How do you fix a customer feedback system that nobody trusts?
7% submission boost, strategic redesign
The problem
Tata Steel's Aashiyana platform had an NPS system that technically worked (data flowed, dashboards updated), but nobody believed the numbers. The system was measuring 'did our website load properly?' when leadership needed to know 'are we delighting customers through the entire purchase-to-installation journey?' Collecting feedback before customers had experienced anything was producing noise.
What I did
Redesigned the NPS approach based on Reichheld's methodology. Moved capture from website popup to post-delivery completion (the first moment customers had experienced enough to form loyalty judgements). Compressed follow-up questions from 5-6 to maximum 2. Added respondent segmentation to ensure only target customers received surveys. Proposed an episode-based architecture breaking the full customer journey into measurable, actionable stages.
Outcomes
What I learned
Measurement architecture is product design. The original system failed because it was designed for data collection convenience, not customer experience truth.
Timing equals legitimacy. Asking for loyalty ratings before customers have experienced anything meaningful trains them to ignore your surveys.
Episodes equal actionability. 'Overall experience' scores are too vague to fix. 'Installation episode' scores tell you exactly which installers need training.
Got a similar problem?
Tell me what you're working on. I'll tell you if it's a fit.
Prefer email? workwithakshatkharbanda@gmail.com →