Share:

One of the biggest fintechs in Latin America.
To maintain a high level of consumer satisfaction and offer increasingly adherent financial products and services. As well as understand this new profile.
The Fintech now had in its hands a true map for defining actions and strategies.

Automated CRM Segmentation for a Leading Fintech

A Latam fintech unicorn grew its user base by around 400% in a single year due to a clear value proposition and fluid user/customer experience.


A fintech reference among digital banks grew by around 400% in 1 year. It acomplished this with a clear value proposition, targeted investments, and the development of new technologies, all focused on customer experience.

The accelerated growth also coused doubts: who were these new customers? How do we serve this new user-base? How do we change our products and services? What innovation opportunities can help us retain these new clients?

In a commoditized market such as the banking sector, user experience and customer relations can make all the difference. Given this context, the commitment was:

  1. Understand the new customers and their unmet needs;
  2. Map out where to invest efforts;
  3. Expand product and service portfolio.

The growth boom experienced by fintechs generated new challenges, mainly regarding the expansion of services. 

In addition to this, emergency aid procedures in the midst of the pandemic added 33 million new entrants to the financial system. An excellent opportunity to embrace expansion. 

The challenge: to maintain a high level of consumer satisfaction and offer increasingly adherent financial products and services. As well as  understand this new profile. 

As the initial challenge was too broad, it was divided into four parts:

  1. Segmentation of the population within banking services;
  2. Discovery of unmet needs;
  3. Creation of innovation opportunities for the high-income market;
  4. Identification of segments in CRM (identified on the basis of who belonged to which segment).

One of the biggest fintechs in Latin America, with an swarm of followers – something previously unthinkable for the banking sector – could not classify these new clients by only income and age group. Something was missing. And we were determined to find it.

It was necessary to define what sets them apart. Put the user at the center of the business strategy and delve delve into their user-base. To know these new customers as well as they knew their loyal ones.

For this, the project was structured in four different interdisciplinary segments:

  • Customer & Market: Benchmarks and market research.
  • Quantitative research: Customer research, data crossing, and final segmentation.
  • Innovation opportunities: Using “Jobs to be done” to list actions in a blue ocean strategy.
  • Data Science: Data quality, exploratory data analysis, hypothesis generation, feature engineering, dashboard development, CRM base segmentation, segmentation comparison, customer base scoring, base classification automation, and segmentation documentation.

The essence of the value proposition was to understand the new consumers. After all, segmenting is nothing more than analysis. 

Afterwardsm we would join Service Design and Data Science, to delve deeply into the customer profile within the database and identify behavioral nuances. In addition to crossing different froms of data, including demographic, behavioral, and qualitative data.

Our innovation team led the Design Thinking process. There were around 300 hours of exploratory and in-depth interviews with customers to find out how digital bank users think and behave.

Transforming data into value

In this case, all areas involved were equally relevant in co-creation but let’s focus on Data Science.

The Data Science project starts with data access. Initially complex for information security reasons, we needed to design strategies that allowed employees to acceess data securely.

We soon realized that this company’s database was enormous. The company itself was, in fact, mature in data and presented us with a base with thousands of variables. 

The initial feature of data science projects is the exploratory phase. But with thousands of variables, we needed to know exactly what we wanted to extract out of their data.

The generation of hypotheses took place in workshops and exploratory interviews, which were the basis for conducting in-depth research.

→ Who would be our heavy users within finance?

→ What characteristics could describe their profile?

Machine learning models were used to classify information within the feature engineering process to extract the following:

  • A thousand-variable data set;
  • Creation of features with profile views, product usage, and financial behavior;
  • Cloud parallel processing strategy to analyze effectively

With these preliminary insights, we created our first deliverables:

  • Creation of interactive customer analysis dashboard;
  • Filters for different customer profiles;
  • Pipeline made available for implementation.

Based on this analysis, it was time to create segmentations. We started the iterative segmentation process by looking at the 127 variables we created.

  1. During the process of generating new hypotheses, 14 iterations were created. We compared these iterations with existing segmentations to pinpoint weaknesses and vulnerabilities.

Meanwhile, the MJV innovation team was conducting immersion and analysis with the client and market. Based on qualitative criteria, the innovation survey sought to segment the ‘banked’ population to identify the key aspects that shape people’s financial decisions.

Subsequently, the Data Science team conducted quantitative research at the national level to help segmentation. The research started using drivers generated in the previous phase. The questionnaire sought to deepen our understanding of financial behaviors and attitudes.

  • Association of questions and drivers;
  • Categorization of macro-drivers and micro-drivers using factor analysis;
  • Cluster analysis;
  •  Segment interpretation.

The Innovation team then analyzed the clustering generated by the Data Science team. This was done to see if it made sense from a qualitative point of view.

The Fintech’s customers were divided based on eight drivers. Other variables collected were used to describe and interpret segments, such as demographics, product history, and financial behavior, among others.

The variables were used to describe segments and provide more insight as to the kind of products those customers were consuming. 

The result was material rich in information, detail, and insights:

  • Identification through fintech respondents;
  • Availability of clusters with fintech customers within the dashboard;
  • Documentation of the entire segmentation process;
  • Record of all decisions made, references, and code used. Available in an easy-to-navigate web interface.

With the data correctly structured and insights generated, the Fintech now had in its hands a true map for defining actions and strategies.

Insights for decision making

The client was now armed with the qualitative insights of the DT process and the segmentation hypotheses of DS. But we also generated a number of possibilities and discoveries about unmet needs.

What we call scalable innovation opportunities

The final delivery consisted of:

  • Creation of 15 customer segments with different characteristics.
  • Market study and benchmarking research;
  • Customer ranking according to behavioral profiles.
  • Mapping of 40 innovation opportunities within a blue ocean strategy (where there is no direct competition from other banks);
  • Prioritization of 26 jobs to be done in search of a competitive edge in this new moment of expansion of the fintech market.
  • Development of tools and dashboards for complete customer analysis.

The project was a milestone for the fintech, as it indicates what the company’s next steps will be. It was  a long-awaited change of pace fore them: the moment when Change the Bank finally becomes Run the Bank.

It is important to emphasize that the process was carried out with a lot of cooperation from stakeholders, who understood that new contexts require new ways of operating – in addition to active participation when prioritizing actions.

When dealing with such a disruptive company, listening to the customer is essential. 

If you too want to get to know your customers and gain valuable insights into products and services, we can help. Get in touch with one of our consultants!

Back