AI Analytics: Video Analysis Of Customer Walk-In’s

Vision AI Empowers an Automobile Giant Personalize Customer Experience & Enhance Revenue

Table of contents
Contributors
Shilpa Ramaswamy

The Client & the Challenge

One of world’s largest two-wheeler manufacturers with a network of 4,000 showrooms and service centers catering to a customer base of over 44 million wanted a custom and scalable AI solution that had the power to track and measure customer engagement and experience across their largest network of sales and service.

Industry Overview

Disruption

The automobile industry is changing rapidly. Autonomous driving, electric powertrains, digital services, and mobility platforms have become the key pivots to an industry on the verge of disruption. Much like tech businesses who have opted for digitization and increasing automation to become customer-centric, automotive and mobility players are reinventing themselves for the future.

Business Challenge

Customer Experience

A smooth in-store experience translates into customer acquisition, retention, loyalty, and brand advocacy. Identifying friction points in the customer’s journey therefore is key to crafting positive purchase and support experiences. The client wanted to profile and segment walk-in prospects, personalize customer experience as well as collect feedback across company-owned showrooms and franchisee outlets with Vision AI, doing away with the manual processes that were slow and error-prone.

A smooth in-store experience translates into customer acquisition, retention, loyalty, and brand advocacy.


Solution

We used a blend of vision AI and Deep Learning to solve the customer's challenge. Here is a breakdown of the steps we used:

Step 1: Capturing the required data

After scrutinizing the existing video feeds, increasing the CCTV coverage was recommended. Fisheye cameras were installed as required.

Step 2: Inferring from the solution

Combining the power of ML and DL, an AI-driven inference solution that captured the entire customer journey including customer walk-ins, behavior, and engagement was developed.

Step 3: Predicting the actionable insights

The data-driven, real-time insights including footfall traffic, demographic profiling, dwell time, emotion detection, and heat maps were transferred to a user-friendly dashboard for the marketing team.


Impact Delivered

  • 95+% accuracy
  • Analyzed 2 years of video data.

Top Benefits

  • Personalized customer experiences
  • Lower customer acquisition costs
  • Higher sales with Multicamera deployment

The Akaike Edge

Inbuilt libraries, DL models with transfer learning capabilities