Symbology That Survives the Print Room
A planning department produces a 36x48 inch map of a city's sidewalk network for a public hearing. The network layer has 240,000 segments classified into four condition ratings. The analyst builds the map in ArcGIS Pro with a single-symbol renderer and manually colors segments, then exports to PDF at 300 dpi. In review, the map is 1.4 GB, takes four minutes to open, and the sidewalk linework disappears when zoomed out below about 1:40,000. The hearing is tomorrow and the print vendor requires a PDF under 200 MB. The analyst proposes to lower export resolution to 96 dpi to hit the size target.
1. Diagnose why the linework vanishes at small scales and why the PDF is so large, and propose a fix that does not sacrifice print quality.
Read the worked answer
Both symptoms trace to the same root cause: the layer is drawn as manually colored individual features rather than as a symbolized, scale-aware layer. At small scales, dense single-symbol linework collapses into sub-pixel widths and visually disappears - a display/scale problem, not a data problem. The PDF is large because every one of 240,000 segments is emitted as its own vector object with its own style, which bloats the file and slows rendering.
The fix is to use a classified renderer on the condition field with four categories, which lets the software draw the layer as a symbolized set rather than 240,000 hand-styled objects. Add scale-dependent visibility or a minimum/maximum scale so the layer is generalized or suppressed at small scales instead of vanishing unpredictably. If the vendor needs a smaller file, the lever is reducing redundant vector complexity and using appropriate layer/PDF settings, not dropping to 96 dpi, which would degrade the very print quality the hearing depends on. Verify at both the overview scale and the detail scale before export.
2. The analyst wants to keep the manual colors because 'they match the legend.' How would you reconcile the requirement for consistent colors with the need for a maintainable, performant map?
Read the worked answer
Manual per-feature colors are not a legend strategy; they are an unmaintainable one. Any data update breaks the match, and no one can reproduce the classification later. The correct approach is to define the four condition categories explicitly in a classified renderer, then style each category once. The legend is then generated from the renderer and is guaranteed to match the map by construction.
If specific colors are mandated (branding, accessibility), set those colors on the four category symbols, not on individual features. This preserves the required palette while making the map reproducible, editable, and far lighter to render and export. Point out the trade-off honestly: manual styling gives the illusion of control but couples the map to a frozen snapshot of the data. A classified renderer with fixed category colors gives the same visual result with none of the fragility.