MapFast (Posts about cartography)

MapFast example files

MapFast — Sun, 18 May 2025 00:00:00 GMT

Create a map with MapFast : example files

Use this page to understand how to structure your data to create a map with MapFast.

💡 MapFast automatically detects the best column to use for geocoding (i.e., the process of converting a textual location into GPS coordinates) and the best columns to display on the map. As a result, the input format is pretty flexible!

I referenced a few examples below, that you can download and reuse as a base file. 🙂

Example data - World map

The following is an Excel file containing the World GDP per country, from the WorldBank.

Preview & download this dataset here: [Google Sheet link](https://docs.google.com/spreadsheets/d/14hsrtEHiO9bXHkFgpgOejF9cqiA-xddp/edit?usp=sharing&ouid=110590424152347804947&rtpof=true&sd=true).

Initially, the dataset contained many years, but I filtered the data and extracted the results into a new sheet to be sure that I had only the values for year 2021. The rest of the file is clean: it has clear country names and the GDP per capita column is numerical. The “code” column could be removed safely.

💡 See also: How to make a World map

Example data - State map (United States of America)

The following dataset is a dataset I used to create a map to count the number of Electric Vehicles registration in the US. The original dataset contained non-EV columns that I cleaned, I also manually computed “per capita” information to be able to create a normalized choropleth map.

Preview & download this dataset here: Google Sheet link.

All the states are named clearly in the “State” column; so MapFast can detect easily that the dataset is entirely about the USA. The number of columns is low (under 20) so it won't cause any issues during import. All columns are numerical.

You can do the same kind of dataset for any country. Instead of States, it can be other administrative areas: Cantons, Regions, Departments, …. for another country.

Example data - categorical colors

The following is a very simple csv file that I created manually to explore categorical maps. It contains Climate Zone categorization for world countries. **

Preview & download this dataset here: Google Sheet link.

This file is very simple, making it very easy for MapFast to understand how to make a map out of it. The “Climate Zone” column will be detected as a list of strings, thus categorical data.

The next step will be to change the colors into the web app!

💡 See also: How to make a colored map

Example data - real estate map

The following is a completely different dataset as it contains precise adresses. It’s an extract of a real estate list of apartments for sale in Antwerp, Belgium in 2025.

Preview & download this dataset here: Google Sheet link.

The dataset contains a clear address column that MapFast will detect and use for geocoding. The rest is a mix of categorical and numerical data, perfect for creating various maps!

There was no difficulty with this dataset as it’s already clean.

Example data - point of interest & street names

This dataset is a list of restaurant in Dubai (original link). It contains heterogeneous adresses: sometimes precise addresses, and sometimes just streets or points of interest.

Preview & download this dataset here: Google Sheet link.

It was a bit messy and needed some cleaning and was also too large, so I just kept and excerpt of it.

It has an “address” column which can be geocoded and a “subCategory” column that can be used to map restaurants types to different colors, which is nice for a map. 🙂 Thanks to the “location” column referring to “Dubai”, MapFast will focus on geocoding addresses in Dubai.

Conclusion

I hope these examples helped you to better grasp what kind of structured data files you can send to MapFast to create your map. Even though MapFast can automatically detect how to read your data, it can't yet clean or rearrange it for you.

Here is a list of general advices:

✅ One row = one location on the map: for now, MapFast does not support multiple values per location (e.g. multiple years of data). You need to filter your data so that each location has exactly one row in the file.
✅ Keep your file simple: remove all unnecessary data from your Excel file and keep only one sheet (if using an Excel file).
✅ Provide clear location names: for example, use city names (e.g. New York, Paris, London) or country names. Avoid abbreviations, and if you use zip codes, make sure to include a clear context (e.g. "94110, USA" instead of just "94110"), so MapFast can properly interpret the location.
✅ Ensure clear column types: if a column contains numbers, make sure it’s recognized as numeric in Excel. Otherwise, it may be interpreted as a categorical value (e.g. “103 meters” instead of 103 in a column named “distance”).

If you still can’t get your map done, contact us in the app within the chat. You can also check out our other tutorials here.

Happy mapping ! ⭐

Choosing the right Classification Method for your Choropleth Map

MapFast — Sun, 03 Nov 2024 00:00:00 GMT

Choosing the right Classification Method for your Choropleth Map

When building a choropleth map, one of the most crucial choices you will make is how to classify your data, i.e. group your data into color-coded categories on the map.

This image depicts 3 different maps with 3 different classification methods : Linear, Quantile, Fisher-Jenks. In this case (italian wine production), Fish-jenks is a better choice.

Illustration of the three classification methods we will compare in this article.

But how to pick the best option? In this article, we’ll explore three common classification methods: continuous linear, quantile, and Fisher-Jenks, and discuss how to decide the number of intervals for your map. We’ll look at examples with two maps for each method.

By the end, you’ll be able to make more informed choices about classifying your data and create maps that tell the most accurate story possible. Let’s dive in!

This article is part of our guide on choropleth maps. Be sure to check it out:

A guide to Choropleth Maps

1. Linear Classification: smooth transitions across values

Linear classification (or “continuous linear gradient”), maps your data along a smooth color scale, where each unique value corresponds to a shade of color between minimum and maximum values.

This approach avoids setting predefined intervals, instead allowing for a seamless transition in color, which can be effective for continuous data without clear clusters.

Map 1: Italian max. temperature in summer (average). Linear classification.

Map 1: The temperature data is a good fit for the linear map: we can see a smooth gradient of red shading gradually darker. On the other side, we have two outliers (low values) that are highlighted more than they should be on the map.

Map 1: Italian real estate median price per province. Linear classification.

Map 2: the linear gradient is not optimal for the real estate housing price map, as it overemphasizes extreme values (like Milan). As we will see below, a quantile or Fisher-Jenks classification might be better.

↘️ When to Use Linear Classification:

Best for data without distinct groupings, providing a visually smooth representation.
Works well when you want to show subtle variations across regions, like temperature or elevation.

However, linear classification can be misleading when there are extreme values, as these outliers might dominate the color scale and obscure smaller differences. For example, in the real estate prices map, extremely high median values cause all other values to appear lighter, making the map harder to interpret for regions with lower price but still significant differences.

💡 Tip: Use linear classification when your data follows a continuous, predictable range and doesn’t have extreme outliers.

2. Quantile Classification: equal number of data points

Quantile classification is all about balance, with each interval containing an equal number of data points. Say you have 100 data values and five intervals—each interval will include 20 values. This approach is useful for highlighting relative differences between regions, especially when you want each color on your map to cover a similar number of areas.

Map 1: Italian max. temperature in summer (average). Quantile classification.

Map 1: The temperature map is now more “clustered”. This is the result of the classification: instead of having a linear gradient (as many classes as there are values), we now have 6 groups. Compared to the linear gradient, differences between minimum and maximum areas are clearer.

Map 1: Italian real estate median price per province. Quantile classification.

Map 2: Quantiles classification now shows the regional patterns of the real estate map. patterns in the data—which is much more interesting! The drawback is that really high values are now mixed in a group with lower values. Increasing the number of groups (>6) might mitigate this problem, to a certain point.

Notice that intervals can be tricky to explain, as they may start small and widen later.

↘️ When to Use Quantile Classification:

Ideal for data with a broad distribution, like income or property prices, where you want to emphasize relative differences between regions.
Useful for audience-friendly maps since each interval has a similar number of areas, providing a balanced visual effect.

One drawback is that quantile classification can exaggerate small differences when values are closely packed together. For instance, in a map of median income, minor variations in middle-income areas could seem more significant than they are.

💡 Tip: Use Quantile classification for visually balanced maps, but be cautious with tightly clustered data, as it might create artificial “breaks” between similar values.

3. Fisher-Jenks Classification: natural patterns

Fisher-Jenks classification groups data into intervals that minimize the variation within each class, effectively highlighting “natural breaks” in your data. It’s a more complex approach but can create a highly intuitive map that emphasizes patterns and clusters.

Map 1: Italian max. temperature in summer (average). Fisher-Jenks classification.

Map 1: The Fisher-Jenks is here a successful mix between the Linear and the quantiles classification : regional differences are clearer than with Linear while the color gradient is smoother than in Quantile.

(Author’s note: while the fact we have 7 groups instead of 6 helps, this readability is still a feature of this method.)

Map 1: Italian real estate median price per province. Fisher-Jenks classification.

Map 2: This map is also clearer than the quantile version, as high values stand out clearly while still showing a smooth gradient.

Just like the Quantile, it produced steps that might be harder to explain. But in this case, IMO, they are more intelligible. This is not always the case with Fisher-Jenks.

↘️ When to Use Fisher-Jenks Classification:

Ideal for data with clear clusters or natural breaks, such as temperature zones, pollution levels, or population density with distinct regional differences.
Great for analysis-focused maps that need to reveal subtle but important patterns.

While Fisher-Jenks can give a more nuanced picture, it’s also harder to explain to a non-expert audience. Additionally, it may create intervals that don’t follow a predictable pattern, making legends slightly less intuitive to interpret.

💡 Tip: Use Fisher-Jenks when accuracy and pattern detection matter most—just keep in mind that the classification may be harder for beginners to interpret and also harder to explain for you.

Choosing the right number of intervals

Deciding on the number of intervals (or “steps”) is as essential as choosing the classification method. Too many intervals can make the map visually overwhelming, while too few might obscure important details. Generally, 5-7 intervals is a good balance, keeping your map informative without being cluttered.

Quantile example

The image depicts a comparison of the number of classes (5 - 7 - 9) for Quantile classification.

Quantile classification example with 5, 7 and 9 groups. Data: Italian real estate prices (€/m2).

As you can see, the quantile classification does already a good job at highlighting the main global differences, even with only 5 classes. Increasing the number of classes to 9 doesn’t change the result much but makes the legend more complex.

However, the classification produced naturally nice steps, easy to understand (i.e. rounded steps, with often a range of 200€/m2).

Fisher-Jenks example

The image depicts a comparison of the number of classes (5 - 7 - 9) for Fisher-Jenks classification.

Fisher-Jenks classification example with 5, 7 and 9 groups. Data: Italian real estate prices (€/m2).

Increasing the number of steps from 5 to 7 also helped the Fisher-Jenks classification to better highlight variations in price. However, we might not need to increase it up to 9, because the map looks almost the same for the added complexity of two more steps.

You can also notice the difference here with quantile 9: the ranges fluctuate between 200-300€/m², then quickly widen. This can be confusing for some readers 🥸

In conclusion

For beginner-friendly maps: 5-7 work well, especially for straightforward data where less detail is acceptable.
For detailed analysis maps: 7-9 intervals can provide more granularity, but be careful not to overdo it—more intervals don’t always mean better insights.

Wrapping It Up

Choosing the right classification method and number of intervals for your choropleth map can make all the difference in how clearly and accurately your data is communicated. Whether you’re working with a continuous linear gradient, quantile, or Fisher-Jenks classification, think about what you want your map to show and who your audience is.

With these tools, you’ll be on your way to creating visually impactful, insightful maps that convey the true story behind your data.

Learn more:

How to make a colored map with MapFast
How to make a World map / How to make a Europe map
Why maps require normalization (coming soon)

Happy mapping! 😊

A guide to Choropleth Maps

MapFast — Sun, 27 Oct 2024 00:00:00 GMT

A guide to Choropleth Maps

Have you ever glanced at a colorful map and immediately grasped which regions were wealthier, more densely populated, or facing unique challenges? Chances are, you were looking at a choropleth map.

These maps use color shading to represent data across specific areas—like population density, unemployment rates, or even levels of internet access—making complex data instantly understandable.

An example choropleth map: World CO2 Emissions per Capita 2022.
Source: How to make a World map

In this guide, you will learn how choropleth maps are made and their strengths and weaknesses. Let’s go!

What are choropleth maps used for?

Choropleth map definition is “a map which uses differences in shading, coloring, or the placing of symbols within predefined areas to indicate the average values of a particular quantity in those areas.”

Said differently, choropleth maps are a visual storytelling tool for geographical data. They can be applied to any domain, but they are frequently used in Politics, Economics, Public Health and Real Estate. They are also easy to read and understand, if made correctly.

Let’s look at some common examples 🙂

Politics

Choropleth maps can be used show political patterns across regions, making it easy to see where support for different parties is strongest and how demographics influence voting trends.

Here is an example from the 2020 U.S. Presidential Election :

Example 1: 2020 U.S. Presidential Election Results Map. (source link)

This map provides a quick overview of which states voted for the Democrats (in blue) and Republicans (in red). The use of each party's signature color enhances the map's clarity, making it easy to see which states supported each candidate and identify regional voting patterns.

However, political maps like this can be misleading, as they don’t account for the Electoral College system. Winning many small states may appear as a large area on the map but doesn’t necessarily lead to an electoral win.

In this case, the issue has been mitigated by CBS News with the addition of a bar chart to the map, as it complements the visual by showing each candidate’s actual electoral vote count and balancing this inherent bias.

💡 This map has only two colors. Even if they provided a legend with divergent shades of blues and red, it could could be classified as “binary choropleth map”, a type of simple maps that only display two colors.

Economics

Choropleth maps are powerful tools for visualizing economic data, as they use color shading to show variations across regions, such as income levels, unemployment rates, or GDP per capita.

In the context of economics, maps are not only used for communication but also for analysis. These type of visualizations are necessary for policymakers, economists and scientists to understand economic disparities and trends, and to communicate about them.

Example 2: A GINI Index Map by Statista. (source link)

The above map successfully illustrates how a choropleth map can show regional patterns. The darker the red, the higher the inequality, making obvious to spot the highest values at a glance.

It’s fascinating how we can instantly see these global patterns—South America and Africa have some of the most intense inequalities, while Europe and parts of Asia present a more balanced income distribution.

However, the map does have its limitations. Smaller countries, like Luxembourg or island nations, are not shown due to their small size. Simplicity comes with some trade offs : adding all countries would certainly make this map more complex to read.

💡 This multicolor choropleth map uses a gradient of related colors—from yellow to dark red—to represent varying levels of inequality. Usually darker shades are typically associated with higher values, indicating greater inequality.

Public Health

Remember that strange time when we were all stuck at home, locked down? I’m sure you remember watching TV and seeing COVID-19 maps on every channel…

They helped us understand how the virus was spreading, whether globally or just in our local areas. Like political choropleth maps, these COVID maps became a part of our daily lives, essential for conveying quick, clear information to people around the world.

Example 3: Reported COVID-19 cases in March 2020, from CNBC. (source link)

The above map shows the global spread of COVID-19 cases as of March 17, 2020, with darker blue shades indicating higher numbers of confirmed cases. At a glance, it highlights the initial hotspots of the pandemic—China, Italy, Iran, and South Korea—as well as emerging cases in the U.S. and parts of Europe.

You’ll notice that the legend is not linear, which is necessary because extremely high values, like China’s 81,000+ cases, would otherwise overshadow lower numbers. This scaling helps highlight variations across countries without letting the highest numbers dominate the map.

💡 This map uses different shades of the same color, it can be categorised as a “**monochromatic choropleth map**”. Again, darker shades are usually associated with higher values. You might also notice this world map does not have the same shape as the previous one. This is because it has a different earth projection. Learn more about this here : [article about projections, coming soon].

And many others…

You now got the idea. It is impossible to restrict choropleth maps to the 3 examples I cited above. Other well known includes real estate prices, demographic data, etc…

The importance of colors

Choosing the right color palette for a choropleth map is crucial to ensure patterns are clearly and easily identifiable by the viewer.

There are multiple choices :

Binary Colors: For two-category data, these colors provide a clear, high-contrast distinction between categories. Binary color schemes work well for yes/no, on/off, or binary classification data (like Democratic vs. Republican).
Qualitative Colors: These colors work best for categorical data where there’s no intrinsic order or gradient. Each category gets a distinct color, ensuring clarity in differentiating categories like land use types or demographic groups.
Sequential Colors: Used for numeric or ordered data that moves in a single direction, often with a light-to-dark gradient. This is perfect for showing intensity, frequency, or amounts, such as population density or income levels.
Divergent Colors: These are ideal for data with a meaningful midpoint, like temperature changes or economic deviation. A divergent color scale uses two contrasting colors with a neutral midpoint (e.g., white, gray) to highlight values on either side of that midpoint.

Binary color scale example

Qualitative color scale example

Sequential color scale example

Divergent color scale example

Colors significantly influence perception. For instance, red can imply negativity, while green often suggests positivity. Darker shades typically represent higher values, while lighter shades indicate lower ones.

Misusing color can lead to viewer confusion, so careful selection is essential!

Learn more in this detailed article: how to choose colors for choropleth maps.

The importance of classification

Beyond color choice, the classification method—how values are grouped into categories or "steps"—plays a key role in shaping the map’s visual impact and readability.

Different methods like linear, quantiles, and Fisher-Jenks classification each affect how data is distributed across color gradients :

Linear Classification (continuous): This approach applies a continuous color gradient across the entire range of data values, with each unique value mapped to a specific shade between the minimum and maximum. It works well for continuous data without clear clusters, providing a seamless transition in color across the range. However, it can be misleading in datasets with extreme values, as these outliers can dominate the color scale and obscure smaller variations, as seen in some COVID-19 maps.
Quantile Classification: Here, each class contains an equal number of data points, creating a balanced visual distribution. This method is useful for highlighting relative differences, making it ideal for datasets with gradual but consistent variation. However, quantiles can exaggerate small differences if most values are close together, potentially overemphasizing minor variations as significant shifts in color.
Fisher-Jenks Classification: This method groups data to minimize differences within each class, highlighting natural breaks in the data. Fisher-Jenks works well when there are clear clusters or gaps, giving a more accurate picture of patterns compared to linear or quantile methods. However, it leads to less intuitive steps harder to explain.

This image depicts 3 different maps with 3 different classification methods : Linear, Quantile, Fisher-Jenks. In this case (italian wine production), Fish-jenks is a better choice.

Illustration of the three classification methods : Linear, Quantile, Fisher-Jenks. For this data, Fisher-Jenks seem to be optimal.

In the above example, Fisher-Jenks seems the best option between Linear (highlights only extreme values) and Quantile (highlights too many small variations). At the end, it really depends on the data you have and the best is to try different options visually!

💡 Classification methods are really important to understand to make meaningful maps. Sometimes the choice of the classification steps is obvious: Linear will not work with outliers, for example. But sometimes it is not and the choice between classification methods will depend on the story you want to tell.

Learn more about classification methods in this article with example maps: Choosing the right Classification Method for your Choropleth Map

Why normalizing data matters

The most common error made with choropleth maps is to forget to normalize data. Normalizing data is the process to adjusting values to a common scale to ease comparisons.

For instance, you would normalize a “GDP per State” by the population per each state and obtain a “GDP per capita”. That way, you’ll effectively get a picture of the states that have the highest standard of living, instead of the states that have the largest population.

Map : Population total - per State, USA

Map : Annual GDP - per State, USA

Map : Annual GDP per capita - per State, USA

Notice how the “Population per State” map and “Total GDP per State” map are similar. On the other side, the per capita, normalized, map is showing a complete other reality.

Normalizing per capita (population count) is the most common. But it is not the only possibility and the choice must be done regarding the data you plot on map.

Here are other examples:

Original	Normalized	How?
Total Income by State	Median Income per household	per household normalization
COVID-19 Cases by Country	COVID-19 Cases per 100,000 people	per capita normalization
Electric Vehicles Count per City	Share of Electric Vehicles count per City	total vehicles count normalization
CO2 Emissions by State	CO2 Emissions by State per km2	per area normalization

Sometimes you don’t necessarily need normalization. For example “Wine Production per Region” can be normalized as “Share of Wine production per Region”, but the wine production counts are not likely to be the same as the population counts anyway. In this case, you have the possibility to make two different maps but complementary maps.

💡 Normalization also reduce the bias introduced by varying geometry sizes : large areas can be misleading, as they often have lower population densities. In election maps, there’s a common saying: “Land doesn’t vote”, highlighting this issue. To address it, the best approach is to normalize by area size or use symbol maps to more accurately represent data distribution.

Learn more about normalization : Why choropleth maps need normalization, coming soon

Pros and cons of Choropleth Maps

Like everything in life, choropleth maps come with their strengths and weaknesses:

→ Pros:

✅ Intuitive: Anybody can understand maps because of their visual aspect. They are more intelligible than raw numbers or bar charts for geographical data.
✅ Great for comparison: They’re awesome for comparing different regions at a glance. They allow discovering regional patterns that could not be seen using other types of graphs (e.g. a bar chart).
✅ Multi-purpose: They can be used for season statisticians driving insights or marketers using them as a communication medium. They are useful for both analysis and communication.

→ Cons:

❌ Prone to bias: Different colors and classification scheme can produce different maps, telling different stories. Adding to this the frequent need for normalization, maps can be easily made incorrectly or convey the wrong message.
❌ Big areas can mislead: Take Australia, for example. It might dominate visually on a world map, but population-wise, it’s mostly empty. So, choropleth maps can make low-data areas seem more important than they are. Different earth projections may also amplify this effect.
❌ Boundaries matter: There are many ways to represent geographic areas : political borders, cultural borders, naturals borders. Many areas are disputed. At the end, there are a lot of choices to do and it might not please everybody.

Wrapping It Up

Choropleth maps are an incredible tool for visually communicating geographic data. Just make sure to normalize your data, choose your colors carefully, and you’ll be on your way to creating insightful, visually appealing maps that tell a true story. 🙂

The good news is, we created a lot of resources to help you create meaningful maps with MapFast!

You should take a look at :

How to make a World map / How to make a Europe map
How to make a binary map (coming soon)
Why maps require normalization (coming soon)
Choosing the right Classification Method for your Choropleth Map

⭐ Happy mapping ⭐

How to create a Europe map with MapFast?

MapFast — Sun, 20 Oct 2024 00:00:00 GMT

How to create a Europe map with MapFast?

💡 This article is a step-by-step guide on how to create your own choropleth Europe map from Excel/CSV data. Let’s dive in!

The goal is to create a map that shows the net contribution to the EU budget. It will highlight member states that are either contributors or beneficiaries:

You will learn how to create a map like this in the following tutorial.

Import the data

We will use data from an article on the Statista website: “Which Countries are EU Contributors and Beneficiaries?”. The data is from 2021, but still relevant to plot.

Since the data was presented as a graph, I manually extracted the 27 values into a table. You can download the exact data from this Google Sheet (File > Download as Excel).

Here is an excerpt of the data (year 2021).

Now, let’s open the MapFast app and select “Europe” for the map area:

The first step asks you to select a country or region.

Once the preview map has loaded, we can upload our data by clicking the “Import Excel/CSV file” button. Once the upload is complete, you should be able to see the data in the table view on the right:

The second step allows you to import any Excel or CSV file into MapFast.

This view enables you to check that you have one row per location. This is a very important requirement. It doesn’t matter what the column names are in the imported file, but there must be a clear column containing the location, and only the location.

Review associations between rows and countries

At this point, MapFast will try to associate each row of the CSV file with a country from its database. Once finished, we need to review the matches, to be sure that everything will be displayed correctly.

After clicking “Next”, you’ll see a table with 2 columns: the first column contains locations from the CSV file we imported. The second one contains the country that was matched with the row. That’s what we need to check.

Most of the lines appear green, but some may show up red because no match has been found. In our case, we only see European states that are not members of the EU: Cyprus, Switzerland, etc.

Be careful when reviewing the associations : a green association doesn’t always mean it’s correct.

The map on the right shows the completeness of our data. In this case, we have nothing to do, all 26 EU state members have been matched correctly ✅

Map preview

The last step is really straightforward: we just need to pick the data column. In our case, it’s ‘Net Contribution (In Million)’.

The final step is to pick the correct data column and load the map!

The map will show a preview of the data. This is useful to check that everything is fine (I actually had to redo the work for this tutorial, because I noticed an error in my data at this screen!).

We will customize its appearance afterward. Click “Load map” when you are ready!

Customization

Once our map is loaded, we can start the fun part: customizing the map ! 🙂

Adding a title, source, and legend or values is really important to give context to the map reader.

Let’s customize it :

Set the appropriate color and scheme.
1. As I want to have a clear distinction between positive and negative values, I picked a Fisher-Jenks & Red-Blue divergent color scheme (in-app : ‘Background > Classification > Fisher Jenks 9’).
2. I also set a transparent color for missing values, with hatch. And modified the stroke color to black see the hatch. (in-app: ‘Stroke > Inside stroke’)
Add map title and source. It is very important to add context to a map, that gives credibility.
1. I’ll go for the same title and subtitle as in the original Statista article (in-app: ‘Text > Title’)
2. And credit Statista and the European Commission for the data source (in-app: ‘Text > Source’).
Add legend or values. The legend is essential if values are not properly displayed on the map. But in this case, I found interesting to display the country name and its value. (in-app: ‘Text > Label column = name’ & ‘Text > labels callout = Manhattan’).
1. Select the country name as label
2. Remove all unnecessary countries that are not in the E.U.
3. Modify all labels to include the value
4. Select manhattan callouts
5. Place them on the map aesthetically

💡 Creating labels like this takes time… we’ll find a more automated solution in the future. Meanwhile, a legend is often as good as displaying the values, while not overloading the map. You can still place all the values on the side of the graph if you need to.

Final result

That’s it! You now have a beautiful Europe map. 🙂

The resulting map of the EU budget contributions and beneficiaries

You can export it using the “Download” button as a PNG or SVG with the resolution that you need.

💡 If something went wrong with your own map… send us a message via the in-app chat. We will help you!

Happy mapping ✨

How to create a World map with MapFast?

MapFast — Sun, 20 Oct 2024 00:00:00 GMT

How to create a World map with MapFast?

💡 This article is a step-by-step guide on how to create your own choropleth World map from Excel/CSV data. Let’s dive in !

The goal is to create a map that shows the actual per capita CO2 emissions. It will highlight countries in the world that have the largest impact on global emissions:

You will learn how to create a map like this in the following tutorial.

Import the data

We will use data from the amazing website “Our World in Data”: https://ourworldindata.org/co2-emissions. I will download the “per capita emissions” data as CSV and modify it.

The file contains one row per year and per country. But as we want to depict countries, we need to ensure that the file contains only one row per country.

So we can filter out every year except the most recent one, 2022. You can download the exact data from this Google Sheet.

Here is what the data looks like, after filtering on year 2022

Now, let’s open the MapFast app and select “World” for the map area:

The first step asks you to select a country or region.

The second step allows you to import any Excel or CSV file into MapFast.

Review associations between rows and countries

Most of the lines appear green, but some may show up red because no match has been found.. In our case, that’s normal : it is because our dataset contains aggregated values for ‘Asia’, ‘Europe’, etc.

This step is about matching locations from our file to a map.

We also need to check green lines, because it might not be associated with the correct country. I found two cases :

‘Asia (excl. China and India)’ incorrectly associated win ‘Republic of China’;
‘European Union (27)’ incorrectly associated to ‘Tokelau’.

To remove both associations, we can click on ‘Republic of China’ and remove the association by selecting the blank option. We do the same for ‘Tokelau’.

Be careful when reviewing the associations : a green association doesn’t always mean it’s correct.

The map on the right shows the completeness of our data. If the map is entirely green, it means we’ve covered the entire world. Good!

Map preview

The last step is really straightforward: we just need to pick the data column. In our case, it’s ‘Annual CO2 emissions (per capita)’.

The final step is to pick the correct data column and load the map!

The map will show a preview of the data. We will customize its appearance afterward.

Click “Load map” when you are ready!

Customization

Once our map is loaded, we only need to customize it the way we want.

Adding a title, source & legend is really important to give context to the map reader.

The basics are:

Set the appropriate color and scheme. For more detailed color differentiation, I’ll pick ‘Quantiles 9’ (in-app: ‘Background > classification’).
Add map title and source. It is very important to add context to a map, that gives credibility. I’ll go for “World CO2 Emissions per Capita, 2022” and cite “Our World in Data” (in-app : ‘Text > Title’ and ‘Text > Source’).
Add legend. The legend is an absolute necessity if values are not properly displayed on the map. I chose a horizontal legend since the map is horizontal (in-app: ‘Elements > Legend > Continuous Horizontal’). I also adjusted the text values.

Final result

That’s it! You now have a beautiful world map. 🙂

You can export it using the “Download” button as a PNG or SVG with the resolution that you need.

map(29).png

💡 If something went wrong with your own map… send us a message via the in-app chat. We will help you!

Happy mapping 🤩

How to make a colored map with MapFast?

MapFast — Sun, 20 Oct 2024 00:00:00 GMT

How to create a map with MapFast?

Have you ever dreamed of creating a colored map to showcase statistics and analytics but never knew where to start? I understand—it’s complicated. But you're in the right place now!

The image depicts a map of the population density per km2 in Northern Ireland.

The example map we will build in this tutorial.

This tutorial will help you to create your own map in 5 steps:

Learn more on color-coded (choropleth) maps: A guide to Choropleth Maps

1. Pick the region or country

First things first, you need to specify either the country or the region that represents the context of your data.

You will have two choices:

Countries: USA, India, France, United Kingdom, Tuvalu... you can choose between 200 available countries.
World areas: North America, South America, Europe, Asia, Oceania are available. You can also pick "world" to create a World Map by countries.

You can choose between 200+ countries or world areas for your colored map. Let’s continue with the United Kingdom.

Once you've picked a country or region, you can choose the type of areas your data will cover. Usually, you'll have administrative boundaries: regions, provinces, districts, or municipalities.

At this step, the choice needs to closely reflect your data. If you want to display every municipality in a country, you will need to have a dataset where each row represents a city.

‘Districts’ are the lowest administrative bounds for cities/electoral districts in the UK.

Finally, you can narrow it down to a specific subset area, allowing you to focus on a particular part of the country or region that you selected.

In this case, we can select ‘Northern Ireland’ to effectively create a district map focused on this part of the United Kingdom.

It can also be left empty, even if you don’t have all the data for all the areas on the map. You’ll still be able to remove it later.

2. Import your data (or edit manually)

It's usually more convenient to have an Excel or CSV file for your data, especially if you have many values to display on the map. For instance, France has over 35,000 cities… nobody wants to map them all by hand!

But if you don’t, you can still manually update the table and fill it with your own values.

MapFast provides two options: option 1) import of an Excel/CSV file, and option 2) manual fill of the table.

This part is the most important part : to ease the creation of the map, your file should at least contain two columns:

the "location" column: It must contain the list of the names for each area that will be added on the map. E.g., the name for each city if you chose "municipality" previously.
the "data" column: It can be either numbers (ex: GPD, sales, or % values) or categorical data (e.g Fruits, Brands, etc.)

Following the previous example, we imported a CSV file containing district information for Northern Ireland (UK). Get it here : https://en.wikipedia.org/wiki/Local_government_in_Northern_Ireland

💡 Good to know :

you don't need to name these columns specifically. The location column will be detected, and you'll be able to choose the data column
you can click on the table and edit the data if needed

Overall, you will minimize import errors if…

✅ Each row of your dataset is consistent with the type of data you chose in step 1 (e.g. each row is a city of you selected cities).

✅ Each line is a unique location.

✅ Remove unnecessary (or blank) lines or columns in your dataset.

After importing your data, you should see it inside the table. You can still click on cells to modify the content to fix some errors you haven't noticed before.

3. Check matching and pre-visualize

Behind the scenes, MapFast associates each row of your dataset with a geometry (a polygon on the map). This crucial step is called the “matching“, between textual locations and their respective representation on a map.

MapFast does this automatically for you, but you can still check that every match is correct. In this case, you can click on a row and add/modify the associated location, or remove it by setting the line to blank.

We had a perfect match here. But for the sake of the tutorial, we removed the match for “Derry and Strabane” so that you can see how it looks and how to fix it.

The map on the right will show all areas in green that have been associated with your data. Usually, the greener the better! But as mentioned earlier, it's okay if there are a lot of gray areas if your data only concerns a subset of the country or region.

Once you double-checked the matching, you can choose the data to see on the map. Select between all the columns of your dataset.

We choose to display the population density in Northern Ireland, although we had other columns available.

The map helps us be confident that all values are consistent. It will show you what to expect, in case you want to go back and modify some values.

4. Customize map

Here we are! Now we are in the most interesting part: customizing the map.

Customizeing the map

You have several options in the sidebar :

background: adjust the colors and how to match colors with values (color scheme). You can also set the color for null values.
stroke : change inside and outside stroke width and color ;
text : add free floating text and labels on the map.
elements : add rectangles, lines to highlight parts of your map. You can also add a legend from this menu.

After 10 minutes of experimenting, we finally have a professional-looking map for our Northern Ireland (UK) data.

💡 What makes a good map ? Usually, maps have an explicit title, a legend, and cite the source of the data. Colors should be picked carefully to reflect the story you want to tell.

5. Export your data

Now that you have a beautiful map, it's time to share it to the world. Or internally, to your fellow colleagues. Or just for you (maybe making maps is your guilty pleasure…).

You have two export options: PNG and SVG. The latter is useful if you need to open the map in more advanced software, like Illustrator or Inkscape.

Exporting the map

At this point, you can freely download your map and enjoy the final result:

The image depicts a map of the population density per km2 in Northern Ireland.

Northern Ireland, Population density per km2 in June 2022. That’s it!

I hope this tutorial helped you create your first map !

Explore our other ressources:

You can also send us a message directly in the app for any question.

⭐ Happy mapping! ⭐