Pecan: Generative experience in the blink of an eye

Pecan, a low-code predictive analytics and data science platform company, was founded by Zohar Bronfman and Noam Brezis in Tel Aviv, Israel in 2018. Pecan’s solutions help companies turn data into valuable insights that lead to improvements in customer lifetime value, customer retention, conversion rates and demand forecasting.

Binci Heeb, editor-in-chief thebroker, talks to Zohar Bronfman, one of the founders and CEO of Pecan AI.

Zohar, it is a pity that you didn’t accompany your MD Dan Goldenblatt to the Swiss Insurtech Hub: Ecosystem Meetup in Zurich? It was very interesting and Switzerland is a beautiful country…

I wish I could have attended this year! I knew you’d be in good hands with Dan, though. I hope to join you for a future event very soon.

You have a PhD in computational neuroscience and one in history of science and philosophy. Do you need the knowledge from both degree programs for Pecan AI?

Every startup founder needs a minimum of two doctoral degrees. OK, I’m joking — but I’ve found that insights from both programs have shaped my approach to both building an AI product and guiding a startup. It’s incredibly helpful to understand the technical and philosophical foundations of both technology and leadership during this time of rapid change and exploration.

Pecan AI promises that its generative experience tool can “predict business needs in the blink of an eye.” How is that possible?

For the first time, we’re making it possible for business and data teams to simply use natural language to start using predictive AI. This process has traditionally been far more involved, requiring extensive collaboration with data science specialists, time-consuming data preparation and exploration, and hand-coding predictive models. But by providing an automated, intuitive interface, we’ve accelerated the predictive process dramatically, helping businesses unlock the value in predictive analytics within hours — not months or more. Behind the scenes, our platform uses unique, patented components that power this impressive outcome.

Until now, the ability to predict customer behavior was based on guesswork. To what percentage can Pecan.AI predict customer behavior?

Pecan’s models can be used to predict a variety of customer behaviors. For example, we have helped many businesses understand and predict customer churn so that they can be proactive about intervening and retaining customers at risk. Prediction accuracy depends on many factors specific to each business and its data, but in general, Pecan’s models are far more accurate than human guesswork. And in reality, even a model that is not perfectly accurate can still provide a business with incredible ROI by empowering them to engage customers in the right ways more often.

Your generative AI can also create customized notebook-style modeling experiences in SQL. Tell us more about this.

Data analysts have used notebook-style interfaces to work with their data for some time. With Pecan’s Predictive GenAI, they can now use a familiar notebook interface to take the next step and build predictive AI. Our platform uses their natural-language input to automatically generate a SQL-based notebook that defines the training dataset for their desired predictive model. They don’t have to guess which data is correct to train precisely the model they need. We get them started with the framework that will lead them to success, using SQL, the language data analysts know best.

How trustworthy are your predictions?

The best testimonial for their trustworthiness is our customers’ success stories. We’ve been able to help a wide variety of companies in different industries and with varying business models adopt predictive AI successfully, with impressive results. Those results include boosting customer engagement, increasing retention, improving marketing effectiveness, and ultimately increasing revenue and growth.

In which areas can your AI help insurance companies?

There are many potential use cases for Pecan in insurance companies. We’ve helped insurance companies with customer churn prediction, enabling them to take action early and increase retention rates. We’ve also provided upsell and cross-sell prediction, supporting insurers’ efforts to grow revenue and share of wallet by selling complementary or upgraded products. Another critical use case for insurers is marketing mix modeling, which offers insights into the impact of marketing spend in various channels and helps insurers optimize their marketing strategy. In addition, we are always open to exploring other predictive use cases with customers.

Why should insurers and insurance brokers use your platform?

We’ve successfully delivered powerful results for insurance companies and are confident we can do the same for others. We understand the industry and its current challenges. We are open to partnering with insurance companies for a more collaborative experience in the predictive AI journey. We can also support and guide companies who would like their own data and business teams to be more hands-on with the Pecan platform.

You have customers in 15 countries. Also in Switzerland?

We would love to add Switzerland to the list of countries where we serve customers!

How is your data integrated into your customers’ business systems?

We help customers integrate predictions from their Pecan models directly into the business systems they already use. For example, they can send predictions into their CRM or ERP systems to guide daily actions and decision-making. This ease of integration means they don’t have to radically change their business processes and everyday work. Pecan makes AI adoption seamless and removes the need for dramatic digital transformation.

And how is the customers’ data protected?

We use robust data protection practices, including enterprise-class security tools and both active and passive security measures. Pecan has ISO 27001 and SOC2 Type II certifications, reflecting our serious commitment to securing customers’ data.

Is your analytics platform open to all users?

Anyone can sign up for a free trial of the Pecan platform. Data and business analysts with SQL knowledge are typically best suited to using the platform themselves. We also offer custom options for companies seeking additional support and guidance in using Pecan as a key part of their AI journey.

Pecan AI offers a marketing mix modeling solution. What is this about?

Marketing mix modeling (MMM) is a time-tested method to understand and optimize the effectiveness of marketing strategy and spending. While it’s traditionally been a time-consuming, highly specialized, and expensive technique reserved for only the largest companies, new approaches incorporating AI have now made it possible for many marketing teams to benefit from MMM. At Pecan, we’ve developed innovative ways to make MMM more accurate and accessible for our customers, who have seen dramatic improvements in their marketing ROI.

Read also: Nemesysco: Wie 007 weiss, ob Sie lügen and Swiss Insurtech Hub: Ecosystem Meetup on February 5th

Dr. Zohar Bronfman is the Co-founder and CEO of Pecan AI, an innovative artificial intelligence company specializing in Predictive Gen AI. Zohar is considered one of the most knowledgeable speakers in the field of artificial intelligence, blending a profound understanding of philosophy, technology, and business.
Pecan has raised approximately $120 million from leading investors such as Insight Partners, GV (Google’s investment fund), Dell Capital, and others. Pecan boasts a large variety of customers, including Fortune 500 and Fortune 100 companies, whose adoption of its platform has impacted revenues amounting to billions of dollars.
With a distinguished academic background, Dr. Bronfman holds two Ph.D. degrees from Tel-Aviv University, one in Computational Neuroscience and the other in Philosophy of Science. He also obtained master’s degrees in Computational Cognitive Neuroscience and Theoretical Biology, showcasing his multidisciplinary expertise. Additionally, he completed his undergraduate studies in Economics and Business at The Open University.
Throughout his academic career, Dr. Bronfman has published 18 scientific papers in leading (Tier-1) scientific journals, establishing himself as a highly-cited professional in these field. He has also served as a lecturer at Tel-Aviv University, imparting his knowledge to M.A. students on the history and philosophy of brain sciences from 2017 to 2019. Dr. Bronfman served in the IDF’s elite 8200 intelligence unit, where he worked in the signal-intelligence division from 2005 to 2007.