@jeroentjepkema
Performance Usability Experience:: ::
Rules of Thumb
Why your users will quit your online (mobile) strategy
Multichannel 2013,
April 24, 2013
http://bit.ly/MW-MC13
Performance?
0"
5"
10"
15"
20"
25"
30"
0" 1" 2" 3" 4" 5" 6" 7" 8" 9" 10" 11" 12" 13" 14" 15"
Abandonment"Rate"(%)"
Page"Load"Time"Band"(sec.)"
Abandonment*Rate*,*
All*Browsers*
Source: Gomez real user monitoring
Abandonment"Rate"Across"200+"Sites"/"177+"Million"Page"Views"
Over"2"weeks"/"All#Browsers#
0"
5"
10"
15"
20"
25"
30"
0" 1" 2" 3" 4" 5" 6" 7" 8" 9" 10" 11" 12" 13" 14" 15"
Abandonment"Rate"(%)"
Page"Load"Time"Band"(sec.)"
Abandonment*Rate*,*
All*Browsers*
Source: Gomez real user monitoring
Abandonment"Rate"Across"200+"Sites"/"177+"Million"Page"Views"
Over"2"weeks"/"All#Browsers#
0"
5"
10"
15"
20"
25"
30"
0" 1" 2" 3" 4" 5" 6" 7" 8" 9" 10" 11" 12" 13" 14" 15"
Abandonment"Rate"(%)"
Page"Load"Time"Band"(sec.)"
Abandonment*Rate*,*
All*Browsers*
Abandonment*Rate*,*
iPhone*Safari*
Source: Gomez real user monitoring
Abandonment"Rate"Across"200+"Sites"/"177+"Million"Page"Views"
Over"2"weeks"/"All#Browsers#vs.#iPhone#Safari#
@jeroentjepkema, MeasureWorks
(inter)action
We
0
0,3
Instantaneous: I like it!
0
1
0,3
Instantaneous: I like it!
Interaction: Let’s conversate...
0
1
3
0,3
Instantaneous: I like it!
Interaction: Let’s conversate...
Mmm, shall I click away?
0
10
1
3
0,3
Instantaneous: I like it!
Interaction: Let’s conversate...
Mmm, shall I click away?
Only if the task/content is relevant
0
10
1
3
0,3
Do I like it?
Do I want to conversate?
This is where the magic happens
Mmm, shall I click away?
Only if the task/content is relevant
0
How we perceive “interaction”?
Sensitive for specific tasks
How we perceive “interaction”?
Always compared to past experiences
Sensitive for specific tasks
How we perceive “interaction”?
Always compared to past experiences
Task completion has positive impact
Sensitive for specific tasks
How we perceive “interaction”?
Slowdowns hurts
Task completion has positive impact
Sensitive for specific tasks
How we perceive “interaction”?
Always compared to past experiences
Experience
the
difference
Tested with webpagetest.org (3G traffic shaping, 2000Kbps Down, 1000 Kbps Up, 150ms latency)
Ultimate goal = “flow state”
Relevant content,
to support
task completion
With fast delivery,
focus on interaction
Flow is an
“optimal experience”
that is
“intrinsically enjoyable”
“”
Myth busted?
Technology vs. Perceived
Experiment 1:
Desktop
Which website would you recommend in terms of
performance?
100 users
Buy a summer holiday
online
Desktop only, fixed
selection of shops
(Travel30)
1
2
3
4
5
Perceived Technology
Sunweb
Globe
Neckermann
D-reizen
Arke
Ranking
Study performed by MeasureWorks for Emerce eTravel, 2013
Technology vs. Perceived
1
2
3
4
5
Perceived Technology
Sunweb Sunweb
D-reizen Globe
Neckermann Neckermann
Arke D-reizen
Globe Arke
Ranking
Study performed by MeasureWorks for Emerce eTravel, 2013
Technology vs. Perceived
1
2
3
4
5
Perceived Technology
Sunweb Sunweb
D-reizen Globe
Neckermann Neckermann
Arke D-reizen
Globe Arke
Ranking
Study performed by MeasureWorks for Emerce eTravel, 2013
Technology vs. Perceived
Experiment 2:
Mobile
Mobile shopping experience setup....
100 mobile users
Mobile browsers only
via 3G, no WiFi
Perform 2 tasks:
‣Buy a book
‣Buy a T-shirt
Speed
0
12,5
25
37,5
50
Zalando HM V&D Tom Tailor
3,97,7
15,4
42,1
1. Buy a T-shirt 2. Review
4
45
2420
Design Speed Mobile Readiness Other
0
10
20
30
40
Zalando HM V&D Tom Tailor
8
34
15,412
6
18
54
21
Design Speed Mobile Readiness Other
Round1Round2
Many functional
issues reported
with Zalando
Mobile Only research:
- Task completion: Only use smartphone to buy a book
- N = 100, users range from 20-65
Nexus - Android - 3G
Tested with webpagetest.org (3G traffic shaping, 2000Kbps Down, 1000 Kbps Up, 150ms latency)
Nexus - Android - 3G
Tested with webpagetest.org (3G traffic shaping, 2000Kbps Down, 1000 Kbps Up, 150ms latency)
Mobile
readiness
0
25
50
75
100
Bol.com Selexyz Bruna Other
1. Buy a book 2. Review 3. Buy at a different store
1616
24
44
Design Speed Mobile Readiness Other
12
46
26
16
Design Speed Mobile Readiness Other
Mobile Only research:
- Task completion: Only use smartphone to buy a book
- N = 100, users range from 20-65
Bol.com Selexyz De Boekerij ECI Bruna Cosmox
Target your audience!
0%
10%
20%
30%
40%
Speed Readiness
18%
40% 40%
25%
< 34 > 35
Mobile Web
42%
Native
58%
Mobile Web
55%
Native
45%
< 34 > 35
Recipe for
performance
You
1.
Eco-System of
me
Deliver the same experience to all your (mobile) users
Regular website Native Application
Deliver the same experience to all your (mobile) users
Regular website Native ApplicationOptimized Mobile Web
Deliver the same experience to all your (mobile) users
Mobile first
experience
Prioritize your business goals
Mobile first
experience
Mobile first
experience
Prioritize your business goals
Deliver content accordingly
0
1
2
3
4
5
6
7
8
9
MoneYou ING IEX Binck Alex ABN Amro
4,84,64,6
2,32,3
3,5
8,9
3,8
8,5
2
3
7,9
1,3
1,8
0,6
3,7
0,91,2
Desktop Mobile Native
PageLoadTime(sec.)
Mobile First experience?
Tested with webpagetest.org & gomez (For desktop 8Mbs up/ 1,5Mbs down, for Mobile 3G traffic shaping, 2000Kbps Down, 1000 Kbps Up, 150ms latency)
#2
Design for
(mobile)
use cases
Focus on
task
completion
With
mobile
conditions
in mind
Optimized for the right screen
Native app
Mobile Web - Landscape
Mobile Web - Portrait
#3
Collecting
data
INTERNET CUSTOMERS
Third-party/
Cloud Services
Content
Delivery
Networks
Local
ISP
Mobile
Carriers
Major
ISP
INTERNAL USERS(CLOUD) DATA CENTER
Storage DB Servers Web Servers
App
Servers
Middleware
ServersMainframe
Load
Balancers
Network
INTERNET CUSTOMERS
Third-party/
Cloud Services
Content
Delivery
Networks
Local
ISP
Mobile
Carriers
Major
ISP
INTERNAL USERS(CLOUD) DATA CENTER
Storage DB Servers Web Servers
App
Servers
Middleware
ServersMainframe
Load
Balancers
NetworkThis is what you control...
INTERNET CUSTOMERS
Third-party/
Cloud Services
Content
Delivery
Networks
Local
ISP
Mobile
Carriers
Major
ISP
INTERNAL USERS(CLOUD) DATA CENTER
Storage DB Servers Web Servers
App
Servers
Middleware
ServersMainframe
Load
Balancers
NetworkThis is what you control... What you’re blamed for..
Measuring performance?
Measuring performance? Outside-in, from
the user’s perspective...
Use
real devices
and/or users
With both 3G
and WiFi
connections
Based on
Real usage
scenario’s
0
5
10
15
20
25
30
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5 6 7 8 9 10
2,9 2,5
3,2
2,4
3,6
2,6 2,5 2,5 2,7 2,8
22,0
2,5 2,6
19,0
2,7 3,0 2,7 2,5 2,8 2,8 3,1 2,6 2,5 2,5 2,8 2,6
Worldticketcenter.nl
31 May 2012 1 June 2012
Worldticketcenter.nl
0
5
10
15
20
25
30
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5 6 7 8 9 10
31 May 2012 1 June 2012
#fail
#fail
#fail
#fail
#fail
#fail
Source: Compuware, 2012, Engaging the tablet user; What to expect from web sites?
Why it matters?
Source: Compuware, 2012, Engaging the tablet user; What to expect from web sites?
Why it matters? 76% of mobile
consumers won’t retry more than twice
Source: Oracle, 2012, Customer Experience Survey
Why it matters? 89% of (mobile)
consumers will leave to the competition
To summarize...
Performance is
a requirement
Net Performance Score
Create great
service
Real users,
real devices,
real data
Detecting
Trends
Don’t let them
click away
Create an
experience
that
performs
Questions?
View slides: http://bit.ly/MW-MC13
More performance?
Join us May 18, webperfdays.org
Thanks! More questions?
M: jtjepkema@measureworks.nl
T: @jeroentjepkema
W: www.measureworks.nl

MeasureWorks - Multichannel 2013 - Why your users quit your online (mobile) strategy