Research News Live

The halo effect understanding biases in CX

Efforts to bring together transnational data with both strategic CX data and other sources of information can be particularly fruitful, writes Dr Violet Lazarevic in an opinion piece.

Customer experience or CX is a major buzzword right now in our industry. It is revitalising the focus on customers, which the insights industry has advocated from its inception. One of the reasons CX is gaining so much attention is because it’s often tied to organisational incentives or key performance indicators (KPIs). This is critical because most businesses wouldn’t exist, and our jobs wouldn’t exist, without customers. Therefore, it’s important that we strive to understand customers and measure the types of experiences they have with our business.

However, the CX narrative often fails to mention it is only one of the available sources of information regarding customers. There are many other sources and types of information that the company will collect about the customer. This will vary by business and industry, but may consist of: customer transaction information, loyalty program information, demographics, customer contact details, call centre interactions, social media interactions, and customer research. Some of these sources of information will be crosssectional or at a point in time, and some will be longitudinal.

Within CX itself, differences also exist. CX-type programs can be transactional or strategic in nature. In the many different businesses in which I have worked, we have often had both. A transactional CX-type survey is one that is triggered by making a purchase or having an interaction with the company (at a specific time). In contrast, strategic CX-type surveys are usually less-frequent and often conducted at a market level (not just among your own customers) to understand experience, as well as broader sentiment and advocacy. The two work well together. The transactional level is conducted at a specific point in time and provides clear organisational actions. The strategic level complements this by being longer term in focus and showing broader market-level sentiment and how that is changing.

Beware of assumptions

Some businesses only measure the transactional CX-type, and this can be challenging. There are undoubtedly many reasons for this. For one, it is easier to action something that is based on a concrete experience and you can typically add in some measures around different elements of that experience (speed, friendliness of staff, etc). The other reason is that the feedback is more timely, so the business can act on it faster. However, this approach to CX measurement makes one very big assumption. It assumes that the feedback you get at that particular point in time is based entirely on that transaction or interaction and nothing else. It also assumes that the feedback gained from that customer at that point in time tells the full story of their experience with the business. Unfortunately, this isn’t always the case. Customers develop relationships with a business and their opinion of that business is cumulative; each experience adds to what they already think about the business. Whether that opinion is negative or positive, customers can find it very difficult to put that aside and be objective at the time of completing the latest CX survey, which is supposed to be ‘only’ about their most recent transaction or interaction.

Efforts to bring together this CX transactional-type data with other sources of information can be particularly fruitful. These efforts around data triangulation lead to interesting findings, illustrating that customers are complex and that they have biases that they are often unaware of. For example, a really significant bad experience with a business can create a negative halo effect for a long time, where a customer will continue to judge that business harshly despite good or neutral more recent experiences. Similarly, with significant good experiences, a positive persistent halo effect can be created.

On the other hand, you may discover that the longevity of the relationship with your business may create some biases. For example, older customers who have had a relationship with a business a long time may feel a fondness for that business largely due to the longevity of the relationship rather than their actual most recent experience.

Data triangulation or bringing together different sources of information highlights that one source of customer data in isolation gives an incomplete picture. To completely understand customers in a holistic manner, multiple data sources are required. A mix of qualitative and quantitative sources works best, so that the limitations of each approach can be balanced. The richness behind the numbers can be bought out and the quantification behind the stories can also be illustrated.

Author: Dr Violet Lazarevic, General Manager of Customer Insights, Telstra

This article reflects research conducted by the author in her personal capacity.

This article also appears in the February – April 2020 edition of AMSRS publication, Research News – CX, UX & Research Design. Check out the rest of the articles in this edition.

Photo by Jongsun Lee on Unsplash


About The Research Society 1079 Articles
The Research Society is the peak body for research, insights and analytics professionals in Australia. It has a diverse membership of individuals at all levels of experience and seniority within agencies, consultancies, client-side organisations, the non-profit and government sectors, support services as well as institutions and the academic community. As well as over 2,000 individual members, the Research Society has 125+ company and client partners, with the number continuing to grow. The Research Society research professionals and company partners commit to and are regulated by the Research Society Code of Professional Behaviour.