Lifetime Value Part 30: Why clumpiness should be one of the KPIs you focus on

When calculating customer lifetime value (LTV),there is one KPI that is often neglected, clumpiness. Clumpiness is a term coined by Eric Bradlow, a marketing professor at Wharton. Think of clumpiness as binge-watching Ozark (or, if you must, Tiger King). Rather than an extended, constant period of consumption, a short, intense buying burst. By understanding and tracking this behavior, you have a significant weapon for improving LTV.

Defining clumpiness

Clumpiness refers to the fact that people buy in bursts and that those customers could be extremely valuable. When calculating customer value and segmentation, we focus on analysing recency, frequency and monetization of the customer (what Bradlow refers to as RFM). This analysis is based on customers making purchases in a regular pattern, i.e. coffee, diapers or milk. Bradlow’s analysis, however, shows that for certain products (and I would classify social and casino games here), customers actually monetize in bursts. Thus, you need to add C for clumpiness to your RFM modeling.

Why clumpiness is important

Bradlow researched how to predict the future value of customers (predicted LTV). Bradlow explains, “[l]et’s imagine you want … to predict who are going to be the valuable customers in the future. And you have four things you can use to predict it. As I mentioned: recency, frequency, monetary value and let’s say the marketing spend towards the customer. Those are the classic ways in which companies build what are called scoring models…. [T]he findings of my research suggest that higher clumpy customers are worth more out of sample, meaning in their future value, even after controlling for RFM and marketing expenditure — which means we have found another variable that firms should track [concerning] our customers and use it to predict their worth in the future.”

This finding is particularly interesting as most companies, especially in the game space, currently base their LTV projections on the RFM model. While this calculation may have worked in the traditional retail economy, consumption has evolved, especially for digital goods. Binge consumption is a fact of life in the entertainment space, and gaming sits squarely at the center of the modern entertainment environment. This analysis is consistent with Bradlow’s findings, where he says, “[i]f you look at historically purchased goods, clumpiness really isn’t there. But if you look in the new wave, the new economy, clumpiness is pervasive in every data set I’ve analyzed.”

Using clumpiness insights

Calculating clumpiness should be easy and not require tracking any new events. It is the same data you are using to calculate R, F, M and LTV. There are then several applications for this insight:

  • Incorporate it into your segmentation to get a better understanding of who your VIPs and high value players are, then focus your premium treatment (and benefits) on these players.
  • Use clumpiness to predict better what players are likely to become VIPs. This will help you concentrate your early retention efforts and reinvestment.
  • Focus reactivation on your clumpy players. Bradlow explains, “[i]f you reactivate them, they’ll come back and be clumpy again, and do a lot of stuff in the future. “
  • Do not simply rely on the KPIs you have always been trusting. Do not rely on simple theories of how your players behave, also look at things about arrival time or when people play. People who come in bursts, then go away and then come back in bursts and then go away are fundamentally different and have a different LTV.

By understanding clumpiness, you have a more accurate predictor of LTV. Once you know how to predict your LTV, you can then impact it by making changes that drive these variables.

Key takeaways

  1. We normally focus on analyzing recency, frequency and monetization of the customer but by adding a new KPI, clumpiness, we get a much better understanding of their expected value.
  2. Clumpiness refers to the fact that people buy in bursts and that those customers could be extremely valuable.
  3. Clumpiness can help you better segment players, predict VIPs and target your reactivation efforts and spend.

Maybe, just maybe, social casino, movie and iGaming will continue to rise post lockdowns

Most people, myself included, believe that the spike in iGaming, social casino and streaming revenue over the past few months will revert back to pre-pandemic levels once casinos and cinemas reopen, however, we may be in for a surprise. Rather than lose the 20-50 percent revenue growth these three sectors have enjoyed, a return to normalcy may actually continue to accelerate revenue.

There are two principle drivers, somewhat related, why companies in the social casino, iGaming and streaming media spaces may be pleasantly surprised: the power of triggers and the weakness of cannibalization.

Triggers are a central driver of activity

The key to building a successful product is creating something that becomes a habit for your customer. Habit-forming products change user behavior and create unprompted engagement. The aim is to influence customers to use your product or play your game on there own, repeatedly, without relying on overt calls-to-action such as ads or promotions. The master of creating habit forming products is Nir Eyal, author of Hooked, who created a model to show how to create such products .

Hook Model

The first step of Eyal’s Hook Model of retention is triggers. Triggers cue the user to take action. There are two types of triggers: external and internal. Habit-forming products start by alerting users with external triggers like an email, a website link or the app icon on a phone. An external trigger communicates the next action the user should take (for more on triggers and how to build them, please see this post).

Triggers show how the reopening of casinos and cinemas can help iGaming and online social casino as well as streaming entertainment, respectively. Going to your favorite casino may trigger in players a desire to continue gaming online. Seeing an advertisement on television for Las Vegas or the local Native American casino will not only create interest in going to the casino but also will remind you to go online to play. Same for movies, going to the cinema to see the latest Disney release may encourage you to spend more time watching Disney+.


Cannibalization or not

I find cannibalization one of the most over-hyped concerns, and it relates to how the social and online gaming spaces are impacted by the reopening of casinos (same for cinemas). Industries, and the economy overall, are not a static pie but a dynamic system that grows (and contracts) based on the interaction of numerous factors. I have been in multiple situations where my company cancelled or throttled new products to protect their existing business. What they ended up learning was that the new products growth actually helped the existing products or competitors came into the market and stole the audience that would have gone to the new product.

The casino space is rife with this thinking. When I first entered the social casino space in 2013, it was virtually impossible to form relationships with land-based casinos. They feared their players would flock online and it would cut into their revenue. Over the next year or so (2013-2015), more and more data showed clearly that land-based casino customers who also played online (in that situation, with social casinos), became more valuable to the land-based casino, not less. Land based casinos became very anxious to form relationships with social casinos because the players not only played both but were the most valuable customers. The MGM-Play Studios relationship behind MyVegas reflects how this attitude evolved.


I saw a similar phenomenon in the real money online gaming space. Players who played for both Real Money and Play Money were one of the most important segments of the ecosystem. Rather than trying to move the Play Money players to real money, companies realized that it was valuable to encourage continued playing on the Play Money and social products.

Pokerstars play mone

A similar phenomenon may occur as land based casinos re-open. Rather than the customers now playing online shifting their spend to land based, these customers may turn into VIPs for both land-based and either social or real money online casinos (depending on the geography).

We just do not know

I am not in the business of predictions and the Coronavirus pandemic is such a unique situation it is impossible to look into the past to know how it will play out but it is worth considering both triggers and cannibalization to understand one future scenario. Rather than seeing the movie or casino market as set at a certain amount with closures and re-openings just shifting where revenue goes, the dynamism of the market could help both sectors grow.

Key takeaways

  1. The re-opening of casinos and cinemas may not hurt online gaming (social and iGaming) or streaming entertainment but continue their momentum due to triggers and the realities of cannibalization.
  2. Triggers, cues for the user to take action, are central to creating habit forming products; land-based casinos and cinemas could serve as triggers for customers to consume more gaming and movie entertainment online.
  3. Cannibalization, one product taking sales away from another, is over-hyped in the casino space and rather than online gaming hurting land-based casinos, past experience shows they help each other.