Case studies / 05

Dashboard styleguides: design & practices

#designsystems #dataviz #guidelines

Client
adidas — Digital Analytics
Scope
Principles, data handling, page types, components, charts, colours
Tools
Zeplin · InVision · Tableau
Dashboard Styleguides — Design & Practices

# Why

After the NPS dashboard, more and more dashboards were being built by different teams — each with its own layout, colours and way of filtering.

The styleguide harmonised them: a living document with guiding principles, best practices for handling and filtering data, page templates, a component library, chart templates and colour palettes, so any team could build a dashboard that looks and works like the others.

01Simplify

Show only what is needed; simplify numbers and views as much as possible.

02Flat

A flat, clean visual language that puts the data first.

03Intent

Every page, tile and chart serves a clear user intent.

# Data handling

RuleExampleNote
Decimals only when needed32.83475 → 33Some KPIs need decimals: 32.83475 → 32.8
Simplify big numbers with units12,345,354 → 12M21,120 → 21K
Units: millions, thousands, %, revenue32.8M · 32.8K · 32.8% · 32.8€Same font size, no blank space
ValueBlack #000000 · 22pt
Positive changeField green #009639
Negative changeWarm red #F9423A
LabelsDark grey #888888 · 12pt / 10pt in graphs

# Filtering

Filtering

A selection of the available data, to focus on specific groups or segments — only the amount of data changes.

Grouping

Arranges the selection by a dimension of the data, creating groups. Filtering then restricts the result even more.

Do / don’t

Always filter first, then group. Double grouping creates too much visual complexity and no clear connection between the dimensions.

# Basic structure

HeaderBranding, dashboard name, main navigation
Sub-headerHome and guide links, general filters, 2nd-level navigation
ContentTiles, graphs, dedicated filters, stories
FiltersData dimensions & segmentation
FooterData updates, feedback
Basic dashboard structure

Three kinds of filtering: normal filtering in the side panel, filtering on details directly from a chart, and filtering through stories.

Normal filtering, on-details filtering, stories filtering

UI colours

Brand · logoDark teal #37A99B
Business unitTeal #04D5C4
Header, active itemBlack #000000
Inactive itemDark grey #888888
Global filters barGrey #D2D2D2
Context filter barLight grey #EFEFEF

# Page types

Three families — landing, internal and guiding pages — each designed for a level of detail and a type of user.

Landing, internal and guiding page
Landing page · internal page · guiding page

# Components

Tiles

A tile is the KPI container: it shows how a KPI is performing and how it changed over the selected time frame. Depending on the user and topic, it can be detailed or really simplified.

Normal

Low density of KPIs — leadership users.

Small

Medium density — PM and PO users.

Minimized

High density — analysts.

Library

# Charts & colours

Chart templates ready to reuse, and colour palettes with fixed meanings — category colours, two themes, and dedicated colours for business units, markets, channels, genders, age groups, devices and more — so the same dimension always looks the same across dashboards.

# In use

Example dashboards built on the guidelines, with components shared in Zeplin and prototypes in InVision.

← PreviousInnovation capabilities in real estate