If you’re a Malayali in the Indian state of Kerala, chances are you’ve visited Manorama Online to read about current affairs or even to check the latest movie listings. The company—and its corresponding web assets—represent the digital branch of Malayala Manorama, a large Indian media conglomerate that specializes in local content for Malayalis. About 30 percent of manoramaonline.com is news; the remaining 70 percent includes content such as entertainment, classifieds, lifestyle services and digital subscriptions.
In 2016, Manorama embarked on a project to develop an in-house analytics pipeline that could unify enormous amounts of raw data from multiple web domains and convert it into meaningful insights. The company currently has 10 major domains, including its matrimonial and real estate sites and it has plans to further expand its digital footprint.
Manorama Online was an early cloud adopter, beginning its journey with Amazon Web Services
(AWS)
in 2011.
Like many media companies in India, the business was also using the free version of Google
Analytics. This
tool served basic analytics needs, but Manorama became frustrated with its limitations—it
analyzes only
data samples for a particular period and was incapable of processing large datasets.
HiFX, an AWS Partner Network (APN) Select Consulting Partner, has been Malayala Manorama’s
technology
partner for more than 20 years and was approached to design its new analytics pipeline,
which it
completed
in late 2018. The Lens analytics pipeline currently handles about 1 TB of data per day and
can
process up
to one million events in less than a minute, with an average of 300 million events processed
per
day. The
horizontally scalable architecture of Lens allows for more events to be processed.
Prior to Lens, Manorama struggled with real-time analytics, a vital activity for news
organizations. Now,
the
API turnaround latency for most workloads is just 12.6 milliseconds. Video editors can now
see
in real
time how
users are engaging with content on the website and make immediate adjustments to an
article’s
title or
publish
videos in different sections to attract more traffic. They can also track the number of page
views per
visit,
the average time spent on each page and the number of quality views, as well as
differentiate
between
unique
and repeat visitors.
For its classified services—especially on its matrimonial site for individuals seeking a
partner—Manorama’s
telemarketing team is able to better segment and target certain communities or religions.
“Using
the Lens
dashboards and recommendation engine, we are able to get a 360-degree view of our customers
and
their
interests
and this helps us to improve our operational efficiency, conversion rates and revenues,”
says
Renjit
Paul,
deputy general manager of IT at Malayala Manorama. The company uses Lens to measure the
effectiveness of a
telemarketing campaign and to get more site visitors to sign up and subscribe for its
matrimonial
services. They
found that conversion rates increased 13 percent with Lens compared to similar previous
campaigns.
As a long-time user of AWS Business Support, Manorama upgraded to AWS Enterprise Support at
the
start of
the
project. “We have been receiving tremendous help from the Enterprise Support team. We are
able
to consult
with
them on our business plans and technical challenges along the way and get real-time
feedback,”
says Paul.
“Before the business embarks on any project, we always work with the AWS Enterprise Support
team
and take
them
through our technical design and implementation plan. They know our existing stack and our
requirements,
which
helps them guide us in the right direction, at times using new services or apprising us of
service
roadmaps.”
Manorama also follows the AWS Well-Architected Framework to ensure that all applications are
designed
properly
for optimal resource utilization and performance.
“Lens is quite a revolution in the media space,” says Mohan Thomas, President of Technology
at
HiFX. The
system
handles both data streaming and batch analytics and it includes a centralized dashboard
accessible across
all
Manorama divisions. “Coming up with a real-time dashboard was not easy, but we found a
solution
using
Amazon
Kinesis, Spark Structured Streaming on Amazon EMR and Amazon Elasticsearch Service,” he
says.
“We were
then
able to show data on concurrent users active on the site—which is a variable process—in less
than 5
seconds.”
The company uses the Spark engine on Amazon EMR for batch data processing, after which data
is
written to
an
Amazon Elasticsearch Service and Amazon Redshift component for dashboard purposes. Before
this
takes
place, data
enters through an Amazon Kinesis stream and is cross-checked for errors. The data is then
written to a
data lake
primarily formed with an Amazon Simple Storage Service (Amazon S3) bucket. Only then does
processing and
analysis begin. “AWS recommended that we use Amazon S3 for the data lake and that saved us
quite a lot of
money
and effort,” Thomas adds.
Marketing teams have also used Lens to boost application traffic and continually assess their
efforts.
Less than two months after launching the pipeline, Manorama saw the daily usage rate of its
mobile
application for matrimonial classifieds increase by 52 percent. Editors can continually
tweak
the mobile
app and the main Manorama application to bring the features that people are using most to
the
forefront or
landing page. On its websites, the company has seen 16 percent more concurrent users since
introducing
Lens and is targeting a 50 percent increase in 2019. “We consider this an effective
measurement
of our
digital marketing efforts,” says Paul. “Furthermore, we can validate our marketing data to
make
sure we
are not losing money or resources on unnecessary campaigns.”
Data security was a nagging concern with the previous free analytics tool, because Manorama
stores a lot
of personal data on its users. With Lens, security concerns have disappeared. In addition,
Manorama can
integrate unique and confidential data into its dashboards, as well as report functions for
complete
visibility into individual usage patterns.
Looking ahead, the company hopes to boost subscriptions to its websites with the introduction
of
a
content recommendation and personalization engine, which is currently in development using
Amazon
SageMaker. Plans are also underway to use Lens for analyzing usage patterns in different
countries to
better cater to those markets. Increased user engagement could also lead to more advertising
revenue and
further increase sales conversions.
Overall, says Paul, “We are very impressed with AWS products and solutions and the pace at
which
new
features and services are added to the platform. We get complete visibility into upcoming
services and the
future roadmap as part of the AWS Enterprise Support and that’s something we haven’t seen
with
any
competitor. The support is top-notch and we see how passionate AWS is to make you
successful.”