Mixpanel Adds Generative AI So Companies Can ‘Chat with Their Data

-Product teams, marketers and executives at 8,000 of the world’s leading digital firms can query their user data with natural language chat

-Mixpanel’s incorporation of OpenAI’s GPT-3.5 Turbo allows accuracy to be validated and prioritizes privacy

Mixpanel, the event analytics* pioneer, has integrated generative AI so companies can ‘chat with their data’. The move makes it easier for companies to quickly understand the impact that product and marketing decisions have on their customers’ experience and their company’s bottomline. Mixpanel’s application of generative AI allows queries built by AI to be reviewed so people can understand the source and accuracy of the report created, without the need to expose their proprietary data to the Large Language Model (LLM).

Now, anyone using Mixpanel can clearly understand every customer’s experience by asking plain English questions to OpenAI’s ‘GPT-3.5 Turbo’ model, removing yet another barrier for companies that want to understand how people use their apps.

For example, a non-technical employee at a ride hailing firm might ask: “Which group of users most frequently convert when we apply surge pricing across our key markets?” A marketer can quickly understand trends relating to advertising spend or a salesperson can see changes in revenue over time. Generative AI builds the required query, which is executed in Mixpanel, returning the relevant chart in an instant to visually display conversion trends for different cohorts across different markets.

Marketing Technology News: MarTech Interview with Kamakshi Sivaramakrishan, Founder and CEO of Samooha

Amir Movafaghi, CEO, Mixpanel said: “Generative AI is the next interface to computing, and it’s unlocking huge productivity gains. In our world, this means it’s much easier for anyone to query their data in plain English by asking the AI a question. Making analytics accessible, so literally everyone can participate, will significantly improve decision making across companies.”

Prioritizing privacy and validating accuracy

Despite the huge opportunity generative AI represents, it isn’t always accurate. That’s why Mixpanel queries built using generative AI allow people to check the source of their answer to understand how a report was generated. While the AI feature will be made available to all Mixpanel users, it will also be optional, so people can choose to build queries using AI or continue with the existing user interface.

Mixpanel is trusted with user behavior data and provides insights to 8,000 paying customers, including large enterprises and the world’s most successful digital leaders, such as CNN, Uber and Yelp! That’s why Mixpanel’s introduction of generative AI prioritizes privacy. Mixpanel customers will not need to contribute their data to the Large Language Model, instead the AI increases the speed and ease with which queries are built, and it’s Mixpanel that actually analyzes the underlying data.

“Generative AI is a bit like electricity, you can build it into other products to make things faster and easier. We’re using it to speed up workflows and simplify how people ask questions of their data. But this is just the start, and we expect LLMs will enhance analytics for years to come.” added Movafaghi.

Traditionally, data analysts using Business Intelligence tools needed to write complex SQL queries when analyzing data, a requirement that made analytics a niche discipline and prevented companies from getting fast answers from data at scale. Mixpanel changed this with its event-based analytics system, which non-technical employees use to ask questions of their data with drop-down menus. The introduction of generative AI reimagines the data analytics process again, so anyone can use Mixpanel to support better decision making by easily asking questions of their data.

The new Mixpanel generative AI interface is currently being used by a select group of customers as part of a closed Beta program. The capability will be made available as an optional interface, at no additional cost, to all Mixpanel users over the coming weeks.

Marketing Technology News: From Hype to Reality: How to Successfully Implement ChatGPT in Your Marketing Campaign

About Event analytics

Event analytics captures every action (or event) that each user performs within a digital product, like an e-Commerce site or a ride hailing app. This very granular view helps companies understand how different groups of users behave at various points during their experience, to answer questions like: which cohorts of users drop off during sign-up? Using Mixpanel, it’s now also possible to understand the knock-on revenue impact of such customer experience issues.

This approach is much faster and easier than traditional Business Intelligence (BI) tools that require data to be prepared and tabulated, with BI queries coded in SQL. While BI analysis can provide a view of how many customers churned, it doesn’t allow exploration of why users are behaving in certain ways. That’s why event analytics is so crucial for product teams that create and continually improve the world’s most successful digital products.