Here's the honest bit. Making a map is easy; making a map that tells the truth takes a little more care. Fortunately, beginners tend to make the same small set of mistakes, so once you know them, you can sidestep almost all of them.

None of these are about being clever. They're about not accidentally misleading your reader (or yourself).

Mistake 1: Mapping raw counts when you need rates

This is the big one. Colouring areas by a raw total usually just shows you where more people live.

• Bad: total number of crimes by district.
• Better: crimes per 1,000 residents.

The fix is normalisation: turn the total into a rate, percentage, density, or value per person before you map it. We covered this in GIS Mapping; go back and re-read it if you're not completely convinced.

Mistake 2: Using a choropleth for everything

The shaded-area map is so satisfying that beginners reach for it constantly. But choropleths are really only right for rates, percentages, and normalised values by area. They're a poor choice for exact locations (use a point map) or raw totals (use proportional symbols, or normalise first).

Mistake 3: Ignoring projection and CRS

A map can look perfectly fine and still measure wrong. If you calculate distances or areas in raw latitude/longitude, the numbers can be well off. When measurement matters, reproject to an appropriate local projected coordinate system first. (See the CRS section in GIS Data.)

Mistake 4: Making the map too busy

More layers, more labels, more colours: it's tempting to show everything at once. The result is a map nobody can read.

If everything is emphasised, nothing is.

Decide what the map is for, make that one thing obvious, and turn everything else down or off.

Mistake 5: Using confusing colours

Colour choices quietly make or break a map:

• Red-green schemes are invisible to many colour-blind readers.
• Ramps that don't follow a logical order (light to dark) confuse everyone.
• Too many classes turn a map into a guessing game.
• Poor contrast makes labels and features hard to pick out.

Aim for ordered, colour-blind-safe ramps, five to seven classes at most, and never let colour be the only thing carrying meaning.

Mistake 6: Trusting data without checking it

Data is not automatically correct just because someone published it. Before you build on a dataset, check its source, date, completeness, accuracy, and licence. Five minutes of checking now beats a week of explaining a wrong map later; the full checklist lives in GIS Data.

Mistake 7: Publishing huge datasets without optimising

Uploading a giant, ultra-detailed dataset straight to a web map is a classic way to make it crawl. Before publishing, think about simplifying geometry, clustering points, tiling large data, and filtering down to what the map actually needs.

Mistake 8: Forgetting the end user

A map for the public, a manager, a field worker, and a GIS analyst are not the same map. They need different defaults, labels, popups, and levels of detail. Before you publish, ask: who is this for, and what do they need to do with it? Design backwards from that answer.

---

Steer clear of these eight and you'll already be making maps that are clearer and more honest than most. Now for the fun part: let's actually build one.