Businesses of all sizes can benefit from market research, so we built a platform that expands access through different approaches, such as the free plan, in our pricing policy, through the entire knowledge base placed at their disposal, through the intuitive usability and the high quality of the content.
Every piece of information you find in Mapfry undergoes a rigorous transformation process until it reaches the state where its characteristics and proportions reflect reality.
And some people still tell us:
“Oh, but just download the data from the IBGE website and put it on the map”
In fact, we are aware of other solutions that are limited to placing the information on the map, which is why it falls far short of Geomarketing.
To give you an idea, just one of the data releases from Census 22 has more than a thousand variables.
Merely understanding them already represents a complex task, but this still could not be called a geo-demographic basis, which are data schemes that reflect population and economic structures.
The quality of Mapfry's results is not just a matter of data collection, but of advanced processing and validation methods.
Stages of the process
Data Collection
We integrate information from a variety of sources, such as censuses, traffic data, mobile devices, and business records.
Survey and collection of relevant geographic databases: Census, PNAD, POF, CAGED, BACEN, IRS, Ideb and hundreds of other sources.
Why it's important: A dataset tells a story that is more complete and closer to reality than any individual package.
This is because each research has its limits of scope and scope, but by organizing them together, we begin to compose a mosaic that represents population and economic dynamics.
Cleaning and organizing the bases
To correct specific errors and inconsistencies.
Why it's important: It's rare, but it happens a lot, for bases to have problems not seen in previous treatments, simply because Geomarketing's use case is very specific.
When we evaluate the original bases against those that are already in our database, we can notice empty fields, with inconsistent and even inverted information.
Standardization
Compatibility of information in comparable proportions
Why it's important: The nature and objectives of the research vary, as does its area of coverage, period of relevance, and level of geographical detail.
Therefore, they need to be placed in minimally comparable proportions and scales.
Quantitative methodologies applied to the population
Segmentation and grouping by tracks to reduce complexity.
Why it's important: After all the previous treatments, what we have is a consistent base, but still very broad in its variables and not fully contextualized with each other.
The information may be out of date and refer to different periods and it is up to us to adjust the time so that they all reflect the same time interval.
That is, they say things that are correct, but have not been tied together in such a way as to allow for a cohesive and continuous analysis.
To this end, we use statistical modeling and spatial projection techniques to estimate the population and its dynamics, such as fertility, mortality, migration, and other demographic projections.
Geoprocessing
We transform properly treated bases into precise coordinates on the map.
Why it's important: We use advanced algorithms to correct inconsistencies and detect patterns. Our process increases accuracy by up to 30% compared to traditional methods.
Rules for reading data
Formation of relevant sets, choice of graphic elements and color palettes that explain the phenomena under visualization.
Why it's important: Thematic maps and interactive dashboards that reveal areas of high potential, population flows, and market gaps.
Real-time processing
Delivery of information as the user navigates. You don't have to select the data and the area you are going to view, just move the map and the information is displayed at the ideal scale.
Why it's important: When an analysis is happening, you don't want your reasoning to flow naturally between the information.
Load capacity
Mapfry has the largest user base and needs to be ready to provide simultaneous access to everyone.
Why it's important: Much more than having a database connected to a system, this integration needs to be fast and reliable.
To this end, we take several measures to distribute requests, reduce response time when any of them are queued, and increase processing availability at peak times.
After all, you never know when you're going to need information to close an important decision.
Documentation and tutorials
Details of the information and the possibility of using it for better use
Why it's important: There are many possibilities offered by Geomarketing, you should find references on the most important ones and how they can be applied to your business.
Mapfry Differentials
- Measurable results: Our platform is open, with the Free plan allowing access to content that can be verified in places where you are familiar.
- Continuous innovation: We started in 2020, we are already among the 3 main Geomarketing platforms and we will continue accelerating.
- Technical authority: Our team combines expertise in technology, statistics, and geography.
- Methodological domain: We are the only team that has a Master Statistician in Demography, which allows us to work at the highest academic standard.
Resources we use
- PostgreSQL/PostGIS
- Python For data analysis, geodata manipulation (libraries such as GeoPandas and Folium) and machine learning
- R for statistics and visualization
- Big Query and Apache Spark
- QGIS for testing and validating geographic data
Theoretical basis
Model Assisted Survey Sampling by Särndal, Swensson and Wretman
Introduction to Variance Estimation by Wolter
Geographic Information Systems and Science by Longley, Goodchild, Maguire, Rhind
Python Geospatial Analysis Cookbook by Joel Lawhead
Applied Spatial Data Analysis with R by Bivand, Pebesma, Gómez-Rubio