Firms that adopt this risk-focused approach will in turn have the scope to deep-dive their business data unlike ever before. Learn how using ConversationReviewAI will help your business gain 100% oversight of customer conversations and revolutionise the identification of problem cases for remediation, whilst providing analytical insight for your firm’s thematic review.
Part 5: Evaluating, evidencing and improving consumer outcomes
Garry Evans
We've talked about the ability to increase QA oversight to 100% by using ConversationReviewAI. But actually there's no point in doing it unless you can use that data – so what can firms do with the data?
Olivia Fahy
ConversationReviewAI isn’t just about producing metrics for MI and board packs - firms will get that data by virtue of using the product. And you can make the data in those packs much richer by demonstrating that 100% AI QA oversight. But I don't think it's enough to just have the pie chart of red, amber, green - even if it is across 100% of the business - because there’s still the question of what does it all mean?
It's about providing that baseline layer of data that firms can use as a springboard to become risk focused, or even more risk focused, rather than random. Firms can use process adherence failures as a foundation or red flag for identifying cases that may be delivering poor outcomes for consumers. If there's something going wrong with the process at the start, that's going to have a knock on impact throughout that whole consumer lifecycle. And thinking about Consumer Duty outcomes, there is likely to be a correlation between process failures and poor Consumer Duty outcomes. For example, if consumers haven't been able to make an informed decision because they haven't been given all of the information they need, there could potentially be a failure in consumer understanding. Or if consumers are getting poor customer service, there's going to be consumer support issue.
The more data you have, the easier it is to highlight where the issues may lie, and in turn then build a thorough picture of your consumer journeys. For example, if a firm's using customer surveys to gather data on consumer understanding, you can then match that data with the AI Assure scores to do a comparison. If this shows you've got a dissatisfied survey but a green AI outcome, that might mean the service associated with following the correct process is unsuitable (because the adviser followed the right process but the customer wasn't happy). Or if you've got a satisfied survey with a red process AI score, there’s a real risk that the customer hasn't been fully informed and they might not have understood the financial decision they made.
There are all sorts of deep dives that you can do when you've got data across 100% of cases. And it helps advisors so they can access and see the breakdown of their calls – and facilitates in immediately remediating issues if needed. For instance, if they notice they didn't do something they should have on a call, they can immediately call that consumer back and rectify the issue. It’s about having a portfolio of data to dive into that adds real value to your business.
You talk quite a bit about QA and compliance monitoring and how ConversationReviewAI can help with that, but does it only help with that everyday monitoring element?
It's a great question - the answer is no. It can help with anything that requires a review and has a process around it. Everyday monitoring is our core use case. That helps firms increase the efficiency and effectiveness of their review teams by enabling them to do their reviews more quickly and focus the majority of their time on high-risk cases.
But we could also help with thematic reviews. For example, we could use AI to perform thematic call analysis, identifying key risk indicators across large volumes of cases with minimal effort. Then firms can focus on areas that might present greater process risk before they become a problem. Or if firms want to perform a ‘look back’ exercise if they were concerned about a particular element, they could easily filter the cases down by that classification, and listen to that section of every call.
We also do a lot in the remediation space, which is a really interesting area where AI can provide huge amounts of value. We can use AI that's trained to specific parameters to support historic call analysis so it can complete past business reviews. To do this it would scan voice conversations and get straight to the cases or content that require the review team's attention. It’s about moving away from people having to manually sift through everything, which the AI can do instead.
Hopefully we've been able to demonstrate how ConversationReviewAI works and why it's a bit different to some of the other systems that are available. And we hope we’ve given a sense of how to help organisations deliver the evidence they need to in order to comply with Consumer Duty - by having access to all of the problem cases. This will help mitigate the future risk of complaints and enable full insight to feedback into the front end of the training process
Don’t miss the other four parts to this Consumer Duty mini-series, catch up with them all now:
- Part one: The role of RegTech to help monitor and evidence outcomes
- Part two: The power of tech-enabled process adherence checks to evidence Consumer Duty outcomes
- Part three: Utilise speech analytics to process risks in customer conversations
- Part four: Transform your Quality Assurance to innovate beyond random sampling