Award Applications

Atidot was founded in 2016 by Dror Katzav and Barak Bercovitz, both of whom served in the technological unit of the Intelligence Corp in the Israeli Army, and Assaf Mizan, former Chief Actuary at the Israel Ministry of Finance. The coupling of technology and data experts with insurance experts places Atidot in the unique position of being a startup dedicated to bringing these two fields together in order to address the strategic needs of the life insurance market. With expertise in both technology and life insurance, Atidot was founded to provide a tool for Life Insurers that allows them to more accurately predict behaviour patterns therefore assisting with company strategies.

Insurance companies face increasing challenges in understanding the markets in which they operate. The demands to attract and retain customers, whilst maintaining core business competency, are becoming more difficult. Data is abundant and analysis, legacy technology and methods lag behind. The need to collate vast amounts of data in different formats and stored in separate locations is imperative if insurers want to understand their policyholders.

Atidot has built a platform that addresses these needs. Our goal is to assist life insurers in their strategic planning, including client development, client retention, underwriting and pricing. The Atidot system can create profile groupings of policyholders by analyzing an unlimited number of categories such as their age, occupation, gender, payment method, average salary for their geographic location and more.

We consume huge amounts of data (internal, external, social, IoT), then we value various advanced models with the target of best explaining customers' behaviour. Atidot's platform then connects to the insurer’s profit model to directly link behaviour to company profits. This allows us to suggest alternative courses of action together with the financial benefit (EV) resulting from each one. This is all connected to your sales force to make it actionable and monitored. Our solution identifies policyholders who are receptive to upselling whilst at the same time estimating their likelihood of lapsing due to this suggested change in their policy, giving strategy makers a choice of two options based on their policyholder’s overall predicted value.

All modeling starts with smart data, and with the combined expertise of our technology and insurance professionals, Atidot has built a platform that makes loading and analysis of all data extremely fast. Insurers are currently burdened with up to 10 operational systems, with data either siloed or stored in a data lake. Accessing this full data is impossible on current systems and the data lake format limits addressable answers. Our technology has been designed so that raw data is loaded from all systems. We then use AI to understand the insurance essence of the data and, by using this technique, can categorise the data accordingly. As an example, once the system understands that a particular data attribute is a premium, it can automatically know that negative values should be cleansed. There is minimal requirement for human interaction at this point as machine learning and artificial intelligence work to complete the sorting and cleansing process automatically irrespective of data format or size. Data can be uploaded in house or in the cloud, depending on country-specific regulations.

Once data has been cleansed and normalised, it is ready for analysis. At Atidot, our standard POC focus is on predicting the value of new business based on prediction of both lapse and profit pattern. Relevant data for a POC is a block of 100k in force policies sold in the past 5 years. The process starts with defining the business goal. This business goal could take the form of a question, such as ‘How does an insurer match a product to a customer to increase profitability?’ ‘Where are the pockets for lapses and how can an insurer detect them in advance?’ With reference to existing business, ‘Who are an insurer’s most valuable customers?’ ‘Who is about to lapse and why?’ ‘How should an insurer bundle products to improve commitment?’ From a European perspective, ‘How can an insurer achieve the above whilst optimizing on solvency requirements?’

Atidot builds a unified view of the customer, coming from all sources, including external ones. This is the model’s input.

Atidot’s technology also allows insurers to optimize. If an insurer wants to improve the value of the book, they can experiment with different strategies. The Atidot system allows insurers to predict what would happen if they change the male/female ratio, or the age distribution. They can skew the age distribution for example and see the effect on lapses, EV and other features. Once an insurer is happy with their new plan, they can save it and score all policies based on the new strategy. This strategy can now add to the operational effort with the new scoring mechanism.

Atidot’s innovative Big Data modelling and decision making tools will accelerate the pace at which senior life insurance executives can confidently and efficiently utilize big data intelligence in their decision making. Companies that take advantage of the Atidot platform will no longer need to make key decisions based on legacy systems and traditional analytic methods, and data cleansing will be dramatically reduced.

Because all data is held on a single platform, different business groups can access one central system for all their requirements. The marketing department will be able to see which products are more suitable for specific policyholders whilst ensuring that upselling will not put the full policy at risk. Actuaries can run multiple simulations simultaneously in an efficient and timely manner. New policies can be priced to better reflect risk based on the myriad of information now available.

Designed as a living tool, the platform is sustainable and can adapt to new situations.