Everyone loves pricing tiers

We’ve all seen this type of image on just about every product or service website:

It’s a common belief that most of your customers will fall into your middle tier, so you try to guess at what to put in that tier, maybe run some experiments and hope to get as much traction as possible. But how do you upsell users into higher tiers? What would happen if you moved some features to higher packages? Would people follow and pay more? How many would stay or even downgrade service?

By analyzing user behavior data, we can actually determine the likelihood of all these scenarios giving you a much more accurate understanding of outcomes. You can better optimize for price and volume within your tiers and extract the most revenue while still keeping customers happy with your service.

You don’t pick your cable package, it picks you

One area where this type of analytics can be easily understood is with cable TV packages. Cable companies pay fees to carry all the channels from the distributors. Those fees and contracts vary, so the cable company wants to bundle content together in a way to maximize both the appeal of those packages and the price they are able to charge. TV is extremely subjective by the end consumer, so it’s impossible to create a commercially viable package that fits all your customer’s needs. So how do you optimize?

One approach is to analyze the viewing data from all customers and to group them using a clustering algorithm. You can then see what segments exist across your customer base and within existing packages.

The real magic then comes in understanding, through further analysis on viewer preferences and licensing fees, which channels could be moved to higher tiers without losing subscribers and therefore generating more revenue.

For example, if your household has kids that exclusively watch Paw Patrol, the cable company could, hypothetically, move Nickelodeon (the channel that carries Paw Patrol) to a higher tier and you are likely going to continue right along paying for that higher tier package to avoid the wrath of toddlers.

This method is especially effective at analyzing the long-tail channels, not necessarily the Nickelodeon’s of the world. You can identify “super viewers” of those long-tail channels and move them to tiers that help maximize revenue. By also seeing trends in the segments of users, you can also create higher customer value by grouping like-channels together. So yes, you might charge customers more for the Bull Riding channel, but if you group it with say the Outdoor channel, you might find those customers are happy to pay for that content with a lot of overlap.

The best part, you don’t have to guess. The segment and behavior analysis will tell you exactly where these super viewers are and what else they might like.

Other applications

This works for all sorts of products and services where tiered pricing strategies are at play. You can look at what features people use the most in each tier, identify the segments within and develop new tiers or move features or pricing around to optimize your revenue and customer satisfaction.

Take cloud storage as an example. All the consumer cloud storage players have tiers mostly differentiated by storage volume. But there are also a host of features and integrations that come with varying tiers of service.

Strategically, a company like Dropbox will want to keep that main differentiator of storage competitive with the market, but the real stickiness is in the features used. They could move mobile app access or Google Docs integrations to a more advanced tier and likely see a small loss in customers, but a large jump in revenue.

Again, if the analysis of user behavior is done correctly, you don’t have to guess. You’d know with some certainty which users would upgrade and how much more revenue would be generated.