“When our NPS suddenly declines, how can a product owner get immediate actionable insight about what to fix, and the potential impact on the NPS score, without having to read hundreds or even thousands of customer verbatim themselves?”
How insights hidden in a tier 1 bank's data allowed them to uncover the source of an NPS decline
When a large Tier 1 Bank’s NPS suddenly declines, how does the product owner get immediate actionable insights and clarity around the issues and their potential impact? Reading thousands of customer verbatim records is time consuming and not feasible, or in this case, spending weeks to analyse feedback using Excel/MS BI only to generate poor, unclear results.
Every minute of every day the bank’s digital assets were being used by customers and businesses, however their NPS score wasn’t reflecting the work they were doing to improve site functionality and usability for their customers
Their Digital NPS had declined rapidly over a couple of months but the reasons for the drop weren’t apparent. Three weeks of traditional analysis by a team lead to some very broad reasons that weren’t granular enough to enable decisions to be made – they had ring-fenced an area of concern but not the issues specifically, nor could they measure the impacts. The stakeholders were stuck and had no way to make efforts to reduce the NPS decline with confidence, let alone turn it around.
The Tier 1 bank onboarded Touchpoint Ipiphany, Touchpoint Group’s AI text analytics software. It was programmed to be respondent to the language used commonly within the banking and finance sector for this bank's needs (Ipiphany can be programmed for any required sector). Within minutes, the AI Analyst picked up and understood distinct and actionable pain points in the bank's data using natural language processing.
The same data that plagued their team of analysts for over three weeks was uploaded to Ipiphany and within 24 hours the reasons for the NPS decline were uncovered with granular clarity. The bank, over a short period of time, had made minor adjustments to customer-facing aspects of their website. Customers had not adapted to the changes positively, many preferring layout and functionality of the previous versions.
Ipiphany was able to quickly draw a detailed conclusion showing the areas for concern and the impact it was having on their NPS score in an Exec level report. It then went on to provide the team of analysts a deep-dive analysis of exactly what aspects of the site customers were unhappy about, why they were upset, how to resolve the issue and most importantly, to quantify the impact change will have.
The detailed evidence-based changes showed impact within a couple of weeks. The NPS improved and even surpassed what it had been before the decline. This was due to not only the issues discovered being a mix of recent changes to the digital assets but also longer-term issues and customer preferences that had not been identified or had been overlooked for the past two years.
Touchpoint Ipiphany was able to collate all their unstructured (open ended feedback questions) and structured data from multiple channels and analyse it together, rather than taking a sample or analysing separate channel feedback. By applying this holistic analysis, the bank was able to gain a richer insight into what customers were saying about their whole business and draw a larger picture, picking up the granular detail being missed by manual analysis and other tools. It drew quicker, more informative feedback, highlighted the areas to target first and provided actionable insights that enabled the team to easily prioritise actions based on the automatic categorisation by Ipiphany to quickly act in areas of concern.
Enter your email address to read more
To read this case study and receive a digital copy enter your email address below.*
Are you trying to determine the cause of a change in your business's customer metrics? Let us know what type of data you'd like to analyse and we'll tell you how Ipiphany can help you find the root cause of issues in your customer data.