Yali Sassoon, Co-founder at Snowplow, looks at how organisations can use behavioural data to create a more efficient customer service function and build predictive models to detect when consumers require human interaction.
For many modern businesses, the ability to deeply understand customers and deliver a more personalised service is critical. To achieve this, many companies are increasing investment in Artificial Intelligence (AI) and feeding these models with structured high-quality behavioural data. However, while transactional and demographic data are well understood and fairly valued, behavioural data is still undervalued. If companies want to truly understand their customers and enhance customer experience, then improving understanding, collection and use of high-quality behavioural data is critical.
Today, companies have the opportunity to collect rich, granular behavioural data at every digital touchpoint with customers, whether that is within apps, websites, or across channels like email and support desk. Importantly, behavioural data captures not just the decisions that people make, but how they make them and the context they are made in, making it much more explanatory and predictive than demographic or transactional data.
For businesses, this helps drive a much deeper level of customer understanding and a better basis to personalise the customer service. For example, B2B SaaS companies can use it to build a picture of how different people in prospect organisations learn about their technology, trial it and decide to evolve their usage. Media companies can understand how users engage with content and what they like, and retailers can understand how customers make purchasing decisions: what information they require and how they consume it.
How behavioural data improves customer service
Behavioural data can improve customer service in many ways. It can help create a single customer view, shorten the time taken to resolve a problem, or by providing a personalised customer service experience. What’s more, it supports a range of use cases. Here we look at three that help deliver an enhanced customer experience:
Use case 1: Delivering customer insight to the helpdesk
Support agents have a difficult task. They have to engage closely with customers to understand their problem and the context, to help them resolve it. And often by the time a customer gets in touch with an agent, they are already frustrated and in no mood to carefully communicate all the information needed to provide the most effective support.
Behavioural data can help enormously here. In the first instance, it can provide support staff with a detailed view of the customer’s journey before submitting the ticket, as well as a view of what the customer is doing in-product. This saves the customer having to explain too much while providing the agent with a good understanding of the problem.
Because behavioural data is so rich, it can also be used with AI to help predict the customer issue and potential routes to resolution. For example, it can be used to dynamically route requests to the agent with the best track record of resolving those types of issues quickly. Or it can be used to proactively spot the issue before the customer has to contact support so that an automated intervention can be initiated.
Use case 2: Capturing and modelling data to drive search optimisation
Behavioural data is also an important enabler of better search optimisation. Companies typically optimise search performance based on data collected by the search engine itself, on which results were selected against each combination of search terms. This enables them to optimise the ranking of results based on click-throughs.
Behavioural data can describe not just the search terms that were entered, but the context that the search was conducted in. By showing customers results they’re more likely to purchase, rather than ones that look promising but on inspection prove to be disappointing, organisations can deliver a much better customer service.
Use case 3: Understanding customer data through customer journey analytics
Organisations can also use behavioural data to deliver customer journey analytics. Successful brands can use this capability to understand customer behaviour across disparate systems and channels. This could be things like the best time to engage a particular customer or what channels are best to engage with them.
Getting the most from behavioural data
The above examples highlight how behavioural data can be used to enhance customer experience. But to get the most from their behavioural data, organisations must optimise its use.
The problem is many organisations lack high-quality behavioural data. This is in part due to factors such as customer opt-outs, technical challenges around implementing tracking, human error and privacy-based measures like Intelligent Tracking Prevention (ITP) which make collecting high-quality data difficult. However, it’s also down to the fact that previously many companies didn’t need high-quality behavioural data: the need only arises when organisations start using it to execute more analytically demanding use cases.
The ability to set up proper tracking, with proper testing and proactively monitor data quality (including identifying and resolving data quality issues at source) is critical if behavioural data is to be used to drive meaningful customer understanding. Companies need to move away from a ‘track everything’ approach and instead adopt a more intentional approach to data collection that enables a more deliberate and transparent use of the data to benefit the customers that the data describes.
Currently, most organisations are using behavioural data to power a limited set of use cases. Typically, this is done via commercially-packaged analytics tools, such as Google Analytics, Adobe Analytics and Amplitude. While these tools are great for measuring and optimising marketing campaigns and improving conversion rates, they are not as good at driving the deep customer insight required to power personalisation and other sophisticated use cases.
Often, they don’t enable the integration of data sources, leaving organisations unable to see the full picture and drive meaningful impact to their revenue and profit. If businesses want to use behavioural data and AI to drive more effective customer acquisition, customer lifetime value, pricing, promotion and personalisation, they need to start by collecting behavioural data that is fit-for-purpose. The only way to do this is through a behavioural data platform (BDP) that provides flexibility and control over how data is collected, rather than relying on packaged tools to collect it and export it as a by-product.
The role of a behavioural data platform
A BDP provides behavioural data management and companies can use it to collect and operationalise behavioural data. A good BDP will allow organisations to capture rich, granular, and accurate event-level behavioural data and power real-time use cases. It will also enable businesses to change what behavioural data they capture as their organisations/customer-service functions evolve.
A key benefit of a BDP is empowerment of the service team across different use cases. By equipping support teams with actionable customer insights, organisations can not only solve customer problems quickly but also predict and proactively solve future requirements. That means responding to particular customer needs and sensitivities, being alerted to upsell opportunities and indicators, and helping the customer achieve their end goal in fewer interactions.
Click below to share this article