Try : Insurtech, Application Development









Dev Ops(2)

Enterprise Solution(13)




AI in Insurance(21)


Product Innovation(28)


Augmented Reality(7)

Customer Journey(7)


User Experience(19)

Artificial Intelligence(81)



Cognitive Computing(6)

Computer Vision(5)

Data Science(12)


Intelligent Automation(19)

Machine Learning(42)

Natural Language Processing(5)

Pushing the Envelope on ICR Accuracy in Hand-written Forms

Himanshu Saraf
5 minutes, 6 seconds read

The need for and consequently the number of solutions for reading hand-written forms in an automated manner has been on a rise for as long as one could remember. Almost all businesses to varying degrees utilize paper-based forms that are filled by customers by hand. Most if not all of these businesses convert this handwritten information into the digital format. Depending on the technological sophistication or the size of the business this digitization might be done manually by one or more data entry specialists or through an automated solution. 

It’s easy to see how the manual route may not be an ideal solution for medium or large-sized business. Some of the apparent drawbacks of manual document processing are:

  1. The cost of having data entry specialists quickly add up as more documents need to be digitized necessitating adding more resources.
  2. Manual data entry is a slow process.
  3. Manual data entry is error-prone and requires a quality inspection which is costly and not fail-proof.

Many businesses have realized this and have transitioned to some form of a partially or fully automated solution to this problem. However, it’s not all rosy for these businesses either. The problems these businesses face is primarily related to the accuracy of the current solutions in the market. 

Shortcomings of Existing Hand-written Document Processing Solutions

The industry average for ICR (Intelligent Character Recognition) accuracy at the character level is about 70% and it will drop significantly if measured at word level which is what matters at the end. Such automation may allow for reducing the number of data entry personnel but with such a low level of accuracy, there will be a need for increased quality check resources, which are often more expensive than data entry resources hence diluting the cost-benefit of automation. Moreover, since the quality check is a slower process than data entry, this kind of automation doesn’t even address the speed problem.

Some of the reasons that result in a low level of accuracy among existing document processing solutions are:

  • Poor form design
  • User input not in line with the format
  • Noisy images
  • Misaligned documents
  • Low-quality scanning of documents
  • Spelling mistakes by the user
  • Overwriting/corrections by user

While we may not have control over some of the above factors such as form design and user input, we can definitely improvise the data extraction models to account for the other factors such as image noise, misalignments, spelling mistakes etc.

Our ICR Solution

The Document Parser solution in FlowMagic provides an intuitive user interface where data can be extracted from any standard form in three easy steps:

Step 1:   The user annotates the form (this is a one-time exercise for each new form) using an easy and intuitive UI. During annotation, each input field can optionally be labelled as mandatory. The user can specify the datatype for each field as alphabets, numeric or checkbox and also set the context for the field e.g. Name, PAN, City, Car Make, Date etc. Once done, the saved template can be used repeatedly for reading forms of the same type as long as there are no changes in the form design. In case of a change, the saved template can be easily modified. 

Step 2:   The user uploads one or more forms and chooses the corresponding template (from previous annotations). The system automatically extracts data from the forms.

Step 3:  The system exports the output in CSV, XML or JSON as desired by the user. If any field was marked as mandatory during annotation, the system also outputs a list of all mandatory fields that are blank.

Salient features of ICR Document Parser

  1. The standard form being annotated can be any number of pages. The input form need not have the same number of pages. If there is a mismatch between the pages in the input form and the template, the system does a matching and runs the data extraction on matching pages only. This also means that the input form need not be sorted correctly.
  2. The system can read handwritten as well as printed forms.
  3. The system corrects for minor misalignments during scanning of documents or documents scanned in the wrong orientation.
  4. The system has inbuilt dictionaries for various contexts such as Name, Cities, States, Countries, PAN, Profession, Marital Status, Relationship, Amount, Car Make, Date, Gender.
  5. The various data types supported by the system are alphabets, numeric, alphanumeric, checkboxes and special characters.
  6. The system corrects user errors or scanning issues by performing data type and dictionary checks (see examples below).
  7. The system checks for mandatory fields to make sure the form is completely filled.

Examples of Data Read/Corrections Made by an ICR

Benefits of ICR

Flexibility – you can annotate a wide variety of forms with complex inputs and data formats using the multiple data types and contexts built into the system.

Speed – Both annotation and data extraction are very user-friendly and fast. The system can extract data from a five-page form in under 30 seconds.

Scalability – The system is highly extensible and once set up for one type of form can easily be scaled for multiple forms or to process documents in bulk of the same format.

Accuracy – The character level accuracy of our model is over 90%. Word level accuracy depends on the form design and quality but in general, varies between 75% and 85%.


ICR (Intelligent Character Recognizer) workflow

No matter what solution you use, you can always benefit from these best practices for form design to improve the accuracy of your ICR:

  1. Have all instructions in bold at the top of the form.
  2. Instruct the user to write clearly in block letters as the form will be processed by a machine.
  3. Provide examples of how to enter data wherever there is a scope for confusion.
  4. Instead of providing a free form space for data entry, it provides a clearly marked space with a specific location to enter each character.
  5. The overall space should be large enough to contain the requisite data to avoid user writing outside of this space.
  6. Have enough separation between the space for two fields to avoid overlap.

To learn more about how FlowMagic can improve the accuracy and speed of your document digitization/Intelligent Character Recognition (ICR) or discuss your broader AI goals, please get in touch with us at


Knowledge thats worth delivered in your inbox

Insurance consumers around the globe are seeking convenience and expecting better customer experience. From millennials to Gen Z, with the agile connectivity, irrespective of the industry has numerous options to choose from. As the competition intensifies the insurance industry has to jump into the bandwagon of technovation in order to provide improved accuracy, cost-saving and excellent customer experience. 

Here is a list of the marketing trends in insurance that will prove to be a game-changer in the year 2020.

1. Robo Financial Advisors

According to a Business Insider Intelligence forecast, by the year 2020 Robo-advisers will manage investment products worth $1 trillion, which will spike up to $4.6 trillion by as early as 2022.

Robo advisors have been around for quite some time. In the year 2008, during the financial crisis, Jon Stein, a 30-year old entrepreneur launched “Betterment”, the first Robo-advisor. In recent years due to its low investment rates and data input based research results, it has increased in popularity. 

It is basically designed for the people who want to manage their finances with low management cost. Based on respective data inputs, the Robo-advisors offer any advisory services. 

The main purpose behind the making of the Robo-advisor is to bring the financial services to the wide range of population with lower investment cost as compared to the traditional human advisors., and are some applications of Robo-advisors.

Behind the scenes of the software of Robo-advisors are actual human beings who track the market regularly and adjust the algorithms based on the current market condition. Robo-advisors are a boon to the end-users as they can invest in direct plans of mutual funds without shelling any commission. However lack of personalization and one-size-fits-all products are the areas of improvement.

2. Data Integration: The Future of Marketing

IDC estimates that, by the year 2020, the digital cosmos will reach 44 zettabytes, further complicating the lives of marketing professionals.

Integrating data sources is vital for any company, whether B2B or B2C to successfully meet Customer Experience expectations thereby drive accelerated sales revenue.

With an integrated source of information, retailers can administer and optimise marketing through KPI’s, metrics and dimensions that would not have been possible with the separate source system. In order to upscale marketing operations, a connected viewpoint is essential to evaluate the campaigns, audiences, events and channels, and drive the strategic goals.

From an operational viewpoint, CRM solution provides the organization with new business and the ERP system allows to manage and drive businesses around obstacles. A good place to start with the data integration is by Integrating these two systems shall provide marketers and the organizational sales-force with vital information, that can be shared with the stakeholders.

3. AI-driven Copywriting

Artificial intelligence can create cancer combating drugs, control self-driving cars, defeat the best brains at incredibly complex board games, but one realm it can’t perform flawlessly is communicating.

To help solve the issue, Google has been feeding it’s AI with more than 11,000 unpublished books, including 3,000 steamy romance titles. 

Autoencoder, a type of AI network, uses a data set to reproduce a result (in this case copywriting) using fewer steps. Insurers can harness this AI capability to create sentences and suggest the best-optimised language to approach the customers.

AI copywriting is evolving to a whole new level. Google granted  €706,000 (£621,000) to the Press Association, to run a news service with computers writing localised news stories. AI with the help of human journalists can write up to 30000 news stories a month and scale up the volume of the stories that would otherwise be impossible to produce manually.  

“Skilled human journalists will still be vital in the process, but Radar allows us to harness artificial intelligence to scale up to a volume of local stories that would be impossible to provide manually. It is a fantastic step forward for PA.”

  • PA’s editor-in-chief, Peter Clifton 

4. Gamification of Insurance

At the nexus of marketing trends ranging from social networking to the IoT to behavioural science and wearable tech;  gamification is a powerful lever for insurers and insurance agents. It creates an enriching digital experience and customer-centric business model.

Gamification offers great potential value to the insurance business process in the realm of consumer engagement and customer experience. From millennials to Gen Z, it has emerged as a useful practice and effective means to target early technology adopters by:

  • Transforming mundane tasks into interesting and fun experiences that keep users returning.
  • Increases brand awareness, brand penetration and affinity.
  • Increase sales by educating customers about product suitability and guide them to buying the product.
  • Motivating people to act in areas of healthcare and wellness, safe driving, financial planning and sustainability.

Ingress and AXA redefined the world of gaming and advertisement. December 5th, 2014, Niantic Labs the creator of ‘Ingress’ partnered with AXA. In the game, AXA Shield was initially only obtainable from AXA Portals, leading you to AXA business locations in person.

5. Advanced AI Capabilities in Insurance

Innovation and technology are the next frontiers in the insurance industry. While automation and IoT are already a reality for insurance, with the advent of AI there has been a holistic approach to Insurance automation. With insurance leveraging AI, it has expanded its reach to more ecosystems than ever before. Deploying AI capabilities in insurance can help make smarter underwriting decisions, fraud detections, risk assessment and create a better customer experience.

AI is driving significant change in business with insurance being no exception. It has the potential to enhance the insurance business model by-

  1. Improving the speed of the workflow: AI and RPA in insurance reduce redundancy of task. Automation of day to day tasks would reduce cost and time consumption thereby increasing accuracy, quality and competency.
  1. Customizing the services for better customer experience: One size no longer fits all, and the same goes for the insurance industry. With focus on individual markets, insurers can create niche usage-based products to sell the packages in a variety of ways.

Parag Sharma, CEO, Manta Labs and AI thought leader is going to speak about the Internet of Intelligent Experiences™: CX for the Digital Insurer at India Insurance Summit and Awards 2020 on March 12, 2020. Catch him live at IISA 2020.


  1. Providing new insights: Insurance is no guessing game. Data in silos is the biggest drawback for any industry. AI in insurance can integrate this data and provide analytics to help actuaries have a better insight while making a decision about a product.

Marketing Trends in Insurance: The Bottom Line

Today, at the core of marketing in Insurance, lies AI, Machine Learning and advanced data analytics to foster better experiences for the end-user. We’ve listed 5 most important trends that have the potential to shape marketing business models for Insurance and InsurTech firms. Be it Robo financial advisors or gamification, impressing customers remains the prime goal for Insurers.

Have thoughts and queries regarding upcoming marketing trends in Insurance? Please feel free to drop us a word at


Knowledge thats worth delivered in your inbox

Data Science is enormous. It brings forth a scientific approach to gather a massive amount of useful data from raw & disordered information (often collected from open sources). According to recent research, over 2.5 million terabytes of data appear daily. In 2020 every person produces 1.7 MB of data per second. Scientists, Analysts, and numerous other specialists use this data to derive decision-ready insights.

Using data science, marketers can get a clearer picture of their target audience. With this knowledge, any organization’s marketing department can formulate strategies to target customers who portray higher chances of conversion. Also, by delivering values, organizations can eventually maximize revenues. Going with the traditional methodologies, data processing can be a daunting task. Data Science offers a cost-effective solution to businesses seeking data-driven insights.

Let’s delve deeper into 5 most profitable and practical use cases of data science in marketing.

1. Budget Optimization

The primary goal of any marketer is to achieve the highest possible ROI from the allocated budget. This objective is undoubtedly difficult and time-consuming. On top of which, because of changing market dynamics and user preferences, strategies often go off the track leading to unanticipated outcomes.

Data science can be a saviour here. By analyzing the marketing department’s spending and acquisition ratio, organizations can build a model to distribute the budget in the smartest way possible. A clear picture will help marketers to invest money in the most relevant and surplus channels, thus optimizing key metrics.

2. Defining Audience Persona

While every marketer is familiar with the process of building the target audience portrait, determining the exact persona of the potential customer can still be a challenge. The lack of proper data insights might lead to ineffective advertiser decisions leading to a waste of resources.

Data science methods help marketers to understand the user persona and their preferred communication channels with data-driven insights. This means that the marketing budget will be spent on the right channels of influence, ignoring the irrelevant media, which a normal human being will think of covering for “just in case”. Such adjustment will inevitably increase the ROI and optimize the entire advertisement campaign. This will also retain brand relevance to the customers.

[Related: Your shopping cart just got a lot smarter!]

3. Brand New Social Media Marketing Strategy

Social media trends change faster than a human can track it. Facebook, LinkedIn, and Twitter define what is popular, and a marketer has to catch up with the trends.

Data science can keep you on track with the changing trends. Using the logic of Data Science in Marketing, one can get a bigger picture of what type of content people like interacting with. Data science allows us to gather and analyze data about people’s online behaviour. It provides the key metrics to adjust the SMM (Social Media Marketing) goals, which include – the time of posting, content type, amount, etc. These simple adjustments using data science insights can help increase the marketing ROI drastically.

4. Clearer Content Strategy

One of the biggest gaps between planning and execution that marketers face is knowing which channels will be affected and what kind of people will interact with their content and with what sentiment. Will be potential customers? Are interactors content gatherers? Are they the competition? Do they intend to ruin your reputation?

Knowing all this information will help streamline your content strategies.

As long as you know who your customers are; what are their perceptions about your brand; what information can attract/repel your customers; what social channels they are mostly active on; what are their sentiments with your content; what they usually do when they like or dislike a content; you’ll know what type of content you should produce.

For instance, some people hate emails, while others adore reading them. Some people want to resolve their queries publicly on social media, which some care about their online image. Data science can help achieve personalization to some extent, which can help humanize the conversations with your followers.

Let’s take another example of how data science in marketing can help stakeholders. It gives marketers insights about what phrases a customer would use while searching for a product/services online. Marketers can utilize this insight and prepare a content strategy that embeds these terms more often in your posts and articles.

Therefore, we can say that data science brings a variety of actionable insights about customer acquisition channels, their preferences, and engagement style, which can help plan content strategy accordingly.

5. Increasing Customer Loyalty

Your best customers are the ones who will not just purchase your product once but also will repeat buying and bring their friends and relatives to your store. Organizations realize that customer retention is easier than acquiring new customers.

But consolidating loyalty may be tricky. Data science can provide the marketing department with all the necessary information that can help boost customer loyalty. Based on purchase history and current search queries, analysts can predict their customer’s inclination towards a product. Accordingly, brands can create the most relevant offers for their customers. With personalized offers, existing customers feel special and will return to your brand and not go to the competitors.

The Essence of Data Science in Marketing

Using data science in marketing may ease the work of employees and uplift your strategies to new heights. We have to admit that the more structured information marketing teams have, the more effective their strategies become. At the core of any marketing efforts, data science can optimize cost for data processing and result in overwhelming conversion rates.

[Related: 5 Deep Learning Use Cases in Insurance]

About the Author: Marie Barnes is a writer for Bestforacar and an enthusiastic blogger interested in writing about technology, social media, work, travel, lifestyle, and current affairs. She shares her insights with the world through blogging. You can follow her on Medium.


Knowledge thats worth delivered in your inbox

Loading More Posts ...
Go Top

May i help you?

Our Website is
Best Experienced on
Chrome & Safari

safari icon