Buyer Persona profiles identify groups of similar customers to personalize communication-based on the propensity to buy of each customer, allowing them to predict and personalize the recommendation.
Buyer-persona analysis name: Define a unique name to identify the new buyer persona profile analysis created.
Previous buyer-persona analysis: Activate this option to use a previously generated dataset. If you activate these options you only will be able to modify the number of clusters to generate.
Entities, attributes, and calculations: Define attributes for buyer-persona analysis by Entity (Customers, Products, Shops, Tickets, Line-Orders), Attribute, and Calculation types.
Transformation options: Configure how you want us to treat the data.
Don't do any transformation if you only use one attribute or all calculations are Boolean.
If you mix a few attributes with similar values we advise normalizing.
In the case that you use different attributes and different types of calculations, the best option is to standardize.
Date filter: You can activate the filter of data to be analyzed given a time interval. By default this option is deactivated, you can activate it by clicking on the next checkbox. The data that is incorporated once launched in the analysis will not be processed.
Entity filters: You can activate the filter for each of the data entities to be analyzed. Using this option you are limiting the customers, products, stores, tickets, or ticket lines to analyze, making the exercise much more specific for the particular case. (Select user, product, shop, or ticket filter)
Number of clusters (K): Define the number of clusters you want to generate. The higher the number, the greater the specificity of the results. We recommend a value between 5 and 15 for an initial analysis in the hierarchy of product families. For CRM enrichment we recommend generating profiles of 3 to 8 attributes with a K between 29 and 151 groups. We advise using odd numbers.
Automatic recalculation: Activate analysis recalculation with a certain frequency to keep the clusters up-to-date with your environment’s new data.
This section of the platform allows for a more visual understanding of the information based on all important aspects of the buyer persona profiles to better analyse the main characteristics of each cluster.
The analysis is based on the classification of client clusters with similar propensities according to the selected variables. In the graph, you can see the level of loyalty of each client, with X being the number of tickets, the total cost per client, and the circle size being the volume of clients in that group. The table shows you the statistical description of each cluster.
You can also find the origin values of the graphic below in a table that shows all the data used for it.
In the following matrix, the rows represent a possible value of the parsed attribute. Each column represents a cluster of clients. The intensity of each cell refers to the propensity towards that characteristic, the more intense the blue, the higher the propensity and the intensity decreases as the propensity decreases. The gray color represents the null propensity.
Gender propensity: represents the propensity towards a variable of gender in each customer cluster.
Axis propensity: represents the propensity towards a variable of different group products in each customer cluster.
Exclusivity propensity: represents the propensity towards a variable of different group products in each customer cluster.
Compare the characteristics that define the buyer-persona profile of the different clusters to identify opportunities for cross-selling, upselling or behavioral changes. Click on the labels of the buyer-persona clusters to add or remove them from the comparison.
The executive summary provides an overview of the results. It provides the overall results, lifecycle, recurrence, frequency, monetary value and acceptance of the GDPR, obtained from each of the Buyer Persona profiles analysed.
Global statistics on the number of customers, total spending, number of tickets, and average ticket for each of the buyer-person profiles.
This information includes all customer data history. To see the evolution of consumption you can enter the details of each cluster or use the comparator. Remember to create a segment for each buyer-persona cluster you want to compare.
The average of the RFM values for each of the buyer-persona clusters is shown below. The recency refers to the number of days since the last purchase, the frequency represents the number of tickets per customer and the monetary represents the total expense.
These graphics show the average values of Lifetime value (LTV) and the days between purchases (TBP).
The volume of customers who have agreed to receive commercial communications for each buyer-person cluster. This information must be ingested in the subscribed field of the client. Remember that you can only send commercial communications to those customers and registries that have agreed to receive commercial communications.
Get customer buyer-persona info: This is the API called to obtain information in real-time about the purchase propensity of a buyer-persona analysis for a specific customer in real-time. With this information, you can customize the experience based on the characteristics identified in the clustering models.
To find more information about the use of APIs: API