Globally, startups and large conglomerates alike have realised the value of data in realising business success. However, sifting through vast volumes of data and extracting insights from it in a timely manner is no easy task. This is where Taran comes into the equation. Founded in 2018 by three mathematicians, this startup is helping companies across 5 continents. By transforming their data with cutting edge solutions, Taran helps deliver a positive impact to businesses everywhere. This is evidenced by their flagship product, the Taran Decision Manager, used by financial institutions in several countries to reshape their processes and better serve their customers.
The origins of Taran
One of the most significant leaps in human innovation has been the ability to crunch massive amounts of data, sometimes in real-time, while organising it using pertinent rules to obtain previously unseen insights. Across industries, this phenomenon has accelerated decision making with an unprecedented degree of accuracy. In doing so, data is now the key that unlocks the door for businesses to streamline, optimise, and scale up their services.
However, startups and large enterprises alike struggle to harness the power of their data fully. Recognising this challenge across industries, a trio of Czech mathematicians, Martin Chudoba, Eva Sobková and Václav Kozmík, set out to tackle it. And so, Taran was born as a one-stop solution for all things related to data. The company began as a consultancy firm, guiding companies on implementing data-driven processes before moving towards product development.
Their first product was a recommendation engine known as TaRank. It’s a fully-customisable algorithm that can regularly process the data of millions of items daily and rank them accordingly. The system has since been utilised by the iPrice Group, an eCommerce aggregator with over 500 million items and is active in Singapore, Malaysia, Indonesia, Philippines, Thailand, Vietnam, and Hong Kong. By using the TaRank algorithm, iPrice saw a 20% increase in click-through rate on its items.
As time went on, Taran continued its consultancy and expanded its footprint across the world. Through these engagements, the company identified the growing demand for a flexible, robust, and customisable decision manager that could work in real-time. It was the spark that led to the creation of its flagship product – the Taran Decision Manager.
Understanding the Taran Decision Manager
In a nutshell, the Taran Decision Manager is a next-gen decision manager, which allows businesses to make data-driven decisions in real-time during customer interactions. The system, built on a modular architecture, is fully customisable and easy to deploy on top of any existing enterprise architecture. Hence, it’s a powerful software that’s allowing businesses to unlock the power of their data.
Taking the example of a person applying for a bank loan to illustrate how the system works, Managing Partner at Taran, Martin Chudoba, explained, “Firstly, the applicant would enter all of the data such as their income, employment status, etc. online or through the app. The bank would then have a series of rules based on their own data models to evaluate this data to decide whether to approve the loan application or not. Sitting in the backend, the Taran Decision Manager helps orchestrate this entire process from start to finish optimally. In doing so, allowing the bank to give an instant response to the applicant.”
By diving into the details of the loan application process, one can see how the Taran Decision Manager eliminates many limitations of similar products on the market.
Obtaining and processing vast amounts of data
One such challenge the Taran Decision Manager overcomes is connecting different data sources. “When the bank is deciding to approve the loan application, it’ll take data from other sources as well, such as the credit bureau. But connecting different data sources is no easy task, especially in South-East Asia where there are many emerging data providers,” explains Martin. But the flexible design of the Taran Decision Manager makes connecting different data sources a breeze.
But once you successfully connect different sources, you’re tasked with processing a colossal amount of data. One of the biggest banks in the Czech Republic, Česká spořitelna, found itself facing this challenge. It had launched an online credit application form, which collected digital data to prevent fraud. While it was only a few megabytes individually, it soon added up, and their existing infrastructure couldn’t keep up. By adding the Taran Decision Manager to their existing infrastructure, the bank could process this data quickly and speed up the approval process.
In capturing data from numerous sources, the Taran Decision Manager can ensure the full effectiveness of machine learning models. But, of course, there are costs associated with accessing some external data sources. The system minimises these costs by querying only the most promising cases and caching data requests to prevent unnecessary duplication. Thus, bringing valuable cost-savings to the use of digital tools for an organisation.
Flexible to meet the needs of startups and enterprises alike
The story of Česká spořitelna also serves as an example of the flexibility of the Taran Decision Manager. Further elaborating on this, Martin highlighted a key difference between fintech startups and established banks.
“A brand new fintech startup seeking to offer loans may not have a system in place to process the data and approve applications. Our system can easily step in and fill this void to help them make instant data-driven decisions.” This was the case with VKPay, a fintech lending platform backed by Russia’s internet giant Mail.Ru Group. Here the Taran Decision Manager serves as the foundation for its credit approval and reporting solution.
In contrast to a startup, an established bank like Česká spořitelna “Will already have its existing infrastructure in place. In such instances, our system would run atop the existing infrastructure, complementing it and optimising it for greater performance.” This level of flexibility was a conscious design choice by Taran.
Martin shared that when they looked at other similar systems, most were relatively rigid in their approaches. Hence, the Taran Decision Manager was built as a modular system on top of existing Python functionalities. Not only does this allow data scientists to test, deploy, and evaluate new machine learning algorithms, but also in a manner without disrupting existing infrastructure. This level of unparalleled flexibility allows organisations to freely explore and adopt approaches that work best to meet their business goals.
A bootstrapped startup with a global presence
With its powerful combination of features and flexibility, Taran has enjoyed glowing reviews that have propelled it to global success. In just three years, the startup has grown considerably. With a team of 15 skilled professionals, it has projects in 20+ countries across five continents, empowering clients across a spectrum of industries.
In addition to those listed above, Taran’s success stories include notable names such as the Rohlik Group – an online grocery unicorn in 3 countries, Wagamama – an international chain of restaurants, Fourth – a provider of software solutions for the hospitality industry. The collaborative efforts sparked by its mathematician founders have changed the way businesses make decisions. Their efforts were initiated independent of any investors and have remained so since the inception of Taran. This formula has certainly worked well thus far.
Looking ahead, Martin states the company will be solidifying its foundation and continuing its global expansion. That means strengthening its products like the Taran Decision Manager while continuing to help companies through its consulting services. As technology continues to reshape our lives, Taran is positioned as a rising leader. One that can guide companies everywhere to unlock the power of their data and seize the opportunities of the digital future.