PlotSet LogoPlotSet
Tous les types de graphiques

Quadrant Chart Maker

Split a scatter plot into four labelled boxes and turn a cloud of points into a decision. Upload your data, set the dividing lines, and publish an interactive quadrant chart you can embed anywhere — free.

Qu’est-ce qu’un Quadrant Chart ?

A quadrant chart is a scatter plot with two dividing lines drawn through it and each of the resulting four regions given a name. The dots are still positioned by two measures, but the labels do the interpretive work: high-effort and high-impact becomes "do first", low on both becomes "drop". It is the chart behind priority matrices, risk grids and portfolio reviews, and its power and its danger are the same thing — those four labels turn continuous data into categories, which is enormously useful for making a decision and enormously easy to abuse.

Quand utiliser un Quadrant Chart

  • You want the chart to prompt an action

    Prioritisation, triage, portfolio review. Naming the quadrants tells the reader what to do about each group, which a bare scatter plot never does.

  • Two measures genuinely define the decision

    Impact against effort, growth against share, severity against likelihood. Both axes have to matter, or one sorted bar chart is the better chart.

  • The dividing lines have a real justification

    A budget threshold, a target, the median. When the boundary means something, the quadrants mean something.

  • You have a manageable number of items

    Enough points to see groupings, few enough that each can be labelled — roughly ten to fifty.

Quand éviter un Quadrant Chart

  • The boundaries are arbitrary

    Sliding the lines a little moves items between quadrants and changes the recommendation. If you cannot defend where the lines sit, do not draw them.

  • Most points sit near a line

    A dense cluster around the crossing point means the categorisation is noise. Show the scatter and describe the distribution instead.

  • One measure dominates the decision

    Then it is a ranking, and a sorted bar chart makes it without implying the second axis carried equal weight.

  • There are hundreds of points

    Unlabelled dots in a named box invite conclusions about groups nobody can identify. Aggregate first, or use a density plot.

Comment créer un Quadrant Chart

D’un tableur à un graphique publié, étape par étape.

01

Start from the quadrant template

Open the PlotSet quadrant chart template. It loads with sample data bound so you can see how the two measures and the labels map onto the grid.

02

Shape your data as label, x, y

One row per item: what it is called and its value on each of the two measures. An optional group column lets you colour the points as well.

03

Upload or connect your data

Import a CSV or Excel file, paste from a spreadsheet, or connect a Google Sheet so the chart can be refreshed later.

04

Set the dividing lines deliberately

Use a real threshold if one exists, and the median if not. Whichever you choose, state it on the chart — an unlabelled line is an unexamined assumption.

05

Publish, embed or export

Publish for a live interactive embed where hovering a point shows its values, or export a PNG, SVG or PDF.

Partez du modèle Quadrant Chart avec des données d’exemple déjà chargées, puis remplacez-les par les vôtres.

Ouvrir le modèle

Les éléments d’un Quadrant Chart

Point
One item, positioned by its two measures. Direct labels beat a legend here, because the reader needs to know which item is which to act.
Dividing lines
The vertical and horizontal cuts. They are the most consequential design decision on the chart and deserve to be labelled with their value.
Quadrant label
The name given to each region. This is what converts position into recommendation, so the wording carries real weight.
Axes
The two measures, each with units. Because the reading is relative, the axes need not start at zero — but they do need to be labelled.
Quadrant shading
Optional background tint per region. Keep it faint; a saturated background competes with the points it exists to group.
Point size or colour
An optional third and fourth variable — budget, team, risk. Useful, but every extra encoding costs the clarity that justified the quadrants.

Erreurs courantes avec un Quadrant Chart

Hiding where the lines came from

A quadrant chart without stated thresholds asks the reader to trust an invisible judgement. Print the values on the lines.

Treating quadrant membership as a fact

An item just over a line is nearly identical to one just under it. Say that out loud when items near the boundary are being treated differently.

Loaded quadrant labels

Calling a region "dogs" or "dead weight" makes the chart an argument. Neutral, descriptive names keep it a piece of analysis.

Unlabelled points

A grid full of anonymous dots supports no decision at all. If there are too many to label, there are too many for this chart.

Quadrant Chart : questions fréquentes

Plotting items against two measures and dividing the result into four named regions, so that position implies a recommended action — prioritisation, risk grids and portfolio reviews.

Créez votre Quadrant Chart maintenant

Gratuit pour commencer, avec plus de 300 modèles. Sans carte bancaire.