Blog for internet marketing

Cases

Case Study: 59,35% “Not Set” Traffic Drop in Google Analytics

3
0
0
0

Specialists who run contextual advertising in Google AdWords and analyze its results in Google Analytics, often face a situation where during handling the main task, minor issues are also occasionally tackled. In this case study we will shed light on how we managed to solve the problem with a small number of conversions and their high cost, as well as the issue with the prevalence of «not set» traffic in Google Analytics reports. 

Service: contextual advertising in Google AdWords.
Topic: software development.
Region: worldwide.
Campaign duration: 6 months.
Budget for 6 months: $19,460 (including agency’s fee).

Goals and Project Features

9 months ago, before all the Google AdWords advertising campaigns were handed to Netpeak agency, the client was running them independently, occasionally turning for our specialists’ advice. Based on the data received from CRM-system, the major problem, in client’s opinion, was the low efficiency of Google AdWords advertising campaigns. Therefore, the main goals for advertising campaigns were the following:

  1. Branding (not to be confused with brand campaign) of all the client’s products with a simultaneous reduction of lead acquisition costs and preservation of advertising share amid the total number of transactions.
  2. Registration for the company’s webinars of those users, who had previously interacted with the company, as well as attraction of new users to viewing these webinars.

Statistics on the effectiveness of advertising campaigns: two months before the client contacted the Netpeak agency: Two months before the client contacted the Netpeak agency This screenshot clearly shows that the main traffic and conversions were registered by Google Analytics as «not set», at the same time the data on paid traffic and relative number of sessions, conversions and other behavioral characteristics didn’t appear in the report. The analysis of «not set» traffic channel revealed that this traffic stemmed from AdWords contextual advertising: The analysis of «not set» traffic channel revealed that this traffic stemmed from AdWords contextual advertising The question still stood as to why Google Analytics interpreted the traffic as «not set». We also had to solve the problem of data not appearing on client’s Google AdWords campaigns.

“Not Set” Case Investigation

1. The first thing we managed to discover after getting a full access to the Google Analytics and Google AdWords accounts – was the unsystematic UTM-tagging to all links and banners, both in advertising campaigns and on all pages, which was carried out by the client’s specialists. At that, the Google AdWords traffic sources were the following:

  • email distribution;
  • referral links;
  • banner links to the website on partners pages;
  • banner links and off-page references on the website;

2. Not all the Google Analytics reporting views, previously used by the client, were linked to the appropriate Google AdWords account, so they didn’t contain the data on the number of clicks and cost. 3. The client systematically conducted the training webinars regarding his products using different platforms. However, not all of the platforms had the resources to pass necessary data to Google Analytics. That is why the Google Analytics reports on these campaigns didn’t contain the conversion data for the respective periods: Google Analytics reports on these campaigns didn’t contain the conversion data for the respective periods In addition, when analyzing the landing pages with Google Tag Assistant, we found that the Google Analytics and Google AdWords remarketing codes had not been properly, and this caused difficulties with the collection and flow of data regarding clicks on ads and filling in the remarketing list: We found that the Google Analytics and Google AdWords remarketing codes had not been properly, and this caused difficulties with the collection and flow of data regarding clicks on ads and filling in the remarketing list

Work Stages

We started tackling problems together with the client’s specialists (there is no getting by without them). 1. We updated the website codes of utilized metrics and retargeting using Google Tag Manager, which was subsequently configured to track all the necessary goals and import them to Google Analytics. 2. In a separately created Google Analytics reporting view we set up the filters which excluded traffic coming from Netpeak, client’s office and referral spam sites’ IP-addresses. 3. We agreed on the system of UTM-tagging applied only when it was necessary to track conversions from external resources with their subsequent transmission to CRM. 4. We gave our client a task to select the webinar platforms permitting to monitor and import the necessary data to Google Analytics.

To find out more about the other work stages on this project within the framework of Netpeak [PPC 2.0] innovative service, click here.

Results

To start with, here is the statistics on advertising campaigns: two months before the client gave us access to his accounts: Two months before the client gave us access to his accounts Almost all the sessions were registered in Google Analytics as «not set» campaign, even though Google AdWords did register the clicks on ads. Moreover, since the data from external resources (including the data on registration for the webinars) was not imported to Google Analytics , it was impossible to evaluate the effectiveness of advertising campaigns. And here is what we saw after four months: And here is what we saw after four months Just look on the bounce rate. High enough, isn’t it? We were also confused because of it for quite a long time. In fact, according to various sources, this field’s «normal» bounce rate for the landing pages with a call to action ranges between 70-90%. Having talked with Google Support, we additionally identified the "weak" spots of the website, however, the company might need much more than one month of technical works to implement the appropriate changes. On top of that, all the necessary information on traffic sources and their effectiveness did appear in the reports.

Besides, in the process of solving the problem of «not set» traffic decrease, it turned out that the company gave one of its pages to a partner in order to attract users by using PPC advertising from a Google AdWords separate account.

Having set up the additional filter to exclude the "partner’s" traffic, we managed to significantly reduce the share of «not set» traffic in the reports on the effectiveness of advertising campaigns. Initially its share amounted to 99.99% of total traffic, but after applying the filter and taking the described actions, the share of «not set» traffic came to 40.64%, that’s a drop of 59.35%. After optimizing the campaigns, we also achieved a significant reduction of conversion cost. For comparison, we would like to show statistics for the two periods that we looked at earlier: Aug. 23, 2014- Oct.23, 2014 Statistics for the two periods that we looked at earlier Feb.02, 2015-Apr.04, 2015 Statistics for the two periods that we looked at earlier

Before we started working on the project, the average conversion cost according to Analytics reports was $22.52, but after optimizing the advertising campaigns it dropped to $16.37.The average cost per click decreased substantially - from $2.28 to $0.26.

Here is a comparative table on the effectiveness of advertising campaigns for the specified periods: A comparative table on the effectiveness of advertising campaigns for the specified periods

Let's summarize

As a result of carried out works, we:

  • reduced the average cost of conversion by 27.3%;
  • reduced the average cost per click by 89%;
  • received 58% increase in the total number of achieved goals;
  • reduced the «not set» traffic share in Google Analytics reports by 59.35%;

Today, as we continue our work on the project and not least because of the client’s commitment to the development of the website, we are reaching higher heights while pursuing new goals. And that’s what we are going to tell you about in one of our upcoming cases.

Comments (0)

To leave a comment, you have to log in.

Subscribe

to the most useful newsletter on internet marketing

Most

discussed popular readable