Best Practices

We have summarized the core principles and practical guidelines for effectively designing dashboards.

KPI placement

By placing key indicators (KPIs) as statistics cards at the top of your dashboard, users see the most important numbers first.

Recommended configuration

1. Top Area: Lay out 3-5 statistics cards horizontally to summarize key KPIs in one line.
2. Central area: Place trend/comparison charts such as line charts and bar charts.
3. Bottom area: Provide detailed data with data table widgets or place auxiliary charts

tip

On statistics cards, use prefixes/suffixes (₩, %, number) to clearly convey the meaning of the numbers. You can adjust the level of emphasis between cards with the value font size (choose from Small, Medium, or Large, or enter 1–120px manually).

Visualization Selection Guide

You should choose a chart based on the characteristics of the data and the message you want to convey.

Purpose	Recommended Widgets	Example
Comparison by Category	bar chart	Sales by department, sales volume by product
Trends over time	line chart	Number of daily visitors, monthly sales trend
Ratio to total	Pie/Donut Chart	Market share, category composition ratio
Relationship between two variables	Scatterplot	Advertising cost vs sales, temperature vs sales volume
Highlight key figures	Stats Card	Total Sales, Active Users
Detailed data inquiry	data table	Transaction History, User List
Explanation/guidance text	text box	Section Title, Data Interpretation Guide
external content	Embed external site	External monitoring tools, in-house wiki

Things to keep in mind when choosing a visualization

• Pie/Donut Chart is effective when there are 5 or fewer items. If you have a lot of items, use a bar chart.
• Line charts are suitable for data with a time axis. A bar chart is clearer for categorical data.
• Scatterplots show meaningful patterns when there are enough data points (at least 30).

Performance optimization

These are guidelines to keep your dashboard loading fast and responsive.

Query optimization

• Use aggregation: Instead of querying the original data as is, use GROUP BY and aggregate functions to reduce the number of result rows.
• LIMIT setting: When searching detailed data such as data table widgets, be sure to specify LIMIT to prevent excessive data return.
• Select only the required columns: Instead of SELECT *, explicitly specify only the required columns.

-- Bad: query all data
SELECT * FROM large_table

-- Good: aggregate and query only the required columns
SELECT category, SUM(amount) as total
FROM large_table
WHERE event_date >= '2026-06-01' AND event_date < '2026-06-08'
GROUP BY category
ORDER BY total DESC
LIMIT 20

Manage the number of widgets

• We recommend no more than 10 to 15 widgets per dashboard.
• More widgets mean more queries running simultaneously, resulting in longer loading times.
• Consider grouping related metrics by topic and separating them into separate dashboards.

Utilizing DRS

• Limiting the query scope using top DRS significantly improves query performance. If the widget does not have an explicit time column setting, DRS is applied automatically.
• Improve initial loading speed by setting the default DRS range to the last 7 days or the last 30 days.
• Locking DRS to a custom fixed range disables auto-refresh, ensuring data does not fluctuate during point-in-time comparison analysis.

caution

Viewing full time period data without a date filter may result in slow dashboard response due to processing large amounts of data. Be sure to use DRS or date conditions in your query.

Select auto-refresh cycle

The auto-refresh cycle (dropdown in the toolbar at the top of the preview screen or settings modal) is determined by taking into account how frequently the data is refreshed and the number of widgets. Choices are Disabled · 30 sec · 1 min · 5 min · 10 min · 30 min (or a custom second value).

• 30 seconds to 1 minute: Real-time operation monitoring.
• 5 minutes: General KPI dashboard.
• 10 to 30 minutes / Inactive: Analysis/reporting dashboard.

Colors and Layout

Color usage principles

• Consistent Palette: Ensure visual unity by maintaining the same color palette within one dashboard.
• Meaningful Colors: Use the same color in all charts for the same categories (e.g. specific departments, product lines).
• Use accent colors: Use high-contrast colors for key indicators or data that requires attention.
• Consideration of color blindness: Avoid red-green combinations, and choose color combinations with sufficient difference in brightness.

Layout principles

• Information Hierarchy: Place the most important information at the top left (Z-pattern reading direction).
• Logical Grouping: Place related widgets close together and separate sections with text boxes.
• Consistent Size: Widgets of the same type will have a neat impression if they are uniformly sized.
• Use Whitespace: Maintain appropriate spacing between widgets to reduce visual clutter.

tip

When first designing a dashboard, you can create a more effective layout by sketching out the widget placement on paper or a whiteboard and then implementing it.

Design Checklist

Before sharing your dashboard, check the following:

• Are the key KPIs clearly placed at the top?
• Is each chart the right type for the message you want to convey?
• Are meaningful titles set for all widgets?
• Is the upper DRS set to a reasonable default range?
• Does the query contain appropriate LIMIT and date conditions?
• Are the colors consistent and accessibility considered?
• Is it focused on key information without unnecessary widgets?