Limitations of Conversion Tracking

Don’t live and die by your conversion tracking numbers. No matter how carefully you set up and tested your tracking code, you cannot build a completely watertight system. There are two things that can skew your numbers: cookies and inaccurate click attribution.

Cookies – How Google remembers your interaction with a website (and what it forgets)

Don’t worry, ‘cookies’ have got nothing to do with those edible treats that we bake and eat. Instead, we’re discussing web cookies. Cookies are small bits of text that Google places on your computer as you surf the Internet.

Conversion tracking uses these cookies to track your interaction with the website. Google then serves this information to you in your AdWords dashboard.

Here is one of the most common uses of cookies in advertising. Two weeks ago, I was on Asos looking for a new pair of white trainers. Until today, the advertisements I see on the Internet point me towards suggested pairs of shoes at and Aldo. It’s quite creepy, and helpful at the same time. So, what cookies were placed on my computer?

  • There is a cookie trail following my shoe shopping spree—what shoes I looked at and where.
  • There are cookies to record when and where I saw the shoe adverts, and whether I clicked on it. Google then decides whether I was interested in those ads, and if it should deliver more or less of these shoe ads.
  • If I click on an ad and bought a new pair of kicks, there’s a cookie recording my sale (or a conversion for the retailer).

Cookies don’t store any private information; it’s just a little unique flag that says, “you’ve been here before!”

While this technology is very cool, there are some technical limitations. Here are some instances when the cookies fail:

  • You’ll track conversions within a conversion window. For example, if you set it at 10 days, it won’t record users who return on the 12th day to buy something.
  • If someone clicks on your ad on one computer, then completes the conversion on a different computer (or a phone), Google won’t be able to connect the two events. There’s a stat which says, 90% of people start an activity on one device and finish it on another!
  • If someone clicks on your ad on Google Chrome, then completes the conversion on Mozilla Firefox, Google doesn’t know either!
  • If your customer deletes their cookies or configures their browser to block cookies, you can’t track them either!

All of these situations would falsely decrease your lead generation and sales. Make sure to compare the number of conversions with your own business records to check whether Google is undercounting.

Quick Tip

How to minimise lost conversions as far as possible

Google, being Google, has made tremendous progress in tackling this technical limitation. There’s a new column on your campaign report called “All conversions”, which calculates cross-device and cross-browser conversions.

You can take the following steps to increase the coverage of cross-device conversions:

  • Add the conversion tracking code to all your sites—mobile, desktop and tablet.
  • Set an appropriate conversion window length, of at least 30 days. This will help you identify more multi-device users and long-tail keywords that they use.
  • Add a Google+ sign-in to your site. This increases the Google sign in rates, and the accuracy of your results.
  • Allow AdWords tags to register all transactions, not just those from Google. In other words, remove any container tags.
  • Use cross-account conversion tracking, if you have more than one AdWords account.

However, you’ll need to have enough conversions in your account before Google can run these calculations. Also, such data can only be collected if the user had signed into Google services on their various devices. If I were signed in onto Google on my phone and laptop, Google can track me as one user. If I used my boyfriend’s phone (where I’m not signed in), my cookie trail breaks off again. Google can only use mathematical models to estimate the actual number of cross-device conversions. Because of this, Google needs lots of traffic data, which a lot of small-businesses do not have.

A Display of Google AdWords cross device conversions

Inaccurate click attribution—When your customer clicks on lots of ads and only makes one purchase.

It is hard to track the behaviour of shoppers who hop around and compare sites—basically, savvy shoppers of the 21st century.

This is because one conversion can only be attributed to one ad.

If searchers click multiple ads before converting, only the last ad is marked as a successful conversion!

Let’s study an example of when this happens. Meet Rob. He’s shopping around for a new camera. Rob was in class when he thought, “Hmm, maybe it’s time to get myself a new camera.” He pulls out his phone and runs a quick Google search for “DSLR camera”. Then, he clicks on your ad and looks around. Since it’s a big purchase, Rob wants to compare more sites and mull over it for a little while longer.

A week later, Rob’s read a couple of reviews and knows exactly which model he wants to buy. Now, he googles the phrase “DSLR Camera Nikon D5300”. He comes across another one of your ads. This time, he clicks “buy”. A conversion is recorded.

How does this behaviour affect your conversion tracking stats?

  • From your conversion data, the keyword “DSLR camera” looks like a lousy keyword! Sure, lots of people click on that ad (which costs you money), but it does not convert a single visitor.
  • In contrast, the long-tail keyword “DSLR Camera Nikon D53300” has fewer clicks and a much higher conversion rate.
  • You may decide to pause ads for the first keyword, “DSLR camera”. Alas, your entire campaign’s performance deteriorates slightly. You’ve lost customers who are at the early stage of their shopping journey. Even though the first visit did not generate the sale, it may have made customers aware of your shop.