Good bye conversion rate - a smarter way to optimising Search Engine Result Pages

Faculty of Computer Science
Professorship Distributed and self-organizing Systems
Good bye conversion rate
a smarter way to optimising Search Engine
Results Pages
MartinGaedke.com
VSR.Informatik.TU-Chemnitz.de
///// IS-SWIS 2017, Rotterdam, February 9, 2017///

2
Search Engine
Search
Search
Search
Search
Martin Gaedke - http://vsr.informatik.tu-chemnitz.de

Search Engine Result Pages (SERPs)
3Martin Gaedke - http://vsr.informatik.tu-chemnitz.de

(simplified) Motivation
4
Evaluation Method Efficient? Effective?
User Testing ✖ ✔
Usability Inspection ✖ ✔
Split Testing ✔ ✖

Split Testing
5
Buy this
Buy this
Version A Version B

Conversion Rate:
proportion of visitors to a website
who take action due to subtle
requests from marketers, SEOs,
(semantic) data providers etc.
The problem with Split Testing
(source: Wikipedia.com)

Version A: Getting the User‘s Zip Code

8
$$$ ≠ Usability
Jakob Nielsen: Putting A/B Testing in Its Place,
http://www.nngroup.com/articles/putting-ab-testing-in-its-place/

Goal:
to leverage the
advantages of split
testing AND to
facilitate effective
usability evaluation

Idea:
Rethink the target measure
from the users point of
view (i.e. development &
user – not sales) :
Usability as a measure
for split testing

11
Idea: Infer quantitative measure of
usability from user interactions.

Usability-based Split Testing
12
Interface A Interface B
Questionnaire
[WaPPU – Usability based A/B-Testing
Speicher, Both, Gaedke, ICWE2014]
WaPPU
Usability System
(Naive Bayes classier for the
learning)
Model
for
Usability

Usability-based Split Testing
13
Interface A Interface B
[WaPPU – Usability based A/B-Testing
Speicher, Both, Gaedke, ICWE2014]
WaPPU
Usability System
TRAINED
Model
for
Usability

Usability Model allows for:
f(cause) = score
(cause is part of usability)

Sure!
f(cause) = score
this function is not bijective
(Not possible to infer optimizations from usability scores )
Limitations?

How to overcome the
f(cause) = score
limitations?
17

Catalog of best
practices
a.k.a. Set of causes
and optimizations
18

19
SERP Optimization Suite
(S.O.S.)
WaPPU
for evaluation
Catalog
of best practices
[S.O.S.: Does Your Search Engine Results Page (SERP) Need Help?
Speicher, Both, Gaedke, CHI 2015]

20
fu (score) = {C, C'}
u = usability item number
C = potential causes
C' = countermeasures

S.O.S. example: "We have a
problem with Distraction"

27
User Study
• 81 participants (~62% male)
• avg. age = 31.08 years
• Task: Solve a problem

28
“Find a birthday present for a good
friend that does not cost more
than 50 Euros.”

32
Results
old SERP new SERP
informativeness -0.17 â < -0.02 â
understandability 0.34 â < 0.45 â
confusion 0.30 â < 0.38 â
distraction* 0.36 â < 0.62 ä
readability 0.45 â < 0.52 â
information
density*
0.04 â < 0.43 ä
accessibility 0.06 â < 0.07 â
overall usability* 59.86% â < 67.50% ä

n S.O.S. significantly improved SERP usability
n Approach developed using HCD/DT
n obtaining usability scores through WaPPU and
n recommendations for optimization from a
catalog for suboptimal scores
n http://vsr.informatik.tu-chemnitz.de/demo/SOS
Conclusions

n Catalogs for other categories of Web pages
n Creating trained model for web page and site
categories
n Crating more generic and reusable catalogs
n Reducing the learning time
n Applying the approach to the quality of linked
enterprise data
n Improving: incorrect, incomplete etc. data
n Dealing with: Co-evolution, Coherence etc.
n Data Quality à Fit for business use
n LEDS project:
http://www.leds-projekt.de/
Future Work

Data Quality can be interpreted as the degree
to which data fits to current requirements
𝐷𝑄#$%&'(& = ∑ ∑ 𝜔-'. / 𝑓𝑢𝑙𝑓𝑖𝑙𝑙 𝑟𝑒𝑞, 𝑜𝑏𝑗-'.<='>'?< + 𝜑

First Inspiring Results
Output of Fitness results
of our Quality Assessment
tool with the means of the
data quality vocabulary
(dqv)
Feel free to visit:
www.leds-projekt.de

Chemnitz University of Technology #VSR
Thank You!
martin@gaedke.com
VSR.Informatik.TU-Chemnitz.de
@gaedke /gaedke

Good bye conversion rate - a smarter way to optimising Search Engine Result Pages

More Related Content

Similar to Good bye conversion rate - a smarter way to optimising Search Engine Result Pages (20)

Recently uploaded (20)

Good bye conversion rate - a smarter way to optimising Search Engine Result Pages