Fair Value Assessment: A smarter approach to customer segmentation
Across the market, firms continue to face challenges with customer segmentation in two related areas: first, in how target markets are defined, and second, in how Fair Value Assessments (FVAs) demonstrate that different groups (or segments) of customers are receiving fair value.
Too many Target Market Assessments are still written at such a high level that they are of limited use when it comes to assessing value for different customer groups. That problem is even more obvious where a firm co-manufactures or distributes a product, but the target market has effectively been set elsewhere in the distribution chain. In those cases, firms can end up relying on a target market definition that is too broad to support meaningful challenge.
In other firms, the opposite is true. The segmentation exists, but the MI does not. That is a problem, but certainly not an excuse!
A firm does not need perfect data before it can form a view on customer outcomes (although it wouldn’t be unhelpful). What it does need is a clear sense of which customer groups matter, why they matter, and how it will test whether value is being delivered.
When considering customer segments, it is important firms do not limit themselves to only those who the product/service is specifically targeted at. Particularly for products/services, which have traditionally been viewed as “mass market”. Instead, another useful way to consider outcomes across different customer segmentations is to look at “who” the product/service is purchased by, how they use it and whether there is evidence that fair value is being delivered.
What do “different customer segments” mean in practice?
In practice, firms can assess fair value through a number of different segmentation lenses. These may include distribution channel (such as direct, intermediary, aggregator or employer scheme), customer behaviour (for example frequent versus infrequent users, or more engaged versus less engaged customers), tenure or lifecycle stage, needs and characteristics including vulnerability, and product holding such as single-product compared with multi-product or add-on customers.
The right segmentation will depend on the business model, the nature of the product, how it is distributed and the outcomes the firm is trying to understand. The key point is that segmentation should be meaningful, capable of shedding light on customer outcomes (and even behaviours), and supported as far as possible by available MI.
For example., if a customer buys through an aggregator, renews year after year, rarely uses the product, and has different support needs from the wider customer base, it is not enough to assume they are receiving the same value as everyone else. The assumption needs to be tested.
Where MI is limited, firms should use what they have. Proxy measures, thematic reviews and targeted deep-dives are often enough to form an initial view, provided the limitations are acknowledged. Waiting for perfect MI usually means waiting too long.
Which customer segments should we assess?
There is no single answer to this question, and my view is that firms do not need to assess every possible segment. The starting point should be to identify the customer groups most likely to reveal whether outcomes differ in a meaningful way. The regulator has been clear that firms should take account of the needs, characteristics and objectives of customers in the relevant target market, including those with characteristics of vulnerability, and should monitor whether good outcomes are being delivered in practice.
My starting point would be to look at the customer segments that are most likely to experience the product differently or most likely to be exposed to poorer value. I would consider things like product/service usage, known characteristics of vulnerability (noting that not all characteristics will always = a vulnerability), distribution channel, tenure, no. of products held etc.
In some firms, that analysis will naturally sit within the TMA. In others, that analysis may not yet exist. Where target market definitions remain broad, firms should supplement them with a more practical analysis of the customers who actually hold the product.
A helpful way to approach which customer segments to assess is to ask yourself three questions:
Which customer groups are most likely to experience the product differently?
Which customer groups may be more exposed to harm or poorer value, including customers with characteristics of vulnerability?
Which customer groups can the firm realistically monitor using the MI, operational insight or proxy data currently available?
What evidence can we use to assess value?
The more credible and perhaps pragmatic approach is to bring together a suite of indicators that shows whether the price paid is reasonable relative to the benefits likely to be received by the relevant customer groups.
The MI and tolerances should be robust enough to support scrutiny and challenge. This is often where many firms fall short. They collect data, but do not organise it in a way that says anything meaningful about outcomes for different customer groups. The FCA’s fair value and outcomes monitoring work points in the same direction: evidence matters, proportionality matters, and firms are expected to act when the analysis suggests customers may not be receiving fair value.
Strive for progress, not perfection
Perfect data should not be the standard firms wait for. What matters is having the right foundations in place: meaningful customer segmentation, proportionate evidence and robust governance challenge to support a credible fair value conclusion.