Introduction

In an era of high connectivity and instant gratification, the expectations of customer experience have never been higher. Customers do not simply want but rather expect accessible and responsive communication across a variety of channels. And for organisations, the risks have never been higher.

Disgruntled users now have the power to share instant feedback online through social networking sites. Poor customer experiences can lead to negative defining moments, where one significant bad experience adversely alters a citizen’s perception of an organisation.

Negative customer experience can have hidden costs

Take tax collection as an example. Taxes are an inevitable and important component of citizenship. They help fund important public goods and services to maintain a well-functioning society. On the whole, citizens understand the importance of taxes and respect their tax authority.

If citizens have a significantly bad experience engaging with a tax authority – for example when asking for help making a compliant tax return – this may alter the relationship permanently. Following the experience, they choose to not reach out for help in the future if they feel they will not get what they need. As a consequence, this creates a compliance issue through errors in their returns. Ultimately, as well as an altered relationship between the citizen and tax authority, this well may have a knock-on effect of compliance costs for the tax authority.

What is composite AI?

The concept of composite AI has been popularised in the Gartner Hype Cycle for AI. Simply put, composite AI recognises the importance and combined value that multidisciplinary analytics can generate for organisations. As the term would suggest, it is not just the use of multiple analytics techniques within an organisation but the effective layering and blending of techniques as a means to best solve a given problem.

As a simple example, you may want to build a machine learning model for whether a customer is likely to make a complaint. The complaints process in itself will generate costs. So the organisation wants to encourage good citizen relations and minimise its costs. A composite AI approach might be to augment and improve model accuracy by blending other techniques.

For example, you could extend the model with outputs from text mining models that identify key terms in an email chain related to customer dissatisfaction. This layered approach may result in a more comprehensive and accurate predictive model.

Putting the right foundations in place

In many cases, a state of composite AI should be seen as a long-term target where synergies are naturally found as analytic maturities develop within the organisation. Organisations embarking on their analytics journeys can still find great value by applying individual analytical techniques to solve business problems.

As well as building towards a state of composite AI, it is important to consider setting the right foundations from a data and platform perspective. By the nature of customer service, there may be peaks and troughs for contact rates across the financial year. Analytics platforms should offer both resilience and scalability. This provides a consistently effective environment throughout the year whilst helping UK government organisations to minimise infrastructure spending. In particular, the benefits of leveraging cloud native architectures can help automate the scaling of the underlying infrastructure to match business requirements.

Making customer experience a priority across government

Historically, customer service has been perceived as a lower priority item when considering budget constraints within the public sector. For many government agencies the transformation to on-demand culture is both an operational and behavioural challenge. Coupled with ever increasing squeeze on budgets, this drives a need for efficiency and innovation within organisations.

The benefit is not just reducing operating costs but also improving customer experience and engagement over the long term.

Conclusion

In this blog we have discussed how analytics can augment, improve, and extend customer engagement for the benefit of both citizens and public sector organisations.

SAS provides a broad platform that covers the complete analytics life cycle across many analytic domains, including machine learning, natural language processing, forecasting and optimization. SAS also has a comprehensive ModelOps and decisioning framework, which allows you to embed your data products into your organisational processes. Find out more about composite AI at SAS here. For more on how SAS works across the UK government, search SAS UK gov.