Skip to main content

AnyDB Chart Cells

Chart cells allow you to visualize numeric data directly inside a record. They are designed for operational use cases where trends, comparisons, and breakdowns need to live next to the underlying data.

AnyDB currently supports:

  • Line Charts
  • Bar Charts
  • Pie Charts

All chart types are configured by referencing other cells that contain array data.

Charts

Common Chart Fields​

These fields appear across all chart types.

FieldDescription
FormatMust be set to Chart to render a chart in the cell.
Chart TypeDetermines how data is visualized: Line, Bar, or Pie.
Chart TitleTitle displayed at the top of the chart.
Series NameLabel shown in the chart legend for a data series.
Series ValuesReference to a cell containing a numeric array.
Add SeriesAdds an additional data series to the chart (Line and Bar only).

Axis-Based Chart Fields (Line and Bar)​

These fields apply to charts that use an X and Y axis.

FieldDescription
X Axis LabelsReference to a cell containing the category or time labels (array).
X Axis TitleLabel shown under the horizontal axis.
Y Axis TitleLabel shown beside the vertical axis.
SmoothedEnables or disables curve smoothing for line charts.
OrientationBar chart only. Controls vertical or horizontal bars.

Pie Chart Fields​

These fields are specific to pie charts.

FieldDescription
Slice LabelsReference to a cell containing slice names (array).
Slice ValuesReference to a cell containing numeric values (array).

Line Chart​

Line charts show trends over time or ordered sequences. They are commonly used for revenue, forecasts, SLAs, and throughput metrics.

Example use case​

Monthly Revenue Growth with Actual vs Forecast.

Line Chart Example

Data setup​

X Axis Labels

["Jan","Feb","Mar","Apr","May","Jun"]

Series 1: Revenue ($K)

[120,135,150,165,180,205]

Series 2: Forecast ($K)

[118,132,148,170,190,215]

Chart configuration​

  • Chart Type: Line Chart
  • Chart Title: Monthly Revenue Growth
  • X Axis Title: Months
  • Y Axis Title: $K
  • X Axis Labels: ["Jan","Feb","Mar","Apr","May","Jun"]
  • Series Values:
    • Revenue: [120,135,150,165,180,205]
    • Forecast: [118,132,148,170,190,215]

Notes​

  • Multiple series are supported.
  • All series must align with the same X Axis Labels.
  • Best used for trends and targets.

Bar Chart​

Bar charts compare values across categories such as vendors, locations, or departments.

Example use case​

Vendor Spend by Quarter. Bar Chart Example

Data setup​

X Axis Labels

["Q1","Q2","Q3","Q4"]

Series 1: Vendor A

[220,240,260,275]

Series 2: Vendor B

[180,195,210,230]

Series 3: Vendor C

[95,110,120,135]

Chart configuration​

  • Chart Type: Bar Chart
  • Chart Title: Vendor Spend by Quarter
  • X Axis Title: Quarters
  • Y Axis Title: Spend
  • Orientation: Vertical Bars
  • X Axis Labels: ["Q1","Q2","Q3","Q4"]
  • Series Values:
    • Series Name : "Vendor A"

    • Series Values: [220,240,260,275]

    • Series Name : "Vendor B"

    • Series Values: [180,195,210,230]

    • Series Name : "Vendor C"

    • Series Values: [95,110,120,135]

Notes​

  • Multiple series render side by side.
  • Best for comparisons, not time-based trends.

Pie Chart​

Pie charts show proportional breakdowns of a whole.

Example use case​

Operational Cost Breakdown.

Pie Chart Example

Data setup​

Slice Labels

["Labor","Materials","Logistics","Overhead"]

Slice Values

[42,28,18,12]

Chart configuration​

  • Chart Type: Pie Chart
  • Chart Title: Operational Cost Breakdown
  • Slice Labels: ["Labor","Materials","Logistics","Overhead"]
  • Slice Values: [42,28,18,12]

Notes​

  • Pie charts support a single series only.
  • Keep the number of slices limited for readability.

General requirements​

  • Source cells must contain arrays.
  • Label arrays and value arrays must have matching lengths.
  • Charts update automatically when referenced values change.
  • Charts are intended for lightweight, inline visualization tied to operational records.

  • Line charts for trends and targets.
  • Bar charts for comparisons across entities.
  • Pie charts for cost, issue, or resource breakdowns.

Charts are most effective when placed directly on the record they describe or using data from the attached children records.