Measuring the effects of branded and non-branded traffic
Posted: Tue Dec 17, 2024 7:00 am
In the example below we see that the difference between 'users' and 'new users' increases over time. In this case a good indicator that the branding has a positive effect.
Difference between new users and users is increasing in graph.
The caveat here is that you need to look at the target group per website. Visitors can return to the website for multiple reasons, and not just for good reasons. So pay close attention to this when analyzing! For example, you can think of:
A product has been sold with defects. The website therefore receives many angry customers.
Service has deteriorated, visitors are filing complaints.
Visitors often return seasonally. Think about insurance at the end of the year.
Technical reasons. Due to the increasingly strict privacy settings of browsers, analytical cookies are reset more often. This has an effect on the statistics that measure visitors. Especially websites with many visitors that use Safari, Firefox and Brave suffer from this.
To eliminate the above risks as much as possible, it is wise to divide the data into channels, so that you can better estimate whether these returning visitors have good intentions.
Furthermore, you should always analyze the metric 'returning visitors' in the long term, not per day or week. I would then go for at least a few months or more, so that you gain insight into the bigger picture. This allows you to say with more certainty that your branding is effective. In addition, you exclude incidents that are only visible in short-term analyses. And if possible, always segment at channel level.
Measuring the direct effect of branded traffic
To see the direct effect of branded traffic on your transactions, you can use segments in Google Analytics. This allows you to apply custom rules that filter the reports.
In this case I have chosen branded in the vietnam telegram data sense of 'paid' (keyword with brand name), 'display', 'direct' and 'organic search users' who land with a keyword on the homepage or on a page of a store location. This rule is of course always open to discussion and depends on the type of website.
Segments in Google Analytics.
In the report below I have compared branded and non-branded traffic. Note that I have chosen transactions based on 'last click'. I will explain the effect of attribution later.
As you can see, this website has a lot of branded traffic, about 34%. This also converts a lot better than non-branded traffic. So apparently a lot is invested in branding here, and with success!
Visitors via branded traffic yield more on average. It is striking that non-branded traffic is on the rise in April and will also yield more on average. Is this a seasonal trend, or a change in marketing strategy?
Difference between new users and users is increasing in graph.
The caveat here is that you need to look at the target group per website. Visitors can return to the website for multiple reasons, and not just for good reasons. So pay close attention to this when analyzing! For example, you can think of:
A product has been sold with defects. The website therefore receives many angry customers.
Service has deteriorated, visitors are filing complaints.
Visitors often return seasonally. Think about insurance at the end of the year.
Technical reasons. Due to the increasingly strict privacy settings of browsers, analytical cookies are reset more often. This has an effect on the statistics that measure visitors. Especially websites with many visitors that use Safari, Firefox and Brave suffer from this.
To eliminate the above risks as much as possible, it is wise to divide the data into channels, so that you can better estimate whether these returning visitors have good intentions.
Furthermore, you should always analyze the metric 'returning visitors' in the long term, not per day or week. I would then go for at least a few months or more, so that you gain insight into the bigger picture. This allows you to say with more certainty that your branding is effective. In addition, you exclude incidents that are only visible in short-term analyses. And if possible, always segment at channel level.
Measuring the direct effect of branded traffic
To see the direct effect of branded traffic on your transactions, you can use segments in Google Analytics. This allows you to apply custom rules that filter the reports.
In this case I have chosen branded in the vietnam telegram data sense of 'paid' (keyword with brand name), 'display', 'direct' and 'organic search users' who land with a keyword on the homepage or on a page of a store location. This rule is of course always open to discussion and depends on the type of website.
Segments in Google Analytics.
In the report below I have compared branded and non-branded traffic. Note that I have chosen transactions based on 'last click'. I will explain the effect of attribution later.
As you can see, this website has a lot of branded traffic, about 34%. This also converts a lot better than non-branded traffic. So apparently a lot is invested in branding here, and with success!
Visitors via branded traffic yield more on average. It is striking that non-branded traffic is on the rise in April and will also yield more on average. Is this a seasonal trend, or a change in marketing strategy?