Drop us a line
If you are interested in the development of a custom solution — send us the message and we'll schedule a talk about it.
Azati Software
Promotional Site Generator for Insurance Corporation
The well-known insurance company wanted Azati to optimize marketing workflow: the process of promotional materials generation according to some internal data and real-time statistics.
CUSTOMER:
It is nearly impossible to establish an insurance company without a wide network of wholesalers and distributors that cover various regions and people groups. If a small company cannot offer some specific insurance policies by themselves, it can ask its partners for help.
Such an approach helps small companies cut down expenses and focus on marketing and sales. It is a typical situation when the client receives similar offers from several agents. Here, it is important to prepare marketing materials that can convince a client to sign the desired policy.
Often small companies ask their partners to help them prepare marketing documents for a specific audience. This process is a headache both for partners and local sales agents.
Our customer is the US national leader in healthcare liability insurance for medical and dental professionals, and hospital healthcare facilities. Azati was asked to optimize general presales workflow and the way managers help external companies prepare marketing materials and presentations.
OBJECTIVE:
The main aim of Azati was to analyze how sales agents prepare the documents and marketing materials to optimize the workflow. Before the customer turned to Azati, he already tried to enhance this process, but this try was unsuccessful.
As it was a pilot project, we did not involve a huge team during the early stages but relied on one business analyst and a system architect experienced in Liferay and TIBCO.
Our specialists discovered that the primary source of information the agents used to prepare materials was a corporate website that contained over a million pages. People manually searched data and extracted the necessary information in a non-optimal way.
It was a time-consuming process, as the existing search engine was not accurate enough to handle the very specific insurance queries. Sometimes it took more than an hour to find the required article, so people even created bookmarks that may be helpful for them.
Our team created an automated tool that finds helpful data, extracts statistics from corporate data sources, and builds small promotional websites where clients may find all the data he/she needs.
The team faced several challenges. Let’s have a look.
#1
CHALLENGE #1:Colossal amount of information
The customer relied on a well-known solution for enterprise portal management – Liferay.
Pages generated with Liferay most often were the starting point of customer interaction. Liferay provided some additional functionality like engagement metrics, cloud analytics, document and content management, etc. The customer’s database contained a hundred thousand various documents, and people could view a considerable amount of these documents online.
The fact that we already developed several applications in the insurance domain simplified the workflow. Our team had already learned most terms, abbreviations, and basic insurance processes.
When specialists analyzed a considerable amount of documents, they figured out that these documents had a common structure, but though they are different in detail.
It means that it is possible to create a unified presentation structure, which can be customized using various content blocks.
CHALLENGE #2:Distributed data
When people make a decision, they rely on proven facts and data. Every person wants to see a personalized offer, not only the general information about the insurer.
As the customer operated across the US, insurance data and statistics were different for each state and target audience. But there were remote corporate databases, which contained the necessary information.
Liferay, by itself, provided a huge number of different metrics. There was info collected and processed via third-party applications. Managers collected and processed all the data manually, so now you may understand how time-consuming was the presales process.
As the team had already created complex data warehouses, so it was easy for us to create additional middleware, that acted like the ETL layer. This middleware extracted the required records from external databases, transformed these records into a suitable format, and loaded the data into Liferay.
#2
#3
CHALLENGE #3:Permission management and data security
Insurance information is considered as sensitive data. The customer used Liferay not only as enterprise CMS but also as an integration platform. It means that the data we have loaded into Liferay are available for some third-party applications and external vendors.
As we already mentioned, managers collected all the information manually as they had the right permissions to operate with these data. We could not provide all the records we aggregate to local sales agents as they may disclose private information to someone.
As the majority of the marketing materials were similar, the team decided to move presentations to the Internet, where we can automatically generate webpages with the necessary information via Liferay.
We can ensure that sensitive information is not accessible to the third-party. But sales agents can still generate a webpage, which already contains valuable insights extracted from the initial data.
CHALLENGE #4:Performance Monitoring
The customer wanted to know what distributors do with the marketing materials. So he asked us to create a basic analytics system with several dashboards to benchmark key performance indicators.
The customer provided access to local resellers, so they can learn how marketing materials are performing: did a client look through the proposal, how much time did he spend on a webpage, a number of entries, bounce rate, information about devices, etc.
Some features were already built into Liferay, while others the team developed by themselves. There were several open-source analytics systems (like Piwik), where all the required metrics are already implemented. Engineers leveled up their knowledge and picked some algorithms from there.
#4
PROCESS:
The client asked Azati take part in this project, as our teams had already completed several projects for the customer. We already proved our Java expertise in general and expertise in Liferay in particular.
At that moment, our primary focus was a third-party services integration. Engineers delivered several portlets (this is how tiny web apps in Liferay are called) that pulled information from third-party platforms and loaded these data into the content management platform.
Our business analysts and system architects were familiar with the internal application structure and ecosystem. So it was easy for us to develop additional features without breaking the workflow.
As the internal development team had already failed this project, the customer was not very optimistic at the beginning. But we convinced him to give our engineers a shot.
We involved two people: experienced business analyst and a system architect. BA collected primary requirements from the customer and the end-users – local sales representatives. At the same moment, the architect analyzed the ways we can implement the required features.
After the team finished the preparation phase and presented the plan to the customer, we got the green light. Azati involved two more Java engineers and an HTML developer.
Business analysts split the entire project into several milestones and six sprints, so we rolled out the first version of the project in a month or so.
SOLUTION:
Liferay is a flexible enterprise portal engine, that can be customizable with various portlets. These modules written in Java can operate both with Liferay Core API and with external services.
According to the initial analysis, the team created many blocks that could be arranged and combined into various webpages. These blocks relied on internal data, and the information extracted from remote databases.
The team also created several portlets:
Portlet for Remote Data Extraction
This portlet acts like the ETL layer. It extracts data from external sources of data once per day – the start was scheduled at 11.30 PM ET. Portlet makes requests to external data providers, processes the results, and places the processed records into the database with Export / Import Liferay API.
Gateway for Third-party Applications
The team created a small portlet, which simplifies the extraction of data used for building interactive reports. This portlet operated with Reports Engine Liferay API, and provided data output in XML and JSON formats. Any BI software or widgets can receive the required data from desktop, web or mobile.
Microsite Generator
Liferay Site API powers this portlet. Engineers created a specific page where distributors may generate a microsite according to the set of parameters. The user fills several fields and receives a functional microsite – a set of static pages with some interactive widgets.
Some white-label capabilities are already built-in. The user could customize the general look, logo, several headlines, colors, structure, minor text blocks, a call to action text. All the statistical information is fetched from Liferay, so the information is always up to date.
Performance Dashboard
This portlet displays usage statistics and key performance indicators extracted from metrics that are streamed from microsites. Every visitor may log in using its username and password to check how the webpages he generated perform. The team also created several master accounts for managers.
TECHNOLOGIES:
SCREENSHOTS:
This is an example of the auto-generated website.
Results:
The development team created several small modules (portlets) and plugged these modules into Liferay. As soon as we launched the project, the customer noticed that this tool became increasingly popular among both local sales representatives and the end-clients.
The customer received instant positive feedback from agents and agencies, as they were almost freed from doing repetitive work. This solution also increased sales in the long-term run, as end-clients operated with the right data and statistics.
A FEW NUMBERS:
~10.000
microsites generated in two years
+4%
growth in B2B sales
5
times less time it takes to prepare basic sales materials
Azati designed, developed, and delivered in time yet another application, that eliminated one specific business issue and positively affects business workflow. Our team also leveled up their knowledge in building analytics software.
We had already created several solutions for this customer, so if you are interested – check out our portfolio.
Featured case studies:
JavaScript, Ruby
HR Planning SoftwareThe customer asked Azati to audit the existing solution in terms of general performance to create a roadmap of future improvements. Our team also increased application performance and delivered several new features.
Python
Stock Market Trend Discovery with Machine LearningAt Azati Labs, our engineers developed an AI-powered prototype of a tool that can spot a stock market trend. Online trading applications may use this information to calculate the actual stock market price change.
Python
Semantic Search Engine for Bioinformatics CompanyAzati designed and developed a semantic search engine powered by machine learning. It extracts the actual meaning from the search query and looks for the most relevant results across huge scientific datasets.
Java, JavaScript
E-health Web Portal for International Software IntegratorAzati helped a well-known software integrator to eliminate legacy code, rebuild a complex web application, and fix the majority of mission-critical bugs.
JavaScript, Ruby
Custom Platform for Logistics and Goods TransportationAzati helped a European startup to create a custom logistics platform. It helps shippers to track goods in a real-time, as well as guarantees that the buyer will receive the product in a perfect condition.
