Improve Campaign Targeting Through User Segmentation

Segmenting users to better target advertising campaigns is a basic task that can and should be addressed at the initial stages of implementing marketing analytics.

Client’s goal:

Increase the ROI of advertising campaigns that aim to reactivate users who have already visited the site. In theory, these users should not be expensive to market to because they’ve already shown interest.

Client’s challenge:

Improve the targeting of advertising campaigns for abandoned shopping carts and lapsed customers. These are people who have performed a certain set of actions on the site but haven’t bought anything. At the same time, our client believes they could buy something.

Our hypothesis:

If we select a segment of website visitors who have added an item to the cart in the past X days and show them advertising, they’re more likely to buy something than are visitors from other segments. Accordingly, by advertising to this segment we can increase revenue while maintaining the same cost revenue ratio (CRR).

Technical task and solution

User segmentation is always required to test hypotheses. We can assume that selected user segments will behave one way or another after seeing an ad, but there’s no 100% guarantee that will be the case.

Our client defined the conditions according to which segments were formed. After forming segments according to those conditions, we transformed our hypothesis into a technical task for our analyst. This task was divided into stages.

Stage 1: Form two user segments. The first segment consisted of users who added an item to the cart in the last X days but didn’t buy anything. The second segment consisted of users who carried out some set of actions on the site that were defined by the client.

Google Analytics samples customer data in response to heavy traffic. Therefore, to build the segments, we used raw data from our client’s site transmitted to Google BigQuery using OWOX BI Pipeline. BigQuery is a cloud storage service that allows you not only to collect information but to process it conveniently using SQL.

OWOX BI collects user behavior data on a website in parallel with Google Analytics and transmits it to BigQuery without sampling. This greatly enhances business opportunities. For example, by collecting data with OWOX BI, you can combine parameters in a single report that exists in different scopes in Google Analytics. You can use OWOX BI to add information to BigQuery from any system, such as your CRM.

Our analyst wrote SQL queries to form the segments using raw website data. These segments contain Google Client ID and Yandex Client ID data, which can be used in the Google Ads and Yandex.Direct advertising services.

Stage 2: Automatically transfer the segments to Google Ads and Yandex.Direct so advertising specialists can use them in setting up advertising campaigns. OWOX BI has an automatic data pipeline that takes segments from Google BigQuery, calculates data for them daily, and uploads that data to Google Analytics. Segments from the web analytics system can be natively exported to Google Ads. A second OWOX BI pipeline loads these segments into Yandex.Audience.

Stage 3: Our client’s specialists set up rules in advertising services that raised the rates for these segments.

Solution characteristics

This solution has several important characteristics. For example, to load a segment to Yandex.Audience, it must contain a minimum of 1,000 lines in order to use a unique Yandex Client ID.

Stage 1: OWOX BI doesn’t collect Yandex Client IDs by default. Yandex.Metrica must be installed on a site to generate Client IDs, and for them to enter Google BigQuery, you need to configure their transmission through OWOX BI. Our analyst wrote SQL queries to form the segments using raw website data. These segments contain Google Client ID and Yandex Client ID data, which can be used in the Google Ads and Yandex.Direct advertising services.

Stage 2: Automatically transfer the segments to Google Ads and Yandex.Direct so advertising specialists can use them in setting up advertising campaigns. OWOX BI has an automatic data pipeline that takes segments from Google BigQuery, calculates data for them daily, and uploads that data to Google Analytics. Segments from the web analytics system can be natively exported to Google Ads. A second OWOX BI pipeline loads these segments into Yandex.Audience.

Stage 3. Our client’s specialists set up rules in advertising services that raised the rates for these segments.

Our client reconfigured the audience in their campaigns, which had already worked for retargeting, sifted out users who didn’t fall into the right segments, and focused the budget on users in their chosen segments. As a result, with the same campaigns and the same investment, ROI increased by 100% to 150%. There were many campaigns, so the indicator is averaged. In fact, the costs of this decision were minimal, and the efficiency was quite tangible. These campaigns are still working today. We haven’t received any feedback from our client that this increase was only temporary.