data-visualization

star 0

Build accessible, performant data visualizations with Recharts, Chart.js, and D3.js. Chart selection, responsive design, colorblind-safe palettes.

medy-gribkov By medy-gribkov schedule Updated 3/3/2026
play_arrow Run Skill in Manus View GitHub

name: data-visualization description: Build accessible, performant data visualizations with Recharts, Chart.js, and D3.js. Chart selection, responsive design, colorblind-safe palettes. user-invokable: true

Data Visualization

Build accessible, performant data visualizations using modern libraries. Choose the right chart type, implement responsive layouts, and optimize for large datasets.

Chart Type Selection

Choose charts based on data relationships, not aesthetics.

BAD - Wrong chart for the task:

// Pie chart for comparing 12 categories - hard to compare angles
<PieChart width={400} height={400}>
  <Pie data={monthlyData} dataKey="value" nameKey="month" />
</PieChart>

// Line chart for categorical data with no time relationship
<LineChart data={productCategories}>
  <Line dataKey="sales" />
</LineChart>

GOOD - Chart matches data structure:

// Bar chart for category comparison - easy to compare lengths
<BarChart width={600} height={400} data={monthlyData}>
  <CartesianGrid strokeDasharray="3 3" />
  <XAxis dataKey="month" />
  <YAxis />
  <Tooltip />
  <Bar dataKey="value" fill="#d4943a" />
</BarChart>

// Scatter plot for correlation analysis
<ScatterChart width={600} height={400}>
  <CartesianGrid />
  <XAxis dataKey="age" name="Age" />
  <YAxis dataKey="salary" name="Salary" />
  <Scatter data={employees} fill="#d4943a" />
</ScatterChart>

Chart selection guide:

  • Bar/Column: Compare categories, rankings, discrete values
  • Line: Time series, trends over continuous periods
  • Scatter: Correlation between two variables, clustering
  • Heatmap: Patterns in 2D categorical data (day/hour traffic)
  • Area: Cumulative values over time, part-to-whole relationships
  • Avoid pie charts: Use bar charts instead (easier comparison)

Recharts Implementation

Recharts provides React-native declarative charts with built-in responsiveness.

BAD - Fixed dimensions, no accessibility:

function SalesChart({ data }) {
  return (
    <LineChart width={800} height={300} data={data}>
      <Line dataKey="sales" stroke="red" />
      <Line dataKey="profit" stroke="green" />
    </LineChart>
  );
}

GOOD - Responsive, accessible, properly labeled:

import { LineChart, Line, XAxis, YAxis, CartesianGrid, Tooltip, Legend, ResponsiveContainer } from 'recharts';

interface DataPoint {
  date: string;
  sales: number;
  profit: number;
}

interface SalesChartProps {
  data: DataPoint[];
}

const SalesChart: React.FC<SalesChartProps> = ({ data }) => {
  return (
    <ResponsiveContainer width="100%" height={400}>
      <LineChart
        data={data}
        margin={{ top: 5, right: 30, left: 20, bottom: 5 }}
        aria-label="Sales and profit trend over time"
      >
        <CartesianGrid strokeDasharray="3 3" stroke="#333" />
        <XAxis
          dataKey="date"
          stroke="#666"
          tick={{ fill: '#666' }}
        />
        <YAxis
          stroke="#666"
          tick={{ fill: '#666' }}
          label={{ value: 'USD', angle: -90, position: 'insideLeft' }}
        />
        <Tooltip
          contentStyle={{ backgroundColor: '#1a1a1a', border: '1px solid #333' }}
          formatter={(value: number) => `$${value.toLocaleString()}`}
        />
        <Legend />
        <Line
          type="monotone"
          dataKey="sales"
          stroke="#3b82f6"
          strokeWidth={2}
          dot={{ fill: '#3b82f6', r: 4 }}
          activeDot={{ r: 6 }}
          name="Sales"
        />
        <Line
          type="monotone"
          dataKey="profit"
          stroke="#d4943a"
          strokeWidth={2}
          dot={{ fill: '#d4943a', r: 4 }}
          name="Profit"
        />
      </LineChart>
    </ResponsiveContainer>
  );
};

Colorblind-Safe Palettes

Rainbow colors fail for 8% of men. Use palettes designed for accessibility.

BAD - Rainbow spectrum, red/green for critical info:

const COLORS = ['#ff0000', '#00ff00', '#0000ff', '#ffff00', '#ff00ff'];

// Red/green for positive/negative - invisible to deuteranopes
<Bar dataKey="change" fill={value > 0 ? '#00ff00' : '#ff0000'} />

GOOD - Colorblind-safe palette with patterns:

// Paul Tol's colorblind-safe palette
const COLOR_PALETTE = {
  blue: '#4477AA',
  cyan: '#66CCEE',
  green: '#228833',
  yellow: '#CCBB44',
  red: '#EE6677',
  purple: '#AA3377',
  grey: '#BBBBBB',
};

// Use blue/orange for diverging data (safe for all types of colorblindness)
const getDivergingColor = (value: number): string => {
  return value > 0 ? COLOR_PALETTE.blue : COLOR_PALETTE.red;
};

// Add patterns for critical distinctions
<Bar dataKey="change" fill={getDivergingColor(value)}>
  {value < 0 && <pattern id="stripes" patternUnits="userSpaceOnUse" width="4" height="4">
    <path d="M-1,1 l2,-2 M0,4 l4,-4 M3,5 l2,-2" stroke="#000" strokeWidth="1" />
  </pattern>}
</Bar>

Recommended palettes:

  • Categorical (up to 7): Paul Tol's bright scheme
  • Sequential: Single hue progression (light blue → dark blue)
  • Diverging: Blue → white → orange (not red → green)

Performance with Large Datasets

Canvas outperforms SVG for 1000+ points. Use data windowing and memoization.

BAD - Re-render entire chart on every update:

function LiveChart() {
  const [data, setData] = useState<DataPoint[]>([]);

  useEffect(() => {
    const interval = setInterval(() => {
      setData(prev => [...prev, fetchNewPoint()]); // Unbounded growth
    }, 1000);
    return () => clearInterval(interval);
  }, []);

  return <LineChart data={data}><Line dataKey="value" /></LineChart>;
}

GOOD - Windowed data, memoized component, canvas rendering:

import { useMemo, useCallback } from 'react';
import { Line } from 'react-chartjs-2';
import { Chart as ChartJS, CategoryScale, LinearScale, PointElement, LineElement } from 'chart.js';

ChartJS.register(CategoryScale, LinearScale, PointElement, LineElement);

interface LiveChartProps {
  maxPoints?: number;
}

const LiveChart: React.FC<LiveChartProps> = ({ maxPoints = 100 }) => {
  const [data, setData] = useState<DataPoint[]>([]);

  // Keep only last N points
  const addDataPoint = useCallback((point: DataPoint) => {
    setData(prev => {
      const updated = [...prev, point];
      return updated.slice(-maxPoints);
    });
  }, [maxPoints]);

  useEffect(() => {
    const interval = setInterval(() => {
      addDataPoint(fetchNewPoint());
    }, 1000);
    return () => clearInterval(interval);
  }, [addDataPoint]);

  // Memoize chart config to prevent re-creation
  const chartData = useMemo(() => ({
    labels: data.map(d => d.timestamp),
    datasets: [{
      label: 'Live Value',
      data: data.map(d => d.value),
      borderColor: '#d4943a',
      borderWidth: 2,
      pointRadius: 0, // No dots for performance
    }],
  }), [data]);

  const options = useMemo(() => ({
    responsive: true,
    maintainAspectRatio: false,
    animation: false, // Disable animations for real-time
    plugins: {
      legend: { display: false },
    },
    scales: {
      x: { display: false }, // Hide labels for performance
      y: { beginAtZero: true },
    },
  }), []);

  return (
    <div style={{ height: '400px' }}>
      <Line data={chartData} options={options} />
    </div>
  );
};

Performance tips:

  • Canvas vs SVG: Use Chart.js (canvas) for 1000+ points, Recharts (SVG) for <1000
  • Data windowing: Keep only visible data in state
  • Disable animations: Set animation: false for real-time updates
  • Remove point dots: pointRadius: 0 saves render time
  • Debounce updates: Batch rapid updates with requestAnimationFrame

Dashboard Layout Patterns

Grid-based responsive layouts with semantic sizing.

BAD - Fixed pixel grid, no hierarchy:

<div style={{ display: 'flex' }}>
  <div style={{ width: '400px' }}><Chart1 /></div>
  <div style={{ width: '400px' }}><Chart2 /></div>
  <div style={{ width: '400px' }}><Chart3 /></div>
</div>

GOOD - Responsive grid with visual hierarchy:

import { ResponsiveContainer } from 'recharts';

const Dashboard: React.FC = () => {
  return (
    <div className="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-3 gap-6 p-6">
      {/* Primary metric - full width */}
      <div className="col-span-1 md:col-span-2 lg:col-span-3 bg-gray-900 p-6 rounded-lg">
        <h2 className="text-xl font-semibold mb-4">Revenue Trend</h2>
        <ResponsiveContainer width="100%" height={300}>
          <LineChart data={revenueData}>
            {/* ... */}
          </LineChart>
        </ResponsiveContainer>
      </div>

      {/* Secondary metrics - 2 columns on desktop */}
      <div className="col-span-1 md:col-span-1 bg-gray-900 p-6 rounded-lg">
        <h3 className="text-lg font-medium mb-4">User Growth</h3>
        <ResponsiveContainer width="100%" height={250}>
          <AreaChart data={userGrowth}>
            {/* ... */}
          </AreaChart>
        </ResponsiveContainer>
      </div>

      <div className="col-span-1 md:col-span-1 bg-gray-900 p-6 rounded-lg">
        <h3 className="text-lg font-medium mb-4">Conversion Rate</h3>
        <ResponsiveContainer width="100%" height={250}>
          <BarChart data={conversionData}>
            {/* ... */}
          </BarChart>
        </ResponsiveContainer>
      </div>

      {/* Tertiary - heatmap spans 2 columns */}
      <div className="col-span-1 md:col-span-2 lg:col-span-2 bg-gray-900 p-6 rounded-lg">
        <h3 className="text-lg font-medium mb-4">Activity Heatmap</h3>
        <ResponsiveContainer width="100%" height={200}>
          <ScatterChart>
            {/* ... */}
          </ScatterChart>
        </ResponsiveContainer>
      </div>
    </div>
  );
};

D3.js for Custom Visualizations

Use D3 for specialized charts not available in Recharts/Chart.js.

BAD - Manipulate DOM directly in React:

function CustomChart() {
  useEffect(() => {
    d3.select('#chart').append('svg'); // React loses control
  }, []);

  return <div id="chart"></div>;
}

GOOD - Use refs, let React manage DOM:

import { useRef, useEffect } from 'react';
import * as d3 from 'd3';

interface Node {
  id: string;
  value: number;
}

interface Link {
  source: string;
  target: string;
}

interface NetworkGraphProps {
  nodes: Node[];
  links: Link[];
}

const NetworkGraph: React.FC<NetworkGraphProps> = ({ nodes, links }) => {
  const svgRef = useRef<SVGSVGElement>(null);

  useEffect(() => {
    if (!svgRef.current) return;

    const svg = d3.select(svgRef.current);
    const width = 600;
    const height = 400;

    // Clear previous render
    svg.selectAll('*').remove();

    const simulation = d3.forceSimulation(nodes)
      .force('link', d3.forceLink(links).id((d: any) => d.id))
      .force('charge', d3.forceManyBody().strength(-200))
      .force('center', d3.forceCenter(width / 2, height / 2));

    const link = svg.append('g')
      .selectAll('line')
      .data(links)
      .join('line')
      .attr('stroke', '#999')
      .attr('stroke-width', 2);

    const node = svg.append('g')
      .selectAll('circle')
      .data(nodes)
      .join('circle')
      .attr('r', (d) => Math.sqrt(d.value) * 5)
      .attr('fill', '#d4943a')
      .attr('aria-label', (d) => `Node ${d.id}, value ${d.value}`);

    simulation.on('tick', () => {
      link
        .attr('x1', (d: any) => d.source.x)
        .attr('y1', (d: any) => d.source.y)
        .attr('x2', (d: any) => d.target.x)
        .attr('y2', (d: any) => d.target.y);

      node
        .attr('cx', (d: any) => d.x)
        .attr('cy', (d: any) => d.y);
    });

    return () => {
      simulation.stop();
    };
  }, [nodes, links]);

  return (
    <svg
      ref={svgRef}
      width={600}
      height={400}
      role="img"
      aria-label="Network graph visualization"
    />
  );
};

Accessibility

Screen readers need context. Provide labels, summaries, and data tables.

BAD - No semantic info, colors only:

<BarChart data={data}>
  <Bar dataKey="value" fill="green" />
</BarChart>

GOOD - ARIA labels, semantic HTML, text alternatives:

import { BarChart, Bar, XAxis, YAxis } from 'recharts';

const AccessibleChart: React.FC<{ data: DataPoint[] }> = ({ data }) => {
  const maxValue = Math.max(...data.map(d => d.value));
  const minValue = Math.min(...data.map(d => d.value));

  return (
    <figure>
      <figcaption className="sr-only">
        Sales by quarter: Q1 ${data[0].value}, Q2 ${data[1].value}, Q3 ${data[2].value}, Q4 ${data[3].value}.
        Highest quarter: {data.find(d => d.value === maxValue)?.quarter}.
      </figcaption>

      <ResponsiveContainer width="100%" height={400}>
        <BarChart
          data={data}
          role="img"
          aria-label="Quarterly sales bar chart"
        >
          <XAxis
            dataKey="quarter"
            label={{ value: 'Quarter', position: 'insideBottom', offset: -5 }}
          />
          <YAxis
            label={{ value: 'Sales (USD)', angle: -90, position: 'insideLeft' }}
          />
          <Tooltip
            formatter={(value: number) => [`$${value.toLocaleString()}`, 'Sales']}
          />
          <Bar
            dataKey="value"
            fill="#d4943a"
            aria-label="Sales values"
          />
        </BarChart>
      </ResponsiveContainer>

      {/* Provide data table for screen readers */}
      <details className="mt-4">
        <summary className="cursor-pointer text-sm text-gray-400">View data table</summary>
        <table className="mt-2 text-sm">
          <thead>
            <tr>
              <th className="text-left pr-4">Quarter</th>
              <th className="text-right">Sales</th>
            </tr>
          </thead>
          <tbody>
            {data.map(d => (
              <tr key={d.quarter}>
                <td className="pr-4">{d.quarter}</td>
                <td className="text-right">${d.value.toLocaleString()}</td>
              </tr>
            ))}
          </tbody>
        </table>
      </details>
    </figure>
  );
};

Accessibility checklist:

  • Add role="img" and aria-label to charts
  • Use <figure> and <figcaption> for semantic structure
  • Provide text summary in .sr-only class
  • Include data table alternative
  • Label axes and provide units
  • Use patterns or textures in addition to color
  • Ensure 4.5:1 contrast ratio for text on charts
Install via CLI
npx skills add https://github.com/medy-gribkov/arcana --skill data-visualization
Repository Details
star Stars 0
call_split Forks 1
navigation Branch main
article Path SKILL.md
More from Creator
medy-gribkov
medy-gribkov Explore all skills →