The word 'customer' is etymologically derived from Old Dutch 'calant' (friend) and French 'chalant' (to be very interested). Its counterpart is the well-known 'nonchalant' (negligent). Treating your customers as friends, being genuinely interested and giving them unsolicited advice when necessary, is a fine ambition. Especially if you want to do that company-wide in an active way. According to Parker& Hudson's ladder of culture, we know several stages of how actively a company operates:

1. Pathological: We do nothing as long as we are not caught (e.g., by the supervisor). It is abnormal behavior that damages relationships.

2. Reactive: We react to problems, so-called firefighting. It is mostly behavior based on an ad hoc policy.

3. Calculative: We answer incoming (inquiries) according to a fixed process. Bureaucratic behavior lies in wait.

4. Proactive: We try to prevent questions before they are asked. Clear (core) values are drivers of behavior and there is a strong will to continuously improve.

5. Generative: We are ahead of others, we generate new demand ourselves(s) (hence generative). We do not give the customer what he asks for, but what he needs.

Legendary in that regard is Henry Ford's statement, "If I had only listened to what customers asked for, I should have given them faster horses." If you are brutally honest, in what level do you see the most behavior within your office? And what does it take to make a move? On the Adfiz site (www.adfiz.nl/dossiers/actief-klantbeheer) you will find a workable roadmap (see also pages 36 through 38 in this issue). When formulating an ambition (step 1 in the plan), the aspiration should be level 4 (proactive) or 5 (generative), with a minimum of level 3 (calculative). Everyone on the shop floor can roughly estimate what level the team is operating in. Spar with each other about this, what is going well, what could be more ambitious?

With most software packages, you can largely control the calculative level. You can create segments and define a contact frequency for each segment, at which the software, based on 'business rules' (if...then) gives a signal to call the customer or pay a visit. For example, for companies at least once a year, for self-employed people once every two years, for the better individual once every three years, etcetera. In some packages you can set up a workflow that automatically sends an e-mail based on the signal, for example asking if a customer wants a maintenance call. In practice, this does not work well. The segments are too large (and thus the message and process too impersonal) and many customers who are approached are not waiting for it. It is very labor intensive, while the response rate (and conversion rate) is often low. This is not profitable and is also demotivating for your employees.

Deployment of smart technology

Adfiz's roadmap indicates that you have to strive for a uniform and consistent customer approach, which is profitable enough for your office and with which signals are automatically generated from the database. If you want to do this without deploying additional advisors, for tens of thousands of clients alike, all of whom have different - often still latent - needs, there is no escaping the use of artificial intelligence. The technology - actually an algorithm - personalizes one-to-one, so that you are always relevant. And stays relevant, because it learns from every interaction. This is necessary because customer needs can change daily. So it's about approaching the right customer, at the right time, through the right channel, with the right message. What concrete applications can we think of?

'Those striving for a uniform and consistent customer approach cannot escape the use of artificial intelligence'

Inform or advise?

The AFM also recommends formulating your "ambition about customer care in the management phase" and defining an approach. There is a plea to go beyond the minimum duty of care. Read through the AFM document "Interpretation informing and advising. There is advice within the meaning of the Wft when a personal recommendation is made to a (potential) client about a new financial product of a specific provider. There is informing when one of the mentioned criteria is not met (such as a recommendation on an existing product). Therefore, informing can be applied to a large extent to management, because this does not (yet) require a full advisory process.

Prevent valuable customers from leaving

This module predicts the future value ( Customer Lifetime Value CLV ) that a customer will bring in. Customers with a high CLV may have low sales now, but have great potential to become good, loyal customers for your firm. They apparently match, without the clients themselves knowing it, well with your firm's advice and product offerings. The CLV can be combined with a module that predicts which customers have a high churn probability (churn=royal). The result is a list of customers with a high CLV and a high probability that they will cancel their package. You can respond to this proactively and have an advisor approach these customers. Because the technique works with thousands of data points simultaneously, the conversion rate if you approach these customers is sometimes increased by 40 percent. This is obviously motivating for your employees. The customer is also happy, because you can proactively help him with personalized advice for a complete package.

Answer frequently asked questions proactively

In customer communication, there are all sorts of possibilities. A module called dynamic FAQ (=Frequently Asked Questions) predicts which specific questions a particular customer will have within a certain period of time. Thus, for example, in his portal, or in a newsletter composed with liquid content, the customer will only see FAQs that are relevant to him at that moment. This is the beginning of personalization (see also the article in VVP 2, April 2022). Research shows that roughly three-quarters of customers expect such a personalized offer from their financial service provider. Worse, three-quarters get frustrated if you don't offer that. Consequence: customers lose trust in your company.

Recommend best-fit coverages or services

This module can recommend a product or service for each customer, what the customer needs at that moment based on data analysis, what content the customer therefore finds relevant to read and what his preferred approach is: mail, mobile, website or perhaps a leaflet with the prolongation. This may sound like future music, but it has been used in retail for years. Technically, this is called contextual recommendations. What does that look like? People like you are more likely to choose comprehensive legal expenses coverage. Click here if you want more information. So here you start with information and depending on whether it is a modification of coverage or a new financial product, an advisory process follows.

Cost-effective

Until now, this complex technology has only been available (and affordable) to large corporates. This is why a number of software players, including InsuranceData, Building Blocks, Bug Business, WeGroup, have joined forces so that service providers (and their affiliated offices), proxies and the larger intermediary can also profitably deploy it.

Finally, the goal of deploying artificial intelligence is not to replace humans, but to use the advisor 's precious time as effectively as possible so that the right customer gets personalized attention at the right time.

Source: this article originally appeared in the VVP, read here the online article.