Excel Charts and Dashboards with XLFill

Charts, sparklines, and conditional formatting turn a wall of numbers into a story. With XLFill, you add all three from the template — no Go code for chart objects, no series definitions, no color hex codes. Design the layout in Excel, drop in a jx:chart command, and the data does the rest.

jx:chart — 8 chart types, one command

The jx:chart command embeds a native Excel chart in your output. XLFill automatically sizes the data ranges based on actual output rows.

Supported chart types

Type	Keyword	Best for
Bar (horizontal)	`bar`	Comparing categories
Column (vertical)	`col`	Comparing categories (when labels are short)
Line	`line`	Time series, trends
Pie	`pie`	Composition, proportions
Doughnut	`doughnut`	Composition with a center label
Area	`area`	Cumulative totals over time
Scatter	`scatter`	Correlation between two variables
Radar	`radar`	Multi-axis comparison (skill maps, balanced scorecards)

Bar chart example

Template layout:

A1: Region          B1: Revenue
A2: ${r.Name}       B2: ${r.Revenue}

Cell A1 comment:
  jx:area(lastCell="H20")
  jx:each(items="regions" var="r" lastCell="B2")

Cell D1 comment:
  jx:chart(lastCell="H15" type="bar" title="Revenue by Region"
           catRange="A2:A2" valRange="B2:B2")

Go code:

data := map[string]any{
    "regions": []map[string]any{
        {"Name": "North America", "Revenue": 1200000},
        {"Name": "Europe", "Revenue": 980000},
        {"Name": "Asia Pacific", "Revenue": 750000},
        {"Name": "Latin America", "Revenue": 420000},
    },
}
xlfill.Fill("template.xlsx", "dashboard.xlsx", data)

XLFill expands the data rows, then adjusts the chart’s catRange and valRange to cover all output rows. If you have 4 regions, the chart references A2:A5 and B2:B5 automatically.

Pro tip: The catRange and valRange use template-relative coordinates (just one row: A2:A2). XLFill expands them to match the actual output. You never hard-code row counts.

Multi-series line chart

Show revenue and costs over time in one chart:

A1: Month    B1: Revenue    C1: Costs

Cell D1 comment:
  jx:chart(lastCell="H18" type="line" title="Revenue vs Costs"
           catRange="A2:A2" valRange="B2:B2,C2:C2"
           seriesNames="Revenue,Costs" legendPosition="bottom")

Multiple value ranges are comma-separated. Each gets its own series with the name from seriesNames.

Pie chart

Cell D1 comment:
  jx:chart(lastCell="H12" type="pie" title="Market Share"
           catRange="A2:A2" valRange="B2:B2")

Pie charts use categories as slice labels and values as slice sizes. Simple and effective for composition data.

Chart styling options

Attribute	Values	Effect
`style`	1-48	Matches Excel’s built-in chart styles
`legendPosition`	`bottom`, `left`, `right`, `top`, `none`	Where the legend appears
`width`	EMUs or cell count	Chart width
`height`	EMUs or cell count	Chart height

jx:chart(lastCell="H15" type="col" title="Sales by Product"
         catRange="A2:A2" valRange="B2:B2"
         style="26" legendPosition="none")

Did you know? Excel’s chart style 26 gives you a clean, modern look with subtle gradients. Try styles in the 20-30 range for professional reports.

jx:sparkline — inline trends

Sparklines are tiny charts that fit inside a single cell. They’re perfect for showing trends without taking up chart real estate.

Cell D1 comment:
  jx:sparkline(lastCell="D1" type="line" dataRange="B2:G2")

Sparkline types

Type	Keyword	Best for
Line	`line`	Trends over time
Column	`column`	Comparing discrete values
Win/Loss	`winLoss`	Positive/negative outcomes

Sparkline in a loop

Add a trend sparkline to each row in a report:

A1: Product    B1: Jan   C1: Feb   D1: Mar   E1: Trend
A2: ${p.Name}  B2: ${p.Jan}  C2: ${p.Feb}  D2: ${p.Mar}

Cell E2 comment:
  jx:sparkline(lastCell="E2" type="line" dataRange="B2:D2")

Cell A1 comment:
  jx:area(lastCell="E2")
  jx:each(items="products" var="p" lastCell="E2")

Each product row gets its own sparkline showing the 3-month trend. The sparkline data range is relative to each output row.

Sparklines vs. charts

	Sparkline	Chart
Size	One cell	Multi-cell area
Interactivity	None	Hover, click, resize
Best for	Row-level trends	Report-level summaries
Types	3 (line, column, win/loss)	8 types
Multiple series	No	Yes

Use sparklines for per-row context, charts for overall picture. They complement each other beautifully.

jx:conditionalFormat — data bars, color scales, icon sets

Conditional formatting makes patterns jump off the page without a single chart.

Data bars

Turn cells into inline bar charts:

Cell B2 comment:
  jx:conditionalFormat(lastCell="B2" type="dataBar"
                       minColor="E6F2FF" maxColor="2F5496")

Every cell in column B gets a proportional bar fill. Higher values = longer bars. The min/max colors define the gradient.

Color scales

Color-code a range from low to high:

Cell C2 comment:
  jx:conditionalFormat(lastCell="C2" type="colorScale"
                       minColor="FF0000" midColor="FFFF00" maxColor="00FF00")

Red → yellow → green. Classic heat map. Great for performance metrics, test scores, or financial variances.

Icon sets

Add visual indicators (arrows, traffic lights, stars):

Cell D2 comment:
  jx:conditionalFormat(lastCell="D2" type="iconSet"
                       iconStyle="3Arrows")

Available icon styles: 3Arrows, 3TrafficLights, 3Stars, 4Arrows, 5Arrows, 3Symbols, and more.

Conditional formatting inside loops

When used inside a jx:each area, conditional formatting applies to every output row:

Cell A1 comment:
  jx:area(lastCell="D2")
  jx:each(items="employees" var="e" lastCell="D2")

Cell C2 comment:
  jx:conditionalFormat(lastCell="C2" type="colorScale"
                       minColor="FFC7CE" maxColor="C6EFCE")

Every employee’s salary cell gets the color scale. The formatting range expands with the data.

Building a complete dashboard

Combine all three — chart, sparklines, and conditional formatting — for a data-rich dashboard:

Template layout

A1: Department    B1: Q1     C1: Q2     D1: Q3     E1: Q4     F1: Total    G1: Trend
A2: ${d.Name}     B2: ${d.Q1} C2: ${d.Q2} D2: ${d.Q3} E2: ${d.Q4} F2: =SUM(B2:E2)

Cell A1 comment:
  jx:area(lastCell="K25")
  jx:each(items="departments" var="d" lastCell="G2")

Cell G2 comment:
  jx:sparkline(lastCell="G2" type="line" dataRange="B2:E2")

Cell F2 comment:
  jx:conditionalFormat(lastCell="F2" type="dataBar"
                       minColor="E6F2FF" maxColor="2F5496")

Cell I1 comment:
  jx:chart(lastCell="K15" type="col" title="Annual Revenue by Department"
           catRange="A2:A2" valRange="F2:F2" style="26")

Cell I16 comment:
  jx:chart(lastCell="K25" type="line" title="Quarterly Trends"
           catRange="A2:A2" valRange="B2:B2,C2:C2,D2:D2,E2:E2"
           seriesNames="Q1,Q2,Q3,Q4" legendPosition="bottom")

Go code

data := map[string]any{
    "departments": []map[string]any{
        {"Name": "Engineering", "Q1": 420000, "Q2": 485000, "Q3": 510000, "Q4": 530000},
        {"Name": "Sales",       "Q1": 380000, "Q2": 410000, "Q3": 450000, "Q4": 520000},
        {"Name": "Marketing",   "Q1": 180000, "Q2": 195000, "Q3": 210000, "Q4": 230000},
        {"Name": "Operations",  "Q1": 290000, "Q2": 300000, "Q3": 310000, "Q4": 325000},
        {"Name": "HR",          "Q1": 120000, "Q2": 125000, "Q3": 130000, "Q4": 135000},
    },
}

xlfill.Fill("dashboard_template.xlsx", "dashboard.xlsx", data)

The output has:

Data table with quarterly figures and formula-driven totals
Sparklines in column G showing each department’s quarterly trend
Data bars in column F showing relative total revenue
A column chart summarizing annual revenue
A line chart showing quarterly trends across departments

All from a single template. Zero chart code in Go.

Tips and tricks

Chart sizing: Place the chart’s lastCell far enough from the data to avoid overlap. A good rule of thumb: leave 3-4 empty columns between data and chart.
Multi-series charts: For charts with 4+ series, use legendPosition="bottom" — it keeps the chart area taller and more readable.
Choosing the right chart type:
- Comparing items? Use bar (horizontal) or col (vertical)
- Showing trends? Use line or area
- Showing composition? Use pie or doughnut
- Showing correlation? Use scatter
Sparklines in summary rows: You can use sparklines outside of loops too — for example, in a totals row that summarizes data from a different range.
Conditional formatting thresholds: For icon sets, the thresholds are based on percentiles by default. The top third gets the green arrow, middle gets yellow, bottom gets red.
Combine formatting types: Nothing stops you from putting both a color scale and data bars on the same range. Excel will layer them, though it can get visually busy.

Pro tip: Design your dashboard template in Excel first with sample static data. Get the chart sizes and positions right visually, then replace the static data with ${...} expressions. This is much faster than guessing at EMU sizes.

What’s next?

Create data entry forms with dropdowns: Data Validation and Forms →
Build financial reports with grouping and aggregation: Financial Reports →
See all chart command attributes: jx:chart Reference →