All posts by : Admin GBA

Despite the fact that in Colombia all workers are obliged to contribute to the social security system, in 2018 35% of employees of private entities and 67% of self-employed workers did not contribute to their pension. This, in part, due to the fact that 44% of Colombians have incomes below the minimum wage and, therefore, the minimum contribution to the system represents a very high percentage of their income. Without sufficient income, many Colombian workers do not have an automatic mechanism to save for their old age.

To improve coverage for this population, in 2015 the Colombian government created the voluntary pension savings program Periodic Economic Benefits (BEPS). When the intervention was designed, in June 2018, BEPS had 1,107,383 linked, of which 352,192 (31.8%) had saved at least once in the program. Moreover, only 19% of those who saved did so consistently: either they had saved in 2017 at least 147,500 Colombian pesos (approx. $115 PPP), or they had made more than six contributions in that year, requirements to access life insurance subsidized by the program in 2018.

From this perspective, we proposed the implementation of Single Customer View (SCV) to allow us to create a complete view of each customer. Integrating all relevant information about their behavior and activity throughout their customer lifecycle.

The Client

From the IDB retirement savings laboratory, the Inter-American Development Bank asked us to develop several processes. Such as, improving the results of linking new savers, increasing the recency of savings, increasing the frequency of savings transactions and improving the average value saved in each transaction. These were the 1.3 million people linked to Colpensiones’ BEPS program (Periodic Economic Benefits).

The program consists of offering people a voluntary savings mechanism for old age, the majority of those linked must be people of low socio-economic levels with an income of less than 1SMMLV (One Minimum Legal Minimum Wage in Force). At the time of the pilot, 68.20% of the portfolio had no savings and 31.80% were considered savers. These are populations with little or no banking coverage, mainly underemployed and informal employees.

The point of focus

The IDB needed to have a comprehensive understanding of its savers, as well as the reasons why other people involved in the program did not save, in order to encourage savings in the most efficient way for both groups.

There are multiple reasons that limit voluntary long-term savings in low-income populations. On the one hand, low experience with the financial sector and lack of knowledge of the program (which in some cases leads to distrust). They may lead many people to prefer to save in financial instruments that are more familiar to them, such as investing in their home, even though the returns on these investments may be lower and more volatile.

On the other hand, psychological biases, most notably inattention, decrease individual willingness to save.

It has been identified that individuals do not have an image of their own old age, especially young populations do not consider it; there are life cycle conditions that facilitate the appearance of this imaginary, so these variables become relevant in the analysis and approach to potential savers.

In order to motivate those linked to the BEPS program who did not save (inactive) to start doing so, and to encourage those who did save (active) to do so more, a strategy of savings incentive campaigns was proposed, executed through different channels, including text messages and high-touch telephone calls, in segments chosen by analysis of massive data (big data) with medium and high propensity to save.

Project Challenges

Low contactability and the impossibility of building multichannel in these groups, a population that, for the most part, changed their mobile number and physical address between two and three times a year.

In addition, we had to take into account a limited budget and the presence of users in more than 1,100 municipalities nationwide. This considerably increased the costs of any face-to-face commercial action that we wanted to propose.

GBA Smart Marketing® Solution:

Single Customer View (SCV).

The task was to increase the number of annual transactions, the average amount saved in each one of them, and to get those who did not save to pick up the habit again. However, we had little knowledge of the customer, which is why all campaigns were segmented in a traditional way and their communication channels were limited to the use of their Contact Center and SMS messages.

We performed a data quality process on the information received from the entity in order to detect outdated and/or inconsistent information. We then updated, complemented and standardized the physical addresses of savers and non-savers, which allowed us to geocode 95% of the database.

Based on the transactional and socio-demographic information and attributes built by GBA Smart Marketing® for the project. Among them, the savings propensity score, income estimator and credit risk score stand out. We were able to segment 100% of the savers and non-savers with more than 51 characteristics for each one.

With the knowledge acquired, we planned and executed differentiated marketing campaigns, through more channels, with messages and frequencies appropriate for each group, maximizing the return on marketing investment (ROMI).

To study the impact of the campaigns, 24,000 program participants were randomly assigned to the pilot, creating a treatment group and a control group, each of which consisted of 6,000 active and 6,000 inactive participants. Since the participants were randomly selected, it was possible to assess the impact of the call campaign on the total number of linkados.

The result:

• • We increased the value saved by 9.47%; the average transaction by 4.11% in the first two months of implementation; the historical number of transactions from 3.31 transactions/year to 5.22 transactions/year; and the contact data update rate by 17.03%.
  • We identified and corrected 17.21% of the database of related parties that presented inconsistencies in the information registered, which affected the results of all marketing actions.
  • We implemented at least 51 new descriptive characteristics that significantly increased the knowledge of the affiliate.
  • We located the areas of concentration of non-members with the highest propensity to save, so that the commercial area of the institution could carry out a more efficient linkage work through its face-to-face brigades with advisory managers.
  • By automating the affiliate journey, we increased the number of campaigns carried out and optimized their management speed.
  • We excluded from marketing management 11.70% of the portfolio of affiliates who had no real possibility of saving.
  • Among the active affiliates, those assigned to the treatment group were 14% more likely (18.41 pp vs. 16.14 pp of the control group) to make contributions in a month and saved 9.4% more per month (approx. US$2.34 vs. US$2.13 of the control group). Thus, among active affiliates, the treatment generated 1.6 dollars of savings for every dollar invested.
  • In the inactive affiliates, although those assigned to the treatment group had a 41% higher probability (0.86 pp vs. 0.61 pp of the control group) of making contributions in a month and saved 46.9% more per month (0.13 dollars vs. 0.09 dollars of the control group), the differences we found are not significant. In other words, the treatment did not generate a significant increase in savings in the inactive affiliates.

Conclusions

High-contact call campaigns in selected groups, using analytical techniques and in accordance with their savings potential, increase the savings of independent and low-income workers, such as those linked to the BEPS program in Colombia. However, their effectiveness depends on whether these workers have already saved voluntarily.

Among those who had already saved, the treatment generated $1.6 in savings for every $1 invested. Among those who had not yet saved, the treatment did not generate significant savings.

One determinant of cost-effectiveness in this case is the price of the calls. Targeting those with a high savings potential allows minimizing the cost of calls while maximizing the potential impact. The savings potential, however, was estimated based on voluntary savings without intervention.

Knowledge of the client through the use of non-conventional variables, or variables that go beyond demographic data, makes it possible to generate personalized campaigns that involve differentiation in the message as well as in the channel and the recipient, maximizing the benefit obtained with the investment. Saving is always possible if it is based on financial education and personalized advice based on an imaginary aligned with the characteristics of the financial product.

Organizations can save money and improve their indicators with a strong investment at the beginning to improve their databases and customer-facing processes, but with a more economical operation in the medium and long term based on updating profiles by observing their reaction to campaigns, their life-cycle changes, emerging events and economic evolution.

For GBA Smart Marketing®, the value and interpretation of data plays a key role in the development and innovation of our portfolio of tailor-made solutions for each of our clients.

Contact us for more information about our services.

Data quality refers to the ability of data to meet user requirements and expectations and to support effective decision making. This includes verifying the accuracy, completeness and consistency of data, as well as eliminating duplicate or incorrect data.

In this opportunity we will share with you how our client, after a digital transformation process, managed to consolidate the Billing, CRM, Collections and S.A.C information of all its consumers. It was key that all this information was located in a single data repository with a single structure and format that would allow practical access for the use of all areas.

Data quality and customer insight

Data quality refers to the extent to which the data is accurate, complete, consistent, relevant and up-to-date for its intended use.

Data quality also relates to data management, which includes data collection, storage, processing and analysis. Good data management can help improve data quality by ensuring that data is clean, organized and accessible.

Our client: a major electricity trading company.

The purpose: it was determined that there was a need to have a single database that complied with data quality best practices, it was necessary to ensure: Existence, Validity, Consistency, Completeness, Accuracy, Relevance. The database had seven fields of information and the new database had to contemplate up to fourteen fields of information that would allow them to have the ability to contact their customers through any communication channel.

The GBA Solution: Smart Data & Holistic Data Quality

Smart Data: We built for our client a database called GOLDEN RECORD in which we integrated all data sources (online/offline). We suggested a new lay out composed of more than forty fields of information. This would allow not only to improve the quality of the existing information but also to increase the knowledge of the final customer. After performing the data quality processes, we applied the information enrichment and complement processes for all the empty information fields.

For identity documents, identity validation was performed for natural and legal persons, guaranteeing the veracity of each piece of information. Finally, having a clean, complete and truthful database, the enrichment of information was carried out in line with Colombia’s personal data protection law.

Smart Analytics & Smart Geo: Using transactional customer information and supported by the Attribute Set by GBA Smart Marketing® we added to the database a segment brand by customer value (RFM). And a sociodemographic segment brand with more than forty attributes of income, expenses, location, life cycle, among others.

Finally, a database that feeds the CRM and maintains the quality of the current information was set up for the customer. As well as the new information that is collected daily not only from current consumers but also from new ones.

Smart Contact: With a quality database, they were diagnosed at the time:

• • Current vs. potential contactability.
  • Current multichannel vs. potential.
  • Permissions (Personal Data Protection Act).

Campaigns were planned to increase multichanneling as well as to obtain and update the permissions of Law 1581 of 2012 (General Regime of Personal Data Protection) as well as the Statutory Law 1266 of 2008 HABEAS DATA that would later allow us to carry out all kinds of campaigns.

Our client’s business, found opportunities in:

1. Identification of customers with high-high and medium-high propensity to purchase other products and services.
2. Opportunities for differentiated attention based on the average ticket of current customers.
3. Identification of industrial and commercial prospects with high consumption potential.

Results

9.11% of the ID’s (Cédulas, Rut, C,E) were inconsistent with the stored information, therefore:

• • Names and/or surnames were supplemented for 63.44% of the records.
  • Erroneous emails were detected in 13.44% of the records.
  • Data were supplemented and corrected in the fields of physical address, physical telephone and cellular phones for 91.22% of the database.
  • 3.11% of the database consisted of deceased persons.
  • 95.66% of the database was geocoded.
  • Fields with information from RRSS were added for 24.66% of the database.
  • 18.76% of the records were duplicated which meant for the client a waste in its marketing budget of 4.55% as well as a significant attrition for the brand.

Good data management improves data quality by ensuring that data is clean, organized and accessible.

In short, data quality is essential to ensure that data is reliable and useful for its intended use, which in turn can improve the effectiveness and efficiency of organizations in decision making.

For GBA Smart Marketing®, the value and interpretation of data plays a key role in the development and innovation of our portfolio of customized solutions for each of our clients.

Contact us for more information about our services.

Multivariate analysis is a statistical analysis technique used to assess the relationship between multiple variables in a data set. In the context of marketing, this technique is used to analyze how different variables, such as consumer demographics, buying behaviors and consumer preferences, relate to each other and how they can influence marketing decisions.

Advanced statistical techniques are used to analyze large data sets and detect patterns and relationships between variables. The goal of multivariate study is to identify patterns and trends in the data that can be used to make informed decisions about marketing and communication strategies.

Multivariate analysis to increase customer insight

By using multivariate analysis techniques, marketers can identify the variables that have the greatest influence on consumer behavior and purchasing decisions. For this client, it certainly presented itself as the optimal action to implement.

Our client presented us with a rather depressed geographical area of a major city with serious problems of mobility, security and public space. There he wanted to build a shopping center, so it was necessary to know what was the right offer in quantity for the different categories of products and services. All in a time span from 2018 to 2030.

The client:

An economic group that within its most important businesses has real estate and the construction of shopping centers.

The problem to solve:

The complexity lay in the fact that, due to the infrastructure works of the regional government and the P. O. T (Plan de Ordenamiento Territorial). This area would go from being the last in consumption potential to evolve in the next 20 years and become one of the most iconic and emblematic places in the city. In addition to being the most interesting from the point of view of consumption potential.

Some of the questions our client wanted us to help him answer were: how many stores should be opened, which categories, which brands, how would we demonstrate to these brands the evolution of the area so that they would be interested in the mall, how many potential customers would we have for each and every store in the next 20 years, how would the current shopping centers in the city impact us, and how would the commercial areas surrounding the mall be transformed, among many others.

GBA Smart Marketing® Solution: Multivariate Study

To respond to this imposing business challenge, we considered that the solution was to know the consumer in the area. At different moments in time and from different dimensions of information. Starting by understanding the main macroeconomic data of the area, the city and the region for the years 2020, 2025 and 2030.

We then conducted different sociodemographic analyses to understand how the population and its age and gender composition would change over time. We also analyzed the consumption behavior of each and every category and compared it with the city’s commercial offer.

We incorporated public transport and its evolution into the analysis, as well as the current and future coverage of the road infrastructure. Its extensions and public transport stations along with the city’s pedestrian and vehicular traffic. In addition to three nearby municipalities from where more than 600,000 people arrived daily for different reasons (work, students, among others).

With these inputs we were able to construct isochrones at different times and hours of the day for the years mentioned, which allowed us to understand the changes in the service area of the future shopping center.

We considered of great value to characterize the floating population arriving to the city for tourism and labor reasons. Not only from nearby municipalities, but also from other cities in the country and even from other countries. Similarly, we incorporated business information to see the impact of business populations on current and future consumption in the area.

Finally, using the Smart Analytics by GBA Smart Marketing® suite of solutions, we segmented the population into eight groups. Each one of them was formed by four sub-segments. Based on this knowledge, we determined the ideal offer by categories, each one with up to three brand options that matched the consumption habits of each group.

The result:

With GBA Smart Marketing®’s consumer segmentation our client established a synthetic reading key of socio-demographic and household behavioral profiles, income and consumption categories.
The analysis of the service area, competitors and POIS (Points of Interest) allowed us to know which categories were most in demand and which were not.
Changes in the service area, based on the evolution of road infrastructure and safety conditions. They made it possible to dimension the size of the target population at each moment in time.
Thus, we understand that the multivariate study is used to identify the factors that have the greatest influence on customer satisfaction, such as product quality, price, customer service and brand. By understanding the factors that influence customer satisfaction, companies can take steps to improve the customer experience and increase customer loyalty.

For GBA Smart Marketing®, the value and interpretation of data plays a key role in the development and innovation of our portfolio of customized solutions for each of our clients.

Contact us for more information about our services.