Quick Start
Last updated
English
Last updated
Welcome to Karsa Labelizer! This guide will show you how to quickly create your first test product clusterization. The goal is to familiarize you with the basic workflow and show you how easy it is to get started. For a detailed understanding of individual steps and settings, we recommend studying the relevant sections of this documentation afterward.
Follow these steps and within a few minutes, you'll have your first custom product segmentation proposal ready:
In the left navigation menu, find the section or link leading to the Cluster Designer. This is your main tool for creating new clusterizations.
In Cluster Designer, click on the open folder icon. Then select the appropriate configuration and confirm your selection by clicking the "Open" button.
Basic settings:
Clusterization name: Enter a descriptive name, e.g., Test_ROAS_ConvValue (the date will be added automatically when the clusterization is created).
Description: Briefly describe the purpose of this test, e.g., First test clusterization based on ROAS and Conversion Value. This description will be displayed in the main part of the screen.
For your first quick test, you can leave many values at their default settings, but check/adjust the following:
Minimum and maximum number of clusters/campaigns: These two settings define the range in which the Karsa Labelizer algorithm will search for the optimal number of clusters for your specific data structure and product diversity. The algorithm always tries to find the number of clusters that best suits your data and maximizes the consistency of the resulting segments.
Minimum number of conversions: For testing purposes, you can start with a lower value (e.g., 15-25) to see more clusters. For production deployment, a higher value is recommended (e.g., 30+).
Parameters for clusterization:
Keep the default (or choose) combination of ROAS and Conv. value (Conversion value). This is a great starting point.
Custom Label:
Select an available Custom Label number (e.g., custom_label_0, if it's available).
Enter a short Prefix (e.g., ktest). The resulting labels will look like ktest_1, ktest_2, etc.
Product movement and new products: For your first test, you can leave these settings at their default values, where product movements are disabled.
After completing the configuration, find and click on the arrow icon to start the clusterization process.
The process may take several minutes depending on the number of your products and data complexity (usually 2-15 minutes).
You can monitor the operation status in Tools > Operation Manager.
Once the clusterization is complete, it appears as a new category in the left navigation menu, containing the individual generated clusters.
Browse through the reports:
Clusterizations Overview/Tab CLUSTERIZATION: Here you'll see your new test clusterization, its status (Test), and basic metrics such as Consistency.
Cluster Detail/Tab CLUSTER: After selecting your new clusterization and then the tab/report for cluster details, you'll see metrics for each individual created cluster.
Products in Cluster/Tab Products Explore which specific products were assigned to individual clusters.
Congratulations! You've just created your first test clusterization in Karsa Labelizer. Now you have a basic understanding of the process.
This quick start was just a taste. For actual deployment and achieving the best results:
Experiment with different parameters and settings in test mode.
Learn how to deploy clusterization to production mode and integrate it with your Google Ads campaigns.
More information:
More information:
Study the individual sections of this documentation in more detail, especially the parts about and .
See:
