When it comes to measuring success, one way of segmenting your visitors is by looking at the how a visitor arrived at your site. This might be by clicking through a search result or following a link from some other location. Typically, most people correlate the source of a visitor’s session to a conversion. However, do people convert one of your website goals simply on one visit? Possibly. But far too often, someone visits multiple times, converting after multiple visits. Each of those visits may be from different sources. For example:
I could search for a product and find numerous vendors. I may look at each in turn to try to determine similarities and differences. Days later, I may see a friend sharing something on social media about one of the brands I discovered in my search. As such, I would click-through and visit a second time. If that brand is employing remarketing tactics, I may start seeing display (banner) ads on other sites I visit, enticing me back to their site. After numerous visits, I may decide to buy. When that time comes, I might ultimately visit their website by directly entering the URL into my browser. After all, having been there so many times, I likely know it by now.
Using most web analytics packages, it’s very easy to suggest that a “direct visitor” converted a goal. After all, the source of the actual visit with the conversion was a direct entry. But is that truly the case? Social media, banner ads and search also played a role. Last-touch attribution, as it is called, would suggest the conversion was caused by a direct visit. But shouldn’t your social media or paid strategies get credit, too? In the example I’ve just described, paid search had an effect on my decision to buy. the cost of acquisition should include those paid display ads.
How Do You Determine Multi-Attribution?
There are many methods to understanding channel effectiveness online. Each method – or attribution model – is useful in its own respective way. However, depending on your campaign goals and business objectives, one model may be more useful than another.
This model states that 100% of the conversion value goes to the last-touch channel. It is the most-often used model and certainly is good for providing benchmarks for online sales. It certainly works for short sales cycles.
First-touch gives all the credit to the first channel source. This might be useful if you are trying to build brand awareness or giving credit to affiliate program partners.
Splitting the Difference
One form of linear attribution model suggests each channel’s interaction receives equal credit toward converting the visitor. Other models work similarly, but may heavily weight the first touch, last touch or even both touches. Some position-based models assign 40% of the credit to the first touch, 40% to the last touch and then splits the remaining 20% credit to the channels in the middle.
Multi-Channel Reports In Google Analytics
Determining all the touch points used to involve a fair bit of coding and set up. Now, Google Analytics makes it fairly straight-forward. There’s a few reports to look at to help determine the real story behind how a visitor to your site became a customer.
Under the Conversions reports is a section labelled Multi-Channel Funnels. The Assisted Conversions Report shows you which online channels helped contribute to your conversions.
As you can see, the report shows the channel source, how many times it was part of a visitors’ conversion, the goal value from the conversion source, how many times it was the last touch point before conversion and other associated values.
You can also view the actual paths visitors took to converting. These show how many visitors followed a specific pattern of behavior to converting. This report shows the actual source of visitors in order, including any sources that may have been repeated.
By default, Google Analytics looks back 30 days from a conversion visit to determine these paths. If you know your average sales cycle is longer, you can change that time. At the top of each report are options to change the number of days in the “lookback window.”
This is just the start of attributing various channels to conversions. Online conversions are inherently easier to track. What about off-line conversions? Next week, I’ll dive into tactics to track attribution of offline campaigns.