marketing dollars. Get this fact on the table with your client and tell her there is a strong probability that your numbers may not match the numbers of her site’s analytical application or customer resource management database. Let’s say you look at your ad server or AdWords tracking that has persistent cookies, and it counts conversions from people who clicked on a banner at any time in the past. Meanwhile, your client looks at Google Analytics and Omniture and doesn’t see the same conversions numbers. That’s because your client is looking at conversions from the last click and you are looking at conversions from the first click.
First-Click vs. Last-Click Conversions
For anyone who manages online media, the use of persistent or long-term cookies is critical. Long-term cookies let us track the real results from our campaigns because people don’t always convert on the first click. If we didn’t count the actions from people who discovered our clients’ sites and return later to perform an actual transaction, the results of our campaigns would be grossly understated.
Explain to your client that to accurately identify the source of high-quality traffic, you use 90-day cookies and that you count all conversions from people who have been introduced to your site via paid media as conversions. This makes sense because people don’t always convert on the first click; sometimes they need time to think or complete their research. Explain that your tracking is about identifying the right combination of sites and ads that deliver an audience that has the highest propensity to convert in the near- to midterm. And explain that long-term cookies are essential in doing this. Then tell your client that if her tracking system attributes conversion only to the source of the user session in which the person converted, your numbers won’t match hers. And that’s OK!
Please reach out to me with a comment or any questions or insights.
Meet Harry at Search Engine Strategies New York March 23-27 at the Hilton New York. The only major search marketing conference and expo on the East Coast, SES New York will be packed with more than 70 sessions, including a ClickZ track, plus more than 150 exhibitors, networking events, parties, and training days.