Evaluating the Effectiveness of Leading Job Portals: A Cross-Platform

Analysis

Mettu Siddhartha, M Ramya Sree,Poli Vamsi Vardhan Reddy, Balam Ruchith Balaji, Parlapalli
Bhargav Reddy and Radha D*
Department of Computer Science and Engineering
Amrita School of Computing, Bengaluru
Amrita Vishwa Vidyapeetham, India
[email protected], [email protected],d [email protected]

Keywords: Job searching platforms, SEO analysis, Website performance, Usability best practices, Core Web Vitals,
PageSpeed Insights.

Abstract: The modern era has seen widespread adoption of online job platforms, connecting job seekers directly with
potential employers. In the proposed work, various key performance evaluation tools—such as Google’s Page-
Speed Insights, Pingdom, SEMrush, and GTmetrix—have been employed to assess metrics like Performance,
SEO(Search Engine Optimization), Accessibility, Best Practices, Load Time, and Traffic, among others. These
tools provide insights into platform strengths and limitations, independent of developer influence, thereby
ensuring objectivity and offering substantial recommendations for enhancing user experience and technical
aspects. The analysis concludes that Job Board-1 demonstrates a high level of competence across nearly all
critical factors essential for traffic, audience engagement, and platform availability, and it presents several
technological strengths. Job Board-3, while performing reasonably well in terms of accessibility and SEO,
faces challenges related to high bounce rates and slow loading speeds, suggesting areas for improvement to
increase user engagement.

1 INTRODUCTION spective, they were insufficiently detailed to compre-

hend the fundamental elements that lead a website to
Job seeking in the digital age has transitioned from successfully furnish a smooth job search encounter.
offline to online, with websites serving as a vital con- Selected job search platforms will be evaluated
duit between job seekers and potential employers. across a range of parameters using industry-standard
These platforms, which now include Job Board-1, Job tools. Starting the evaluation, the performance, SEO,
Board-2, Job Board-3, and Job Board-4, are essential accessibility, and other best practices are checked
resources for anyone looking to advance in their ca- with Google PageSpeed Insights. This involves con-
reers. Each platform offers unique features and ser- sidering factors such as title tags Meta description,
vices tailored to meet the needs of its diverse user server response time, loading speed of both the mo-
base, ranging from job listings and company ratings to bile and the desk top version of the Website. The
networking opportunities. The effectiveness of these properties like metadata, mobile-responsiveness, and
platforms relies not only on the number of job post- structured data are analysed in order to understand the
ings but also on their overall performance, usability, websites’ visibility and their capability of SEO opti-
and accessibility. mization. Moreover, the issues of usability and acces-
Page load speeds and user interface design were sibility are also verified by such criteria as the exis-
the main emphasis of traditional website performance tence of alt text for images, keyboard navigation, and
monitoring. However, a more thorough study is now color perception violations on the platforms. About
achievable because to the development of increas- performance I use Pingdom by giving tests from a
ingly advanced techniques and technology. Conven- server in Japan to examine performance in the Region
tional approaches frequently involved basic mobile of Asia. To analyze the traffic of various platforms,
compatibility tests and broad user satisfaction sur- the SEMrush tool is used, including the total visits,
veys. Although these approaches offered a wide per- visits by unique visitors, bounce rates, and the average
time spent by a user on a site. This site goes further in electronics companies[2]. It evaluates performance
giving different general performance indicators like indicators like visitor attrs, bounce rate, average stay
LCP, TBT, CLS, total page weight, and the quantity duration, most popular pages, ctr, backlink, and re-
of requests to give a full report in how well these Plat ferring domain. The AIDA model, developed by St.
forms run. By adopting a multi-dimensional perspec- Elmo Lewis, is used to evaluate promotive activities
tive to analyze each platform, the strategies for how of some brands. The findings help brands under-
to optimize them are also shown in detail. stand consumer behavior trends and their digital foot-
The remaining part of the paper consists of re- prints, guiding their marketing approach. Web ana-
view on the background information and introduction lytics have developed rapidly, with SEMrush being a
to the topic is discussed in section 1.and Section 2 user-friendly tool.
presents literature of related studies and prior work A web-based solution that uses WebpageTest,
done in similar field. Section 3 provides the method PageSpeed Insight, and GTmetrix tools to automat-
explaining the method used, methods employed and ically gather and compare e-commerce site perfor-
instruments applied in the study. Section 4 displays mance metrics was presented by Hossain et al.[3].
the results in form of findings, data and any analysis The application uses PHP, MYSQL, CSS, and HTML,
done on them. Last, Section 5 provides a summary of and allows users to input the URL of a site. Tests on
the findings and the implications of the research for ten Bangladeshi e-commerce sites showed site7 had
practice. the lowest Total Blocking Time, while site10 had the
lowest Load Time. The application currently supports
computer systems, with future research focusing on
2 LITERATURE REVIEW mobile versions and testing limits.
According to Palacios-Zamora et al. assessing
The section outlines prior work done on measur- university websites’ effectiveness is crucial to en-
ing the performance of the website, search engine op- hancing their reputation [4]. It discusses various qual-
timization, keyword study and usability studies across ity assessment models and highlights the impact of
multiple industries. It underlines the role of recogniz- parameters like response time, throughput, utilization,
ing the shifts of keywords and users’ behavior and the extensibility, data transfer rate, concurrency, and reli-
influence made by SEs’ algorithms on the website. ability on website performance. Measures like opti-
Such as SEMrush, PageSpeed Insight, or GTmetrix mizing images, layouts for desktop, tablet, or mobile
that is used to review the performance and to deter- devices, and content caching can enhance site effi-
mine slips. Largest Contentful Paint and First Input ciency. The research also suggests improving mobile
Delay are among the most important metrics config- device performance and using the increasing, usabil-
ured within the Core Web Vitals to address user expe- ity, and accessibility brief.
rience and site performance needs. Various aspects of When evaluating user experience, the authors em-
the website should be optimized on a continuous basis phasized importance of PageSpeed Insights (Web-
and the results should be compared with industry tar- PageTest), First Input Delay (FID), and Largest Con-
gets to ensure that a high efficiency with subsequent tentful Paint (LCP) [5]. It suggests that web develop-
improved search engine ranking is attained. ers should focus on improving these parameters to en-
Nanda et al. study is to examine the trend analy- hance website performance, efficiency, and business
sis of the keywords, questions, and website domains results. This will ultimately lead to increased conver-
most frequently associated with melanoma and skin sion rates, creating a competitive advantage in online
cancer search [1]. All the information was obtained business.
using different search engines and findings indicated Bernine et al. provided an analytical model
higher search terms based on skin cancer as com- grounded on Petri nets in assessing the effectiveness
pared to melanoma. The clinicopathologic classifi- of a Web services structure if requests and services
cation and diagnosis constituted the largest group of adhere to an exponential server[6]. This mainly ap-
the ten most frequently used melanoma keywords. plies since the arrival of user as well as the web ser-
The survival query was the most popular but gen- vice requests follows the Poisson distribution only.
eral/melanoma or diagnosis type query type yielded The model is solved analytically, and the response
the highest number of searches per query. To help re- time and mean number of clients for the system are
solve common issues related to melanoma and skin determined. The limit number of clients in the system
cancer, the study suggests explaining. is derived from which the system becomes congested.
Using the SEMrush analytics tool, Afroz et al. According to Kumar et al. website performance
evaluated and analysed the websites of consumer has a significant impact on user experience, search
engine optimisation, and overall business success tion, then the research highlights that the Decorate Al-
[7]. Key automated tools such as Google Page- gorithm an ensemble classifier is efficient and reliable
Speed Insights, GTmetrix, Pingdom, WebPageTest, than other algorithms.
and Apache JMeter are reviewed by the authors, Sujee et al. explained Modern educational
who highlight their usefulness in locating perfor- databases have expanded, and it is possible to find
mance bottlenecks and streamlining web applications. many hidden resources that can contribute to improv-
The authors support a methodical approach to per- ing the results achieved by students [12]. This work
formance evaluation that enhances site efficacy and explains how predictive modeling, grouping, and as-
efficiency by using a variety of instruments, ongo- sociation rule mining can be employed in order to
ing monitoring, and benchmarking against industry discover information beneficial to students and tutors.
norms. RBF model enables one to predict which students are
Akgül et al. highlighted the significance of public performing well and which group could benefit from
value, usability, and readability in improving user ex- a boost in instructions. It helps the instructors in the
perience and public service delivery of Turkish gov- settings to present teaching methodologies that fits ev-
ernment website performance [8]. The authors eval- ery student’s needs successfully.
uate how well these websites meet the needs of cit- The domain, domain age, web impact variables,
izens, encourage participation, and guarantee that all and Alexa traffic rank of Indian universities were
users can access the material. Using a mixed-methods compared with the NIRF ranking 2021 by Meghwal et
approach and making use of automated technologies al. [13]. It discovered that all the websites in the study
such as Google PageSpeed Insights, GTmetrix, and employed SEO tools. Out of the universities, Amrita
WebPageTest, the study advocates for a user-centred Vishwa Vidyapeetham university had the oldest do-
strategy in e-government projects and offers useful main registration in December 1988 and the Indian
suggestions for enhancing government websites. Institute of Science university had the highest Domain
The idea of SEO and the methods by which stan- Authority score of 62. A National level survey found
dard tools can be used to evaluate a website’s search it that three Universities of Karnataka were in the top
engine optimisation were described by Simec et al. ten Universities in India.
[9]. As an example, it proposes to apply three tools Subbulakshmi et al. explained how a framework
to evaluate the same site and compare their func- for extending the assessment of Web portal pertinence
tions and options. In the same token, SEO is a very can be used to measure the reliability of Web portals
fluid field where one is always in assimilating new and to calculate their credibility score [14]. Main pa-
information in the field. Semrush is an online tool rameters taken into account include the content, links,
that helps analyze SEO, content, a market, web ad- spam information, the frequency of updating the web
vertising, and social networking. It gives a percent- portals under concern. The credibility score is deter-
age bound to each item where the important mistakes mined using metrics such as page rank, blacklist sta-
fixed and the warnings given are said in terms of a tus, average page hits, and two key quality factors: ,
percentage. namely credibility and relevance. A web crawler col-
Simunic et al. conducted a study on SEO factors lects page links from different websites in response to
to improve internet visibility [10]. The research in- specific queries and in the end, presents the Web sites
volved literature analysis and an analysis model. The in order of reliability so that users can easily deter-
study found no significant correlation between web- mine which sources contain reliable information.
site quality and search query points. However, there The Deep Cyber Threat Situational Awareness
were qualitative deficits in variable optimization. The Framework (DCTSAF), developed by Soman et al.
study suggests that detecting and optimizing impor- uses deep learning to detect malicious domains and
tant Google factors for ranking can lead to higher on- URLs [15]. Traditional methods like blacklisting
line direct sales, enriching new scientific data knowl- and signature-based strategies don’t work, especially
edge and creating potential for further research. when faced with fresh or more sophisticated threats.
The authors Considers the rise in traffic from ser- Deep learning is the foundation of the framework,
vices like HTTP, FTP, and SMTP, network traffic which uses character-level embeddings to work with
classification is crucial for monitoring and managing the raw data and automatically determine the most rel-
data flow in networks [11]. In this work, the perfor- evant features. The use of both hierarchical features
mance of using different sorts of supervised learn- and long-distance linkages in domain names/URLs
ing algorithms: Ensemble learning, Decision tree makes CNN-LSTM networks better than other mod-
and Bayesian classifiers for detecting network traffic. els. The system is extremely powerful and can handle
When applying the wrapper method in feature selec- two million events per second, which leads to early
threat detection and an alert. 3.5 Load Time
Malathi et al. pointed out In the current genera-
tion to form social network they must family social The loading time for this website was determined by
network site such as Linked In, Google Plus and face Pingdom Tool [0]. Namely, the tests were performed
book [16]. Social networks can be represented by from an Asia-Japan-Tokyo server to consider the load
graphs and their most significant figures, connections, effect in this region. when using Pingdom, which de-
and interaction. Other uses of graphs include the rep- livers specific results based on several aspects such as
resentation of relations and processes of different sys- server response time, resource loading time, and total
tems, chemical and physical, biological, information page loading time. This assists in finding out some of
as well. This work examines a database of US politics the constraints in performance that might hamper on
books as a network with such elements as between- user experience within certain espoused zones.
ness, eigen vector, degree and closeness.
3.6 Traffic Analysis
3 EXPERIMENT For the assessment purpose of monitoring website
traffic and its behavior index, the SEMrush Tool [0]
In conducting the analysis of website perfor- was used. It also offers additional traffic details such
mance, several industry-standard tools were used as the visits, the visitors, the bounce rates and the av-
to gather insights across key metrics such as Per- erage session time. The following metrics were ana-
formance, SEO, Accessibility, Best Practices, Load lyzed:
Time, and Traffic
To assess performance, SEO, accessibility, and 3.6.1 Total Visits
best practices, Google PageSpeed Insights [0] was
used. This tool provides detailed insights into the The over all the number of time the site was visited in
speed characteristics of a website, scoring and of- the given time of the study.
fering suggestions for improvements on both desktop
and mobile versions. 3.6.2 Unique Visitors

3.1 Performance The number of people who visit this website without
repetition of any user.
This metric complies with how quickly and active the
website is once opened. The tool breaks down fea- 3.6.3 Purchase Conversion Rate
tures such as response time of the server, resource
loading and rendering speed. This was measured wherever possible showing the
rate which visitors made a purchase or achieved a de-
3.2 SEO sired end result.

SEO score is defined as the capability of the website 3.6.4 Pages per Visit
to appear in search engines results and evaluate meta-
data, mobile-friendliness, and structured data. The quantity of unique web page visits by each user
on average per session.
3.3 Accessibility
3.6.5 Average Visit Duration
This metric deals with the accessibility of users with
complications for example; this measures the number The specific time that the users stay connected to the
of images that have the alt text; users who can only site/ portal.
use the keyboard to get round the website; users who
have problems with color perception. 3.6.6 Bounce Rate

3.4 Best Practices The rate at which visitors leave the site after visiting
only one of its pages.
Google PageSpeed Insights assess the current state of GTmetrix [0] was also used, offering additional
website according to modern web technologies and performance insights. The tool includes metrics such
standards like security, response, and coding. as GTmetrix Grade, Performance Score, Structure
Score, Largest Contentful Paint (LCP), Total Block- lost due to a specific resource. The site’s TBT values
ing Time (TBT), Cumulative Layout Shift (CLS), To- must be lower since this represents a more responsive
tal Page Size, and Total Number of Requests. website.

3.7.1 GTmetrix Grade 3.7.6 Cumulative Layout Shift (CLS)

This is a composite score that reflects the overall per- CLS gives the total scrambled layout/flush that hap-
formance and structure of your website. It’s broken pens inadvertently when you load a site. For instance,
into two components: if images or ads take time to load that they across text
and make it shift about it will contribute to the CLS
Performance Score (70% of the grade) Reflects score. A CLS score below 0.1 guarantees that users
how well the website performs based on loading have reliable and consistent client-side experience.
speed and interactivity, derived from Core Web Vitals
3.7.7 Total Page Size
metrics.
This metric demonstrates the sum of the sizes of all
Structure Score (30% of the grade) Measures files required for rendering the webpage – images,
how well your site follows best coding and optimiza- scripts, stylesheets and others. Factoring for larger
tion practices to ensure faster load times. page sizes is that greater page size decreases the load-
ing time, particularly for internet users with low band-
3.7.2 Performance Score width. Therefore the best page size is derived from
minimizing its resources since a decrease in page size
This is the segment of the GTmetrix Grade that looks has a positive effect on page performance.
at how effectively users can access and interact with
what your website has to offer. While it also incorpo- 3.7.8 Total Number of Requests
rates data from other performance indicators such as
Core Web Vitals to provide a transparent indication of The number of HTTP requests or browser clicks an
user experience scores. HTML page makes to load all required elements,
including images, java, css, and fonts, is shown by
3.7.3 Structure Score this measure. The longer it takes for a website to load,
the more requests the page receives. The website
This metric emphasizes various aspects of a website’s loading time can be extended if script requests can be
architecture or design. Often it shows how well your made more frequently or at a later time.
site can actually perform and details like too much
JavaScript, ineffective CSS, and uncompressed im- Several objective tools were chosen because they
ages discover. The greater the structure score, the are universal and can give more or less exhaustive
more your site prepares appropriately regarding load- information about various aspects of a website
ing speed and maintainability. taking into account technical parameters and users’
experiences. These metrics provide a birds-eye
3.7.4 Largest Contentful Paint (LCP)
view of the website and this not only considers
such factors as speed of loading of the website,
LCP determines the page load time on the screen tak-
ease of use of the website, availability of Search
ing the biggest factor that is the size of the largest vis-
Engine Optimization and identified best practices but
ible content element (image, video or large text block)
also factors such as structure or organization of the
into consideration. This is an essential UX value, as
website, responsiveness of the website, and optimum
increased LCP can give a user a sensation that a site
use of resources amongst others. In any case, it
is slow. Google advises LCP should happen within a
integrates supports the identification of specific areas
time frame of 2.5 seconds from the page time.
of improvement, providing a fair assessment based
3.7.5 Total Blocking Time (TBT) on both, technology and users.

TBT quantifies the time the browser is occupied by

tasks that hinder or completely halt user interaction 4 EXPERIMENTAL RESULTS
(such as script loading). It measures the time a web-
site’s JavaScript or any other resource takes to hamper Analyzing the performance of four job board web-
interactivity and quantify the amount of interactivity sites, their strengths and weaknesses differ signifi-
 Table 1: Performance analysis of various webpages
Job Board-1 Job Board-2 Job Board-3 Job Board-4
Performance 100 64 60 93
SEO 100 85 100 77
Accessibility 100 90 90 75
Best practices 74 96 100 93
Load Time 767ms 3.30 s 853 ms 112 ms
Table 2: Performance and Structure Metrics of Various Job Boards as Analyzed by GTmetrix
Job Board-1 Job Board-2 Job Board-3 Job Board-4
GTmetrix Grade A (91%) C (76%) D (61%) C (78%)
Performance Score 89% 71% 39% 76%
Structure Score 93% 83% 93% 82%
Largest Contentful Paint 770ms 686ms -83ms 4.5s +3.7s 3.0s +2.2s
Total Blocking Time 260ms 1.0s +775ms 1.2s +919ms 172ms -88ms
Cumulative Layout Shift 0 0.01 +0.01 0 0.01 +0.01
Total Page Size 585KB 3.98MB +3.41MB 2.43MB +1.86MB 2.98MB +2.41MB

cantly. However, Job Board-1 stands out by excelling whereas other platforms, such as Job Board-4, Job
in multiple categories, performing superbly with high Board-3, and Job Board-2, scored above 90. When
scores in traffic, audience engagement, and easy ac- considering visits, Job Board-1 led with 1.8 billion
cess. It claims the highest visit count, unique visi- visits, followed by Job Board-2 with 43.9 million, Job
tors, and pages per visit, which are strong indicators Board-4 with 29.5 million, and Job Board-3 with 7.1
of higher user engagement and a lower bounce rate. million. Engagement rates show that Job Board-1 had
Job Board-2 receives an average number of visitors the highest page views per visit at 8.8, with an average
but experiences slow loading times, which may lead visit duration of 12 minutes and 1 second. However,
to potential user loss. Job Board-4 shows good per- Job Board-3 had the highest bounce rate at 58.41%,
formance in terms of accessibility and compliance, indicating that users leave the site quickly.
but overall visits have decreased, which could be im- Fig 1 represents the traffic analysis of Job Board-
proved by enhancing SEO and reducing load times. 1, Job Board-2, Job Board-3, and Job Board-4, reveal-
Job Board-3, the smallest platform by traffic, boasts ing distinct performance metrics. Job Board-1 leads
the highest accessibility and best practice rates. How- significantly with 1.88 billion visits (up 3.62%) and
ever, it has a relatively high bounce rate and lower 423.6 million unique visitors (up 6.44%). Job Board-
interaction rates, indicating room for improvement in 2 follows with 43.9 million visits (up 9.14%) and 24.7
user experience (UX). The findings from this analy- million unique visitors (down 4.49%).
sis highlight the importance of technical performance,
search engine optimization, and user-centric metrics Table 2 provides a comparative analysis of four
for driving effective traffic flow. job boards (Job Board-1 to Job Board-4) based on
performance and structure metrics from GTmetrix,
According to the data presented in Table 1, the covering six key aspects: GTmetrix Grade, Perfor-
following important aspects of website performance mance Score, Structure Score, Largest Contentful
can be evaluated. Regarding performance, both Job Paint (LCP), Total Blocking Time (TBT), Cumulative
Board-1 and Job Board-3 posted faster loading times, Layout Shift (CLS), and Total Page Size. Job Board-1
allowing customers to engage in quicker interactions, achieved the highest overall performance with an ’A’
unlike Job Board-2, which took 3.30 seconds to load. GTmetrix Grade (91%), the best Performance Score
The overall SEO report is almost perfect for most (89%), and a strong Structure Score (93%). It also
sites, with Job Board-1 and Job Board-3 scoring 100 had the fastest LCP (770ms), minimal blocking time
points, while Job Board-4 and Job Board-2 scored rel- (260ms), and the smallest page size (585KB), indi-
atively lower. In terms of accessibility, Job Board- cating quick load times and efficient resource usage.
1 achieved a perfect score of 100%, indicating full In contrast, Job Board-3 performed the worst, with
compliance with accessibility standards for disabled the lowest GTmetrix Grade (’D’, 61%), the slowest
users, while Job Board-4 received the lowest score. LCP (4.5s), the highest TBT (1.2s), and relatively
Job Board-3 performed moderately well. For best large page size (2.43MB). Job Board-4 also lagged
practices, Job Board-1 had the lowest score of 74, in some areas, but performed well in blocking time
 Figure 1: Traffic analysis of various platforms

while Job Board-1 has the lowest at 41.11%. Over-

all, Job Board-1 outperforms its competitors across
all key metrics, highlighting its effectiveness and user
engagement.

5 CONCLUSION
In conclusion, analysis indicates that Job Board-
1 surpasses other job platforms in performance, user
engagement, SEO, and accessibility, positioning it as
Figure 2: Visualization plot for website analysis a benchmark job platform. Another popular plat-
form, Job Board-2, exhibits slower loading times,
(172ms). Overall, Job Board-1 stands out as the best- suggesting potential improvements in user experi-
ence. Job Board-4, recognized for strong accessibil-
performing website in terms of both speed and user ity and adherence to best practices, may enhance traf-
experience. fic through targeted SEO optimizations. Job Board-
Fig 2 presents a comparison of key performance 3 ranks highly in accessibility and best practices,
metrics across five different platforms, focusing on though its bounce rate remains elevated, with moder-
ate user engagement relative to larger platforms. Con-
areas such as Performance, SEO, Accessibility, and sistent with prior findings, this study emphasizes that
Best Practices. Two platforms show the highest per- a balance of efficiency, user satisfaction, and findabil-
formance scores, reflecting quicker load times, while ity is essential for the success of online platforms.
one lags behind with the lowest performance score
and a longer load time of 3.30 seconds. In the SEO
category, two platforms achieve a perfect score, in-
dicating strong optimization, while others have room
REFERENCES
for improvement. Accessibility is another area where
one platform excels with a perfect score, suggesting Nanda, J.K., Hay, J.L. and Marchetti, M.A., 2021.
full accessibility for disabled users, while another has Analysis of keywords used in Internet searches for
the lowest score. Regarding best practices, three plat- melanoma information: observational study. JMIR
forms consistently score high, but one unexpectedly dermatology, 4(1), p.e25720.
has the lowest score in this category. The chart of- Afroz, S., Riyazuddin, Y.M. and Jadda, V., 2023,
fers a visual comparison that highlights strengths and December. Website Traffic Trends and Performance
areas for improvement, particularly in SEO and best Evaluation of Selected Consumer Electronic Com-
practices for some platforms. pany Websites. In 2023 Global Conference on Infor-
Job Board-4 sees a decline in visits to 29.5 mil- mation Technologies and Communications (GCITC)
lion (down 13.36%) and 13.1 million unique visitors (pp. 1-4). IEEE.
(down 5.61%). Job Board-3 shows modest growth Hossain, M.T., Hassan, R., Amjad, M. and Rahman,
with 7.1 million visits (up 2.38%) and 3.5 million M.A., 2021. Web performance analysis: an empirical
unique visitors (down 3.63%). In terms of engage- analysis of e-commerce sites in Bangladesh. Interna-
ment, Job Board-1 has the highest pages per visit at tional Journal of Information Engineering and Elec-
8.8 and an average visit duration of 12:01 minutes. tronic Business, 11(4), p.47.
Job Board-4 and Job Board-2 show increases in pages Palacios-Zamora, K., Cordova-Morana, J., Mendoza-
per visit, while Job Board-3’s metrics decline. The Cabrera, D. and Pacheco-Mendoza, S., 2023. Mea-
bounce rate is highest for Job Board-3 at 58.41%, surement on University Websites: A Perspective of
Effectiveness. JOIV: International Journal on Infor- Soman, K.P., Poornachandran, P. and Menon, V.K.,
matics Visualization, 7(3-2), pp.1995-2006. Vinayakumar R Center for Computational Engineer-
Dobbala, M.K. and Lingolu, M.S.S., 2022. Web Per- ing and Networking (CEN), Amrita School of Engi-
formance Tooling and the Importance of Web Vitals. neering, Coimbatore Amrita Vishwa Vidyapeetham,
Journal of Technological Innovations, 3(3). India vinayakumarr77@ gmail. com.
Bernine, N., Nacer, H., Aissani, D. and Alla, H., Malathi, A. and Radha, D., 2016, October. Analy-
2020. Towards a performance analysis of composite sis and visualization of social media networks. In
web services using Petri nets. International Journal of 2016 International Conference on Computation Sys-
Mathematics in Operational Research, 17(4), pp.467- tem and Information Technology for Sustainable So-
491. lutions (CSITSS) (pp. 58-63). IEEE.
Google PageSpeed Insights
Kumar, N., Kumar, S. and Rajak, R., 2021, Decem-
ber. Website Performance Analysis and Evaluation SEMrush
using Automated Tools. In 2021 5th International Pingdom Tool
Conference on Electrical, Electronics, Communica- GTMatrix
tion, Computer Technologies and Optimization Tech-
niques (ICEECCOT) (pp. 210-214). IEEE.
Akgül, Y., 2024. Evaluating the performance of web-
sites from a public value, usability, and readability
perspectives: a review of Turkish national govern-
ment websites. Universal Access in the Information
Society, 23(2), pp.975-990.
Simec, A. and Križanić, M., 2022. Comparison of
tools for analyzing the degree of optimization of web-
site search engines.
Šimunić, M., Stifanich, L.P. and Car, T., 2023. Hotel
web site SEO analysis: Segmentation and valoriza-
tion as a precondition for discovering and understand-
ing insights to improve online visibility. Ekonomski
vjesnik/Econviews-Review of Contemporary Busi-
ness, Entrepreneurship and Economic Issues, 36(2),
pp.299-311.
Archanaa, R., Athulya, V., Rajasundari, T. and Kiran,
M.V.K., 2017, January. A comparative performance
analysis on network traffic classification using super-
vised learning algorithms. In 2017 4th International
Conference on Advanced Computing and Communi-
cation Systems (ICACCS) (pp. 1-5). IEEE.
Sujee, R. and Padmavathi, S., 2021, October. Per-
formance Analysis and Prediction of Students Re-
sults Using RBF Networks. In 2021 2nd International
Conference on Smart Electronics and Communication
(ICOSEC) (pp. 333-338). IEEE.
Meghwal, J., Joshi, K., Chaparwal, N. and Rajput,
P.S., 2022. NIRF Ranking 2021: A Webometric Anal-
ysis of Top 10 University Websites of India. In-
ternational Journal of Research in Library Science
(IJRLS), 8(2), pp.191-205.
Subbulakshmi, S., Gopika, A. and Thomson, L.,
2019, July. Enhanced Ranking of Websites Based
on Credibility Evaluation. In 2019 International Con-
ference on Communication and Electronics Systems
(ICCES) (pp. 2035-2040). IEEE.