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 selections primarily based on real-time efficiency. Lotus’s, considered one of Thailand’s largest retailers, acknowledged that regardless of gaining access to considerable information, actionable insights remained out of attain for a lot of frontline groups. The corporate sought a strategy to empower its 3,000+ retailer managers to make smarter selections 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 IASY for Non-Technical Usters to RESER THETIE SHESST THE THE THE THE THE THE BEISTE SIB Or English in Plaine Language Like “What’s the gross sales right this moment in comparison with yesterday?” Or “What Are Underperforming this Week?”
A Basis for Scale, Velocity, and Retailer-centric Choices
With Amity’s NLQ-powered answer, Lotus’s has redefined how insights are accessed and acted upon on the retailer stage. Moderately than navigating a number of programs or studies, retailer groups now have interaction straight with their information utilizing pure questions, decreasing dependency on head workplace and rushing up response time to on-ground occasions.
The chatbot interface, built-in with each desktop and cell platforms, ensures that information is all the time accessible on the go. Whether or not figuring out underperforming SKUs or validating promotional effectiveness, retailer leaders at the moment are outfitted to make selections immediately and independently.
Actual-Time Retail Intelligence in Below 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 answer gives solutions inside 5 seconds to 1 minute. This has reduce 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 after took a good portion of the day to compile at the moment are only a query away.
Whether or not checking on gross sales anomalies, inventory provisions, or weekly recognized loss studies, managers can act sooner, rising store-level effectivity, agility, and efficiency.
Determine 1. Instance UI Web page of Actual-Time Retail Intelligence: Comply with-up Query
Determine 2. Instance UI Web page of Actual-Time Retail Intelligence: Gross sales Report
The Growing Utilization of Actual-Time Retail Intelligence
Determine 3. Day by day Consumer Efficiency Breakdown
This chart illustrates the rising 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 consumer base (distinctive customers), showcasing elevated adoption and utilization of the AI capabilities throughout the group.
Remodeling Information Entry with Morning To-Do Lists
To transcend conventional query-based insights, Amity launched a To-Do Record (TDL) function, a day by 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 options
Recognized loss monitoring with detailed merchandise breakdowns
Promotion feasibility primarily based 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 look forward to IT studies.
TDL provides me insights I’ve by no means seen earlier than. It helps me focus instantly. I can spot irregular gross sales, perceive inventory points, and even see margin gaps earlier than morning rounds are over. — Retailer Supervisor, Lotus’s
Determine 4. Instance UI Web page of To-Do Record
Safe and Scalable Information Infrastructure with Databricks
Behind the scenes, the complete information and analytics engine is constructed on Databricks, guaranteeing velocity, scalability, and robustness throughout hundreds of places. Amity carried out a Row-Stage Safety (RLS) mannequin throughout the Databricks Platform, guaranteeing 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, guaranteeing reliability and governance at enterprise scale.
This structure helps real-time information pipelines, day by day AI job era, and safe user-specific querying, all managed on a unified lakehouse structure. The result’s a seamless steadiness between enterprise-scale governance and on-the-ground usability.
Scaling Perception Throughout Hundreds of Shops
Amity Options presently powers over 3,000 shops throughout Thailand. The AI chatbot receives 1,000+ messages day by day on common, serving as the first interface for retail intelligence that matches into retailer managers’ routines and units.
By making information accessible 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 Information-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 hundreds of places, this answer has redefined how Lotus’s groups work, making information not simply accessible however actionable, personalised, and embedded into their day by day workflow.
As Lotus’s continues to steer in digital transformation, the NLQ, TDL, and Databricks-powered infrastructure stand as a testomony to what’s attainable when expertise is constructed for the individuals who use it most, delivering readability, velocity, safety, and confidence on the frontlines of retail.
