Making On a regular basis Retailer Choices Smarter, Quicker
Within the fast-paced world of retail, retailer managers are beneath fixed strain to make fast choices based mostly on real-time efficiency. Lotus’s, one in all Thailand’s largest retailers, acknowledged that regardless of getting access to plentiful information, actionable insights remained out of attain for a lot of frontline groups. The corporate sought a option to empower its 3,000+ retailer managers to make smarter choices from the palm of their hand.
Amity Options partnered with Lotus’s to introduce a Pure Language Question (NLQ) platform built-in with a chatbot interface, making it simple for non-technical customers to retrieve insights simply by asking questions in Thai or English in plain language like “ยอดขายวันนี้เทียบกับเมื่อวานเป็นอย่างไร?” or “Which SKUs are underperforming this week?”—no dashboards, no delay, simply immediate solutions.
A Basis for Scale, Velocity, and Retailer-centric Choices
With Amity’s NLQ-powered resolution, Lotus’s has redefined how insights are accessed and acted upon on the retailer stage. Relatively than navigating a number of methods or studies, retailer groups now interact immediately with their information utilizing pure questions, decreasing dependency on head workplace and dashing up response time to on-ground occasions.
The chatbot interface, built-in with each desktop and cellular platforms, ensures that information is at all times obtainable on the go. Whether or not figuring out underperforming SKUs or validating promotional effectiveness, retailer leaders are actually outfitted to make choices immediately and independently.
Actual-Time Retail Intelligence in Beneath a Minute
This near-instantaneous response is powered by Databricks Mosaic AI Mannequin Serving, which ensures low-latency, real-time inference. This permits retailer managers to obtain solutions in seconds with out mannequin deployment and upkeep complexities. The platform additionally makes use of Mosaic AI Vector Search to search out probably the most related information for every question effectively.
With a backend powered by scalable infrastructure, Amity’s resolution offers solutions inside 5 seconds to 1 minute. This has minimize perception supply time from a median of 2-3 hours to beneath a minute, decreasing guide report dependency and enabling leaner ops groups throughout areas. Insights that when took a good portion of the day to compile are actually only a query away.
Whether or not checking on gross sales anomalies, inventory provisions, or weekly recognized loss studies, managers can act quicker, growing store-level effectivity, agility, and efficiency.
The Growing Utilization of Actual-Time Retail Intelligence
This chart illustrates the growing utilization of an AI-powered system by monitoring two key metrics over time (from September 2024 to April 2025).
The info demonstrates an upward pattern in system engagement (message quantity) and person base (distinctive customers), showcasing elevated adoption and utilization of the AI capabilities throughout the group.
Reworking Knowledge Entry with Morning To-Do Lists
To transcend conventional query-based insights, Amity launched a To-Do Listing (TDL) function, a each day AI-generated briefing tailor-made for every retailer. It highlights anomalies, efficiency gaps, and high-priority actions throughout:
- Gross sales spikes and underperformance
- Inventory gaps and overstock alerts
- Margin points and adjustment strategies
- Identified loss monitoring with detailed merchandise breakdowns
- Promotion feasibility based mostly on present inventory
Retailer managers begin their day at 8:00 AM with the TDL, which now replaces the necessity to open a number of BI instruments or watch for IT studies.
TDL provides me insights I’ve by no means seen earlier than. It helps me focus straight away. I can spot irregular gross sales, perceive inventory points, and even see margin gaps earlier than morning rounds are over. — Retailer Supervisor, Lotus’s
Safe and Scalable Knowledge Infrastructure with Databricks
Behind the scenes, all the information and analytics engine is constructed on Databricks, making certain velocity, scalability, and robustness throughout 1000’s of areas. Amity carried out a Row-Degree Safety (RLS) mannequin inside the Databricks Platform, making certain that every retailer supervisor solely accesses the info related to their retailer, preserving privateness and operational integrity.
To streamline deployment and observability throughout its rising variety of LLM-based instruments, Amity leverages Mosaic AI Gateway. This permits for unified logging, efficiency monitoring, and fallback routing throughout basis fashions, making certain reliability and governance at enterprise scale.
This structure helps real-time information pipelines, each day AI activity era, and safe user-specific querying, all managed on a unified lakehouse structure. The result’s a seamless stability between enterprise-scale governance and on-the-ground usability.
Scaling Perception Throughout 1000’s of Shops
Amity Options at the moment powers over 3,000 shops throughout Thailand. The AI chatbot receives 1,000+ messages each day on common, serving as the first interface for retail intelligence that matches into retailer managers’ routines and gadgets.
By making information obtainable immediately and in plain language. Amity has democratized information throughout each stage of the Lotus’s group. From frontline decision-making to regional technique, the insights delivered have shortened response instances, elevated focus, and empowered each retailer to carry out at its finest.
A New Commonplace for Knowledge-Pushed Retail
The collaboration between Lotus’s and Amity Options marks a brand new chapter in fashionable retail operations. Each retailer supervisor has entry to AI-powered insights, and each determination is backed by information delivered in actual time.
With confirmed success throughout 1000’s of areas, this resolution has redefined how Lotus’s groups work, making information not simply accessible however actionable, customized, and embedded into their each day workflow.
As Lotus’s continues to guide in digital transformation, the NLQ, TDL, and Databricks-powered infrastructure stand as a testomony to what’s doable when know-how is constructed for the individuals who use it most, delivering readability, velocity, safety, and confidence on the frontlines of retail.
