In a world where groceries arrive at the doorstep in ten minutes, why shouldn’t brands get insights and make decisions just as fast? With markets such as quick commerce redefining delivery speed, waiting days or even hours for insights can be outdated.
That’s the premise on which GobbleCube, an AI-powered agent for brands, was founded. Its 250 strong customer base includes brands from both large and small businesses as well as D2C brands such as L’Oreal, P&G, Dabur, ITC, Reckitt, Minimalist, Farmley, etc.
Quick commerce, in particular, has disrupted the way retail works, which means that the older ways of looking at data doesn’t work. Brands have gone from seeking data to dashboards to insights. Now, they even expect intelligent agents that can act on their behalf,” says Satyam Krishna, Founding Engineer & Head of Data Engineering, GobbleCube.
‘Why is my brand not performing well’, ‘where should I spend’, ‘how do I price’, ‘why is my campaign not giving me the same value as earlier’, ‘how can I increase my margins’ are among the many questions that GobbleCube’s AI layer helps brands find answers to within minutes.Whether it is detecting revenue leakages, identifying underperformance or surfacing untapped opportunities, data is at the core of this AI layer that helps churn out insights, strategies and decisions by processing billions of data points and signals across various e-commerce platforms.
“With LLMs, workflows are evolving into decision flows. It’s no longer about ‘help me find this data’ but ‘help me make this decision’ or even ‘make the decision for me’,” explains Krishna.
GobbsGPT AgentGobbsGPT is one of GobbleCube’s most prominent agents. Currently in a beta stage, it is an AI co-pilot built exclusively for CXOs. “GobbsGPT goes beyond answering the ‘what’ to uncovering the deeper ‘why’ and even shaping the ‘what next,” Krishna explains. “It can diagnose why a campaign underperformed and suggest concrete next steps, such as adjusting bids or regional spend strategies.”
By combining billions of data points with its proprietary SLM models, GobbsGPT turns raw data into clear, actionable strategies, enabling business leaders to make decisions with speed and precision without having to wait for weekly reports or wading through dashboards.
The right data platform is imperative for GobbleCube to maintain this speed and precision, and do it cost effectively with the limited resources available. Behind its AI engine lies the Snowflake data platform that powers everything from ingesting sales, campaign and market data to running models that uncover revenue leaks, underperformance and growth opportunities. It enables near real-time onboarding, instant dashboards and faster experimentation with strategies.
Earlier, the company was using the standard Redshift clusters for its data platform. However, significant resources went into managing the clusters and their maintenance, which made the platform untenable as the data volumes grew.
Snowflake’s scalability and cloning features have streamlined operations across teams:
- Data engineering teams use it to simplify pipelines.
- Analytics and data science teams rely on it for rapid model training and iteration.
- Customer success and product teams use it to deliver instant insights to brands.
Onboarding new brands has gone from hours to just five minutes, enabling the company to scale to over 250 brands without adding a large infrastructure team. Faster iteration has also unlocked new product capabilities like real-time alerting and conversational insights, allowing GobbleCube to deliver value to customers almost instantly.
All this has helped the company expand its market reach, speed up innovation and serve brands at scale with the same lean team.
“For GobbleCube, the future of commerce isn’t about more dashboards, it’s about decisions that move as fast as the market does,” concludes Krishna.
