AnalyticsWhen Clicks Don’t Get Credit for the Sale

When Clicks Don't Get Credit for the Sale

Analytics & ROI expert Eric Enge explains the common problems associated with tracking pay per click campaigns through to the final conversion points. Get tips on how to investigate conversion tracking problems on your Web site and within your online marketing campaigns, and find methods to get more accurate data.

Properly crediting your PPC campaigns with all the sales that they are responsible for can be quite challenging. It is quite common to have sales being driven by PPC keywords that don’t get properly credited to those keywords, and may not get credited to your PPC campaigns at all.

This can cause situations where you decide to lower the bids on, or turn off, campaigns or keywords because they look unprofitable to you. Yet once you are done you end up scratching your head because your total margin drops. Let’s talk about why this may be happening, and what you can do about it.

Problem Scenarios

There are a couple of major ways that this can occur. These are:

1. A user comes to your site through a paid click, and then comes back to your site later through a bookmark, or via organic search, and buys the product. Conventional bid tracking systems will not credit that sale to your PPC campaign.

To be fair, you might think, some of the credit should go to the bookmark, or your organic campaign. But the problem with that is that if you turn off the keyword responsible for the initial click, you would never get the bookmark click, or you might not get the organic search that followed it. So you would lose the sale, and your margin on your overall margin generated by your Web site would drop. Not the desired result!

2. A user comes to your site through a paid keyword, gets to a page where they would like to order a product, and then goes to lunch (or gets a phone call). They come back from lunch (or their phone call ends) and they order the product. The mechanism for this is a bit more complex, and relates to the methods used by analytics packages to track what’s going on with your site.

Analytics Software and Cookies

The techniques for tracking user behavior on your Web site are inherently tricky. Analytics packages track user behavior through cookies. The user’s machine receives a cookie from the analytics software once they get to your site. The cookie contains information on when they arrived at the site, how many times they have been there before, and a user tracking ID.

There is nothing in this mechanism to let the analytics software know when the user leaves your site, and then comes back again later, perhaps by directly typing in a URL address on your site. Or maybe, as we outlined above, the user simply went to lunch. The two scenarios are impossible to tell apart.

So what has become the industry standard is to use the cookie to track the time of the last visit to your site. If the next page request from that user comes within 30 minutes or less, it is considered part of the same session. If the next page request comes in more than 30 minutes, it’s considered a new session. This method provides a really good estimate for purposes of counting the number of users who visit your site. However, it’s not perfect.

Here are some examples of errors that occur in this process:

  • In Problem Scenario #1, if a user comes to your site through a paid click, and then returns via a bookmark, and the time between the paid click visit and the bookmark visit is less than 30 minutes, this will still be considered a single visit, even though it’s really two different visits. However, this will cause your PPC campaign to be credited with any sales that occurs on the second visit, and should not be a problem.
  • In Problem Scenario #1, if the return visit via the bookmark occurs more than 30 minutes later, it will be counted as 2 different visits. But this is how our problem with PPC tracking develops. The paid keyword, and its campaign, don’t get credited with any sales that occur in the bookmark visit.
  • In Problem Scenario #2, if the user arrives at your site from a paid keyword, and is then idle for more than 30 minutes before placing an order, your PPC campaign will not get credit for the sale.

Note that if your user blocks cookies, or your cookie gets deleted in the meantime, all bets are off, and there is nothing you can do to track these scenarios. Fortunately, if you have setup the analytics software to use first-party cookies, cookie deletions happen a small percentage of the time – but do make sure you are using first-party cookies, and not third-party cookies.

What to do?

Fortunately, there are some analytics packages that allow you to still credit your PPC keyword and campaigns for these scenarios. Because the analytics software has an ID for each cookie it has set, it can track the referring information from each visit and keep that in an internal database. So when the user makes that second visit (be it Problem Scenario #1, or Problem Scenario #2), it knows that there was an earlier visit involving a paid click. Some packages allow you to configure them so that they will still attribute the sale to the paid click.

Here are a few analytics tools that offer this capability:

Some of these programs offer more capability than the others, and some require a bit more configuration than the others, but they all offer ways to address this issue. You just need to know to look for this issue and make sure you address it.

The Bottom Line

There are more complex scenarios that we have not dealt with here. For example, if someone comes to your site via paid clicks multiple times, which keywords get credited? To date, no true solution for that problem exists, so we simply need to be aware that it exists and then forget about it. The above problem can help you improve your measurement of how your PPC campaigns are doing, and help you generate more sales. Just remember: better measurement gives you a competitive edge.

Resources

The 2023 B2B Superpowers Index
whitepaper | Analytics

The 2023 B2B Superpowers Index

8m
Data Analytics in Marketing
whitepaper | Analytics

Data Analytics in Marketing

10m
The Third-Party Data Deprecation Playbook
whitepaper | Digital Marketing

The Third-Party Data Deprecation Playbook

1y
Utilizing Email To Stop Fraud-eCommerce Client Fraud Case Study
whitepaper | Digital Marketing

Utilizing Email To Stop Fraud-eCommerce Client Fraud Case Study

1y