How Big Data is Changing User Engagement
-
Upload
mounia-lalmas -
Category
Internet
-
view
954 -
download
1
description
Transcript of How Big Data is Changing User Engagement
How Big Data is Changing User Engagement
Mounia Lalmas Yahoo Labs London [email protected]
Strata + HW Barcelona – 19 November 2014
This talk
§ User engagement › Definitions › Characteristics › Metrics
§ Case studies: Absence time › Within sessions › Across sessions
User engagement
What is user engagement?
“User engagement is a quality of the user experience that emphasizes the phenomena associated with wanting to use a technological resource longer and frequently” (Attfield et al, 2011) self-report: happy, sad, enjoyment, …
emotional, cognitive and behavioural connection that exists, at any point in time and over time, between a user and a technological resource
analytics: click, upload, read, comment, share …
physiology: gaze, body heat, mouse movement, …
4
focus of this talk … Big Data
Why is it important to engage users? § In today’s wired world, users have enhanced expectations
about their interactions with technology … resulting in increased competition amongst the
purveyors and designers of interactive systems. § In addition to utilitarian factors, such as usability, we must
consider the hedonic and experiential factors of interacting with technology, such as fun, fulfillment, play, and user engagement.
(O’Brien, Lalmas & Yom-Tov, 2014)
Online sites differ with respect to their engagement pattern
Games Users spend much time per visit
Search Users come frequently and do not stay long
Social media Users come frequently and stay long
Niche Users come on average once a week e.g. weekly post
News Users come periodically, e.g. morning and evening
Service Users visit site, when needed, e.g. to renew subscription
(Lehmann etal, 2012)
Why is it important interpret user engagement measurement well?
CTR
new version
Characteristics of user engagement
Focused attention
Positive Affect
Aesthetics
Endurability
Novelty
Richness and control
Reputation, trust and expectation
Motivation, interests, incentives, and benefits
(Attfield etal, 2011)
Large scale measurements – analytics
within-session measures across-session measures
• Dwell time • Session duration • Bounce rate • Play time (video) • Click through rate (CTR) • Number of pages viewed • Conversion rate • Number of UCG (comments). • Time between visits • …
• Fraction of return visits • Total view time per month (video) • Lifetime value (number of actions) • Number of sessions per unit of time • Total usage time per unit of time • Number of friends on site (social
networks) • Number of UCG (comments) • Time between visits • …
• within-session engagement measures success in attracting user to the site.
• across-session engagement measured by observing user long-term value.
loyalty popularity
activity
Two case studies: Absence time
news facebook mail news news news
absence time
within sessions across sessions
Case study I Absence time within sessions
The context – Online multitasking Browsing the “old way”
facebook news news news news mail
1min 2min 1min 3min
Dwell time during a visit on news site: 7min on average
news site
news facebook mail news news news
1min 2min 1min 3min
Dwell time during a visit on news site: 2.33min on average (1min | 3min | 3min)
Browsing “nowadays”
The dataset … Big Data
› July 2012 › 2.5M users › 785M page views
› Categorization of the most frequent accessed sites
• 11 categories (e.g. news), 33 subcategories (e.g. news finance, news society)
• 760 sites from 70 countries/regions
short sessions: average 3.01 distinct sites visited with revisitation rate 10% long sessions: average 9.62 distinct sites visited with revisitation rate 22%
Time spent at each revisit
0.23
0.28
0.33
mail sites news (finance) sites news (tech) sitessocial media sites
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
100%
50%
0%
decreasing attention increasing attention constant attention complex attention
BackpagingHyperlinkingTeleportation
Per
centa
geof
nav
igat
ion t
ype
Pro
por
tion
of to
tal
dw
ell ti
me
on s
ite
ith visit on site ith visit on site ith visit on site ith visit on site
p-value = 0.09m = -0.01
p-value = 0.07m = -0.02
p-value = 0.79m = 0.00
Four patterns: decreasing, increasing, constant and complex Backpaging increasingly used at each revisit
Metrics: 1. AttShift: 1 (-1) when time spent at each revisit increases (decreases);
0 when no pattern 2. AttRange: 0 when time spent at each revisit is constant
Time spent between each visit
• 50% of sites are revisited after
less than 1min interruption of a task?
• There are revisits after a long break
returning to a site to perform a new task?
0.00
0.25
0.50
0.75
1.00
10ï2 10ï1 100 101 102
mailsocial media
news (finance)news (tech)
Cum
ula
tive
pro
bab
ility
Absence time [min]
2nd
1st 3rd … visit
absence time
Metric: 1. CumAct: increases with time spent between visits
§ Clustering of sites using multitasking and standard engagement metrics: › CumAct, AttShift, AttRange › Visit, Session
§ Five models
Models of online multitasking
C4: 74 sites
0.25
-0.25
0.75
-0.75
C5: 166 sites
0.25
-0.25
0.75
-0.75
C3: 156 sites
0.25
-0.25
0.75
-0.75
C2: 108 sites
0.25
-0.25
0.75
-0.75
C1: 172 sites
0.25
-0.25
0.75
-0.75
Visitdt
[min] CumActdt,3
AttShiftdt,4
AttRangedt,4
Sessiondt
[min]
One task during a session
Visitdt
[min] CumActdt,3
AttShiftdt,4
AttRangedt,4
Sessiondt
[min]
C2: 108 sitesauctions, front page,
shopping, dating
0.25
-0.25
0.75
-0.75
C1: 172 sitesmail, maps, news,
news (soc.)
0.25
-0.25
0.75
-0.75
§ High dwell time per visit and during
entire session § Users return to continue a task
(short absence time) § C1: attention is shifting to another
site § C2: attention is shifting slowly
towards the site
C4: 74 sitesfront page, search,
download
C3: 156 sitesauctions, search,
front page, shopping
0.25
-0.25
0.75
-0.75
0.25
-0.25
0.75
-0.75
Several tasks during a session
§ Users perform several tasks during
a session
§ No simple activity pattern
§ C3: Dwell time per visit is low, but dwell time per session is high
Visitdt
[min] CumActdt,3
AttShiftdt,4
AttRangedt,4
Sessiondt
[min]
C5: 166 sitesservice, download,
blogging, news (soc.)
0.25
-0.25
0.75
-0.75
Sites with low activity
§ Users do not spend a lot of time on
these sites § Time between visits is short § Attention is shifting towards the site
Visitdt
[min] CumActdt,3
AttShiftdt,4
AttRangedt,4
Sessiondt
[min]
C2: 108 sitesauctions, front page,
shopping, dating
0.25
-0.25
0.75
-0.75
C3: 156 sitesauctions, search,
front page, shopping
0.25
-0.25
0.75
-0.75
Browsing behavior can differ between sites of the same category
§ C2: users visit site once to perform
their task
§ C3: users visit site several times to perform task(s)
Visitdt
[min] CumActdt,3
AttShiftdt,4
AttRangedt,4
Sessiondt
[min]
The message – absence time within sessions
Towards a taxonomy of online multitasking One main continuous task versus user loyalty New metrics that account for time spent between visits and at each visit Applications: story-focused reading & network of sites
Online multitasking happens How it affects a site depends on the site
(Lehmann etal, 2013)
Case study II Absence time across sessions
The context – search experience
The context – search experience
Absence time and survival analysis
story 1story 2story 3story 4story 5story 6story 7story 8story 9
0 5 10 15 20
0.0
0.2
0.4
0.6
0.8
1.0
Users (%) who did come back
Users (%) who read story 2 but did not come back after 10 hours
SURVIVE
DIE
DIE = RETURN TO SITE èSHORT ABSENCE TIME
The dataset … Big Data Ranking function on Yahoo Answer Japan
Two-weeks click data on Yahoo Answer Japan: search One millions users Six ranking functions 30-minute session boundary
Examples of search session metrics
§ Number of clicks
§ Click at given position
§ Time to first click § Skipping
§ Abandonment rate § Number of query reformulations
survival analysis: high hazard rate (die quickly) = short absence
5 clicks
cont
rol =
no
clic
k
Absence time and number of clicks on search result page
3 clicks
Absence time – search experience
1. No click means a bad user experience 2. Clicking between 3-5 results leads to same user experience 3. Clicking on more than 5 results reflects poorer user experience; users
cannot find what they are looking for 4. Clicking lower in the ranking (2nd, 3rd) suggests more careful choice from
the user (compared to 1st) 5. Clicking at bottom is a sign of low quality overall ranking 6. Users finding their answers quickly (click sooner) return sooner to the
search application 7. Returning to the same search result page is a worse user experience than
reformulating the query
search session metrics absence time
The message – absence time across sessions
short absence is a sign of loyalty
important indication of user satisfaction with search portal
(Dupret & Lalmas, 2013)
Easy to implement and interpret Can compare many things in one go No need to estimate baselines But need lots of data to account for noise Applications: direct display & churn rate & advertising
Some final words
§ “If you cannot measure it, you cannot improve it” – William Thomson (Lord Kelvin)
§ “You cannot control what you cannot measure” – DeMarco
§ “The way you measure is more important than what you measure” – Art Gust