The VAULT Model for Customer Segmentation

Many of the models and methods that are available for developing a customer segmentation strategy are often highly convoluted and overlook the basics of customer motivations, lacking appreciation of the customer’s considerations and needs as they migrate through the product experience. The VAULT Model provides a simple framework that enables the marketer to align strategies and tactics to customers based on their level of product engagement at any given point.

The model puts forward five specific ‘stages’ in a customer’s progression as they encounter and experience a product. At each stage, it is critical to have a comprensive understanding of the customer’s behaviours, beliefs, motivations, and attitudes towards the product or service. Based on these, the Marketer can define strategies and tactics to either further develop the customer’s level of engagement, or to re-engage where necessary.

Vault Customer Segmentation Model

V – Virgin: “Users who are about to, or have recently experienced the product for the first time.”

Customer retention starts before a customer first actively experiences or engages with a product. Their future behavior and attitudes towards the product are conditioned by their initial encounters with a brand through word-of-mouth or marketing. It is therefore imperative for the marketer to consider how their brand and acquisition marketing leads into the initial product experience and consequently, how the user is first ‘welcomed’ to the product. For websites, this requires understanding where the ‘doors’ are into the site, i.e. key pages through which new users first enter. How does the site messaging introduce and explain the product? How intuitive is the navigation to the next step in the product adoption process? How do you reward repeat visits and behaviour? Get the welcome right and it makes keeping users easier in the future.

A – Active:
 “Users who have actively engaged with or adopted the product in a repeated fashion.”

The second stage in the VAULT Model is to define what constitutes ‘activation’ and to determine how to move a user from being a ‘Virgin’ into being ‘Active’. These answers will vary depending on the product offering and user profile / lifecycle. An ‘Active’ user could be one that purchases a number of items (one or multiple?), or registers on the site to access additional content or benefits. It is imperative that the definition of activation relates to the core objective of the product. Activation Rate, i.e. the number of new users meeting the definition of activated, should be monitored on an ongoing basis as a Key Performance Indicator (KPI). As with Virgin users, attention should be paid to not only how a user is ‘converted’ into becoming Active, but also how they are ‘welcomed’ into this new level of engagement. Site messaging should reassure the user that they’ve made the correct decision.

Activated users can be further analysed and understood based on their degree of activation. At a basic level, this can take the form of ‘Light’, ‘Medium’ or ‘Heavy’ users. These definitions should again be based on the core objective of the product and tracked as part of the suite of KPIs. Accordingly, strategies and tactics using a mixture of messaging and offers can be applied to transition users between these three strata to increase product usage. An alternative segmentation identifies ‘Occasional’, ‘Regular’, ‘Advocates’ and ‘Super Fans’. These terms not only reflect the nature of the user, but also suggest how they evolve and can be used strategically to ‘up-sell’ and ‘cross-sell’ the product, ultimately leading to greater loyalty and product evangelism. These Active segments are defined as follows:

Occasional – “Users who have activated their product experience, yet are sporadic in their usage.”
Regular – “Users who demonstrate frequent and on-going product engagement and usage.”
Advocates – “Users who frequently use the product and who actively tell non-users about its benefits, so spreading word-of-mouth.”
Super Fans – “Users who proactively market the product in an organised fashion.”

Moving users along this continuum – Occasional > Regular > Advocates > Super Fans – is the key to driving repeat adoption, loyalty and evangelism. How do you encourage this? Marketers can look to offer benefits to users who behave in a certain fashion: make multiple purchases, return frequently, or pay a premium, etc. These benefits can take the form of exclusive content, early access to new features or products, discounts, and so on. The marketer should also consider how to make it easy for the user to evangelise about the product to non-users through referral rewards or affiliate programmes; adding content dissemmination and bookmarking tools such as Digg, Reddit, StumbleUpon, etc.; or simply ‘Share With A Friend’ links and incentives.

U – Unstable: 
“Existing users who are becoming erratic in their behavior towards the product and / or disillusioned towards it.”

The central pillar of a CRM strategy is to identify ‘at risk’ users before they end their relationship with the product, and to put in place tactics in order to dissuade them from doing so. VAULT defines this group as ‘Unstable’. Per Active users, the definition of an Unstable user will vary depending on the nature of the product and it’s core objective. Generally, these definitions will relate to the behaviour and / or attitudes of the user. Product satisfaction will invariably be a key component where if this drops below a certain threshold, it is likely that the user will be lost. Using datamining and on-/off-site research techniques, the marketer can track a user’s behaviour and attitudes to ascertain the point at which they become ‘Unstable’. For example, on-site surveys or social listening techniques can track users’ declared satisfaction and align dissatisfaction with specific user behaviour or profiles. These can then be targeted across the wider site audience to trigger corrective measures in the form of messaging or promotional offers.

An extension or alternative metric to satisfaction is the concept of “Net Promoter Scores”. This measures the difference between the proportion of users who give a high response (9 or 10) to the question: “on a scale of 0 to 10, how likely is it that you would recommend us to your friends or colleagues?”, termed “Promoters”, and those who give low ones (0 to 6), “Detractors”. Some companies have found that Net Promoter coefficients correspond highly to customer loyalty and future behaviour and, as such, Net Detractors can help in the opposite manner by identifying potential Unstable users. Both satisfaction and Net Promoter models have been questioned: satisfaction in that it reflects only vague feelings rather than a telling action, and net promoters as the model is unbalanced given it’s scoring system. Used in unison, satifaction and net promoters / detractors can guage the levels of and specifically identify Unstable users; what’s important to the Marketer however, is to follow-up to understand what’s going wrong with their offering and how they can correct this.

L – Lapsed: “Users who have not engaged with the product for a sustained period of time, yet who have not fully terminated their relationship.”

‘Lapsed’ users are those who have dropped off in their activity with the product. This could be for a host of reasons: dissatisfaction, a superior competitor offering, lack of further need for the product, etc. It is important for the marketer to understand why this is the case so that they can take the necessary measures to avoid this with future users. Again, the number of Lapsed users should be tracked based on a time period of in-activity, the threshold set using datamining / research to understand the likelihood that they will return within respective timeframes.

If the user has provided some form of contact details through a registration process, it may be possible to re-connect with them. This action can take two forms: research to understand the reasons behind their lack of activity, or some form of promotion to entice them to re-engage and switch back into the product. The optimum solution would be to use a combination of the two in order to encourage a greater research response.

T – Terminated: “Previous users who have not engaged with the product for an elongated period and have thereby ended their relationship with the product.”

The final outcome in the VAULT progression is full termination. This can be defined based on two possible scenarios: straight closure of an account / request to un-register, or a lack of activity for a period of time at which it is accepted that they will not use the product again. As with the previous stages, it is important to monitor the number and proportion of users terminating, otherwise defined as the product’s ‘Churn Rate’ (for any given period of time, the number of participants who discontinue their use of a service divided by the average number of total participants). A key objective of a CRM programme is to reduce and minimise both Latent and Terminated users. Again, if the user has provided re-contact details, it may be possible to win them back with a reward offer or serve them with a research questionnaire to ascertain their motivations for leaving.

Applying the Definitions

As highlighted, the above definitions will vary depending upon the nature of the product, user profile and lifecycle of adoption. The VAULT model is intended to be built around a level of understanding and insight derived from a programme of datamining and research that enables users to be identified, segmented and tracked such that specific actions can be applied in the form of: follow-up research, site usability design changes, targeted messaging, or tailored promotions. It is this integration of data and activity built around the five stages of adoption and termination which enables the marketer to ‘manage’ users from product consideration, through activation and ultimately to prevent potential deactivation, all of which leads to long-term user retention.

Please feel free to add comments and let me know your thoughts on how this model can be applied and improved.

, , , , , , , , , , , , , , , , , , , , ,

No comments yet.

Leave a Reply

hello@get2growth.com