Creating Dashboard Visualizations

Elisabeth Engel

0%

99%

Easy, RiGHt?

Raw Big Data

Which

data?

Which representation?

Icon made by freepik from www.flaticon.com

time

$$$

Chart Sketch

Variant B

Variant A

For a/B Testing

time

$$$

Chart ResuLT

Variant A

Variant B

After A/B Testing

Clearly Variant B

Readability

IS NOT about the Font size

?

What's the better variant?

Dimensions

Encoding

Interaction

Integrity

Icon made by freepik from www.flaticon.com

Dimensions

Key InfluencErs

Time

Location

Sucess / Failure

Type

Target Group

EnCoding

Spot Information at a glance

ColoRs

Semantics

Color blindness

ColoR scHeme

colorbrewer2.org

Get Inspiration

Pinterest informationisbeautiful.net • behance.net • dailyinfographics.com

coolinfographics.com • visualcomplexity.com • flowingdata.com

improving-visualisations.org/visuals

Interaction

Zoom

Drag

Filter

Sort

Link

Highlight

Dashboard

10 Second Rule

Image taken from http://www.dashboardinsight.com/CMS/3f7edc6b-609f-4b54-9e81-8ac7629a2798/Healthcare-performance-dashboard.png

Dashboard

So what...

?

10 Second Rule

Dashboard

Sketch

25

Orders

99

Fails

25

15

70 k

30

k

Calls

Sales / Week

Revenue

Dashboard

Action!

Sketch

Get STARTED

LET OTHERS INVENT THE WHEEL

blackrockdigital.github.io/startbootstrap-sb-admin-2

FRAMEWorkS?

 

What do we need?

 

So test therefore, who join forever

 

Who will use it?

 

 

What do we want to spend?

 

JS FRAMEWorks

ENGINEERING FLOWCHART

Special requirements?

Spend

Money?

Completely crazy new stuff?

Fusion Chart

Highcharts

Zing Charts

...

NO

YES

NO

YES

Google Charts

Flotcharts

Morris.js

ChartJS

...

NO

YES

D3.js

Library based on D3.js

JS FRAMEWorks

HOW TO CHOOSE THE RIGHT ONE

jsgraphs.com 

Currently offline :(

Main Difference

Most Libraries

<script> 

 d3.select('#content')
    .append('svg')

       .selectAll('rect')

       .data([...])

       .enter()
       .append('rect')
           .attr('x', xValue)
           .attr('y', yValue)

           .attr('height', height)
           .attr('width', width);

</script>
<script> 

new Chart({
    type: 'bar-chart',
    target: '#content',

    settings: {
        width: '400',
        height: '300',
        caption: "A / B Testing",
        xAxis: "Time",
        yAxis: "$$$"
    },

    data: [...]

}).render();

</script>

VS.

D3.js

Let's explore

ScalableVectorGraphics

<rect>

height

width

(x, y)

<circle>

(cx, cy)

r

<line>

(x1, y1)

(x2, y2)

<text>

Description

(x, y)

text

Group

<g>

D3.js Example

<html>
   <head>
      <script src="http://d3js.org/d3.v4.min.js"></script>
   </head>
   <body>
      <div id="content"></div>

      <script> 
       d3.select('#content')
            .append('svg')

             .selectAll('rect')

             .data([100, 200, 150, 60, 50])

             .enter()
             .append('rect')
                .attr('x', 0)
                .attr('y', (data, index) => index * 25)

                .attr('height', 20)
                .attr('width', (data, index) => data);

Visible Result

Browser Result

<html>
    <head>
        <script src="http://d3js.org/d3.v4.min.js"></script>
    </head>
    <body>

        <div id="content">
            <svg>
                <rect x="0" y="0" height="20" width="100"></rect>
                <rect x="0" y="25" height="20" width="200"></rect>
                <rect x="0" y="50" height="20" width="150"></rect>
                <rect x="0" y="75" height="20" width="60"></rect>
                <rect x="0" y="100" height="20" width="50"></rect>
            </svg>
        </div>

        <script>[...]</script>

    </body>
</html>

Next STeps

const X_SCALE = (data) =>

    d3.scaleLinear()
        .range([0, 500])
        .domain([0, d3.max(data)]);

Scales

Next STeps

AXIS

chart.append('g')
        .attr('transform', 'translate(0, 100)')
        .call(d3.axisBottom(X_SCALE(data)));

0

200

100

50

150

Next STeps

bar.on('click', () => 

    bars.selectAll('rect')
            .sort((a, b) => d3.ascending(a, b));
);    

Interaction: Sort on Click

COMMON MISTAKES

d3.select('#content')
    
    .append('svg')
    .selectAll('g')
    .data(data)
    .enter()
    
    .append('g')
    .append('text')
           .attr('x', 0)
           .attr('y', (d, i) => {i * 37})
           .text((d, i) => d.title)

    .append('rect')
           .attr('height', 25)
           .attr('width', (d, i) => d.value * 25)
           .attr('x', 152)
           .attr('y', (d, i) => {i * 37});

Missing Return

Wrong Nesting

Duplicate Functions

Magic Numbers

Done Better

const MARGIN = 12;
const HEIGHT = 25;
const WIDTH_TEXT = 140;

const Y_SCALE = (d, i) => i * (HEIGHT + MARGIN);
const WIDTH_BAR = (d, i) => X_SCALE(data)(d.value);
const TITLE = (d, i) => d.title;

let svg = d3.select('#content').append('svg');

let chart = svg.selectAll('g').data(data).enter().append('g');

chart.append('text')
           .attr('x', 0)
           .attr('y', Y_SCALE)
           .text(TITLE);

chart.append('rect')
           .attr('x', WIDTH_TEXT + MARGIN)
           .attr('y', Y_SCALE)
           .attr('height', HEIGHT)
           .attr('width', WIDTH_BAR);

Lie Factor

Proportion

Scales

10 %

Variant B

20 %

Variant A

50

51

52

100

1000

EARLY Feedback

different data sets & People

Variant A

Variant B

Variant C

Blue Button Test 

Red Button Test

Yellow Button Test

  • Sun shine
  • Some circles
  • People sitting around a table
  • Strange logo

time

$$$

Better approach

Variant A

Variant B

For A/B Testing

Go 4 Gold!

General

  • policyviz.com/helpmeviz

  • infoactive.co/data-design

  • Information Dashboard Design: The Effective Visual Communication of Data

  • datavizcatalogue.com

  • improving-visualizations.org

  • storytellingwithdata.com

  • blog.tr3dent.com/2016/03/21/top-10-data-visualization-experts-to-follow-on-twitter

  • lisacharlotterost.github.io/2016/04/22/Colors-for-DataVis

GO 4 Gold!

D3

  • github.com/mbostock/d3/wiki

  • bost.ocks.org/mike/

  • bost.ocks.org/mike/chart

  • Getting Started with D3 by Mike Dewar

  • Interactive Data Visualization for the Web by Scott Murray (online for free: alignedleft.com/tutorials/d3/)

  • dashingd3js.com

  • http://techslides.com/over-2000-d3-js-examples-and-demos

  • http://christopheviau.com/d3list/gallery.html

QUESTIONS?

 

RECOMMENDATIONS?


REMARKS?

What do you think?

!?!

!?!

!?!

!?!

ThankS For

your Attention!

@_lizzelo_

slides.com/elisabethengel/dashboard-visualizations

Credits

comSysto GmbH

GUTEFrage.net GmbH

SKeTCH

Example

IN Depth

SVG

VS.

Canvas

Vector based

 

Faster on large screens

Events

Browser Inspection

CSS Styling

Import / Export

Accessibility

 

Pixel based

 

Faster for many objects

 

D3.js can

handle both

Updating Data

Document

A

B

C

D

E

NEW DATA

F