Towards Appraising Online Stores

1
Towards Appraising Online Stores

• Ernest Cachia
• Mark Micallef
• Software Engineering Process Improvement Research
  Group
• (SEPI)

2
Introduction
3
Overall Goals
4
The First Step
5
Initial Targets

• Identify most important quality attributes in
  e-commerce systems
• Limit Scope to the Online Store variety of
  E-Commerce Systems (for the time being)
• Success of such systems depends largely on users
  accessing a website and buying products or
  services
• Therefore, users opinions should be given a
  large weighting

6
Overview of Activities

• Research into existing work
• Discovering how e-commerce systems differ from
  generic systems
• Designing a questionnaire
• Conduct a survey amongst 350 e-commerce users
• Analysis of results
• Extraction of important e-commerce attributes
  (and their importance ranking)

7
How do E-Commerce Systems Differ?
8
Why was this information required?

• Identifying why e-commerce systems differ was an
  important step
• Further justifies the need for an e-commerce
  metrication framework
• Allows us to identify quality attributes based on
  these differences
• 5 main differences were identified

9
How E-Commerce System Differ

• Content Driven (Navigability)
• Exposed to the world (Security)
• Browser-Based (Portability)
• Enormous User Base (Scalability, Multilinguality)
• Frequent Change (Maintainability, Reliability)

10
Survey Organisation
11
Survey Design

• Designed to rank importance of quality attributes
  defined above
• Divided into two parts
• Section 1 Collection of demographic information
  about users
• Section 2 Collection of information about which
  attributes are deemed most important by users

12
Example Questions from Section 1

• Age Group
• Main use of e-commerce (personal or business)
• Education Level
• Frequency of online shopping
• Products/Services bought online
• OS, Browser, Devices

13
Example Questions from Section 2

• Abandoned Transactions
• Reasons for abandoned transactions
• First Impression Factors
• Intentions when visiting an e-commerce site
• Ranking of attributes/features

14
Survey Results
15
Our Sample Space

• 350 Respondents
• 94 of participants were under 50 years old
• Browser usage, OS usage and Device usage compare
  to other surveys carried out

16
Browsers
17
Operating Systems
18
Devices
19
Most Popular Products
20
Abandoned Transactions
21
Reasons for Abandoned Transactions
22
Security and Privacy Issues (1/3)

• Security turned out to be an important factor
• For example, when asked to identify a reason why
  they would prefer not to use e-commerce, users
  replied as follows

23
Security and Privacy Issues (2/3)

• How sure would you have to be about a sites
  security capabilities before you buy from it?

24
Security and Privacy Issues (3/3)

• When asked to rank the importance of quality
  attributes in order of importance, security
  received and average scored of 6.24 (out of a
  possible 7).
• This made it the most important quality attribute
  as perceived by users of e-commerce systems
• We recommended security as an essential quality
  attribute in e-commerce systems

25
Portability Issues (1/4)

• Refers to the extent to which a system is
  accessible from different operating systems,
  browser and devices
• Participants ranked portability as the 5th most
  important attribute (out of 7)
• Case for portability not too strong if one relies
  on the results of this survey

26
Portability Issues (2/4)

• 98 of respondents use WindowsTM -based OSs
• Almost 93 use Desktop PCs
• 95 use Internet ExplorerTM
• Less than 4 of users who abandoned transactions
  did so because of compatibility problems

27
Portability Issues (3/4)

• Results seem to indicate that portability is not
  important
• However one must be careful
• 88.7 of users will not change browsers so as to
  be able to view a particular site
• A small percentage of customer lost due to lack
  portability might still amount to large amounts
  of lost revenue
• New mobile technologies are likely increase the
  importance of portability issues in e-commerce

28
Portability (4/4)

• Portability is recommended as a necessary
  attribute in e-commerce systems
• However, in the current environment,
  compatibility with dominant technologies would
  seem to ensure a sufficient reach

29
Performance and Scalability (1/3)

• Speed is important to users
• 33 of users who abandoned transactions mid-way
  through did so because the site was too slow
• Only 18.4 would remain unconditionally loyal to
  their favorite site if it became slow

30
Performance and Scalability (2/3)

• What is the most important first impression
  factor you look for in a site?

31
Performance and Scalability (3/3)

• Participants ranked performance as the 4th most
  important quality attribute
• We are recommending Performance as an essential
  quality attribute

32
Navigability (1/3)

• Why did you abandon transactions mid-way through?

Also more likely to simply find another site
33
Navigability (2/3)

• 72 of users know beforehand what they are
  looking for when visiting an online store
• This indicates that search and navigation
  features are important
• 30 of participants also chose navigation as
  their primary first impression factor

34
Navigability (3/3)

• Navigability was ranked as the 3rd most important
  attribute by users
• We are recommending usability with an emphasis on
  navigability as an essential quality attribute in
  online stores

35
Multi-Lingual Features

• Not being recommended as an essential attribute
• The importance of such features is highly
  dependant on the context in which an online store
  is operating
• 51 of participants claim e-commerce systems
  could still be classified as a quality system if
  it had no multi-lingual features
• Respondents ranked this feature in 6th place (out
  of 7)

36
Low-Bandwidth Version of Site

• Some e-commerce sites have a text-only or
  low-bandwidth version available
• Useful for users with slow connections
• 52.5 of participants deem such a feature
  unnecessary
• Users ranked this attribute as the least
  important
• Considering increasing popularity of broadband
  connections, we are not recommending this as a
  necessary attribute

37
Ranking of Essential Attributes

• Security
• Reliability
• Navigability
• Performance
• Portability

38
Conclusions and Further Work
39
Conclusions and Further Work

• These results are an important first step to our
  overall goals
• Since this research we have explored
• Information Architecture
• Graph Theory
• Security
• Further work will include
• Investigation into each of the above attributes
  and how they can be measured
• Development of a metrication framework that will
  combine metrics for different attributes together
• Development of a list of guidelines
• Incorporation of guidelines into framework

40
Questions and Discussion