It seems that each quarter this debate arises. One company counts one way and the other counts another way. One company has longer OEM initial periods, and the other has shorter. One company pays a low revenue share and the other pays a high revenue share.


What matters is the BOTTOM LINE. However, in order to satisfy the unyielding demand for debate, here is a breakdown of how subscribers are counted.


This category of OEM subscriber is used with Sirius partners such as Ford and Chrysler. These auto manufacturers pay for a subscription at the time the car rolls off the assembly line. Because the subscription has been paid for, Sirius counts these radios as subscribers. These vehicles then ship to dealers and are sold. During the time which they sit on a dealer lot, these radios do not generate "countable" revenue. The revenue from the OEM payment received by Sirius is deferred until such time that the consumer who buys the car starts receiving payment. The benefit is seen in cash flow and churn. The negative is in a lower ARPU. The key is understanding the various components of this type of subscriber. Analysts such Jacoby say that the current churn numbers and subscriber numbers are not "real" due to the cars being on lots. However, did Jacoby address that ARPU is lower than it would be in this same situation? No, he did not, and therein lies one of the problems in this sector. A general lack of understanding of how various deals and deal structures impact certain metrics.

For the initial subscription, be it 6 months or 1 year, the subscriber is not counted as a self paying subscriber.

Sirius subscribers who elect not to continue service are counted as churn.


This type of deal comes from the early days of OEM installations. A radio is installed in a car, and when that car is sold, it is counted as a subscriber. In the case of GM, payments for 2 out of the three months are made, and the third month is provided by XM. during this period the subscriber is not counted as a self paying subscriber. The benefits to this are that the OEM is paying part of the initial subscription that exposes the consumer to satellite radio. The downside (at least for the GM deal) is that the revenue share is quite high, and the length of the trial is shorter.

In XM's case, a subscriber who does not elect to keep the service in this category in NOT counted as churn.

This is a newer metric. In some cases, a consumer buys a car and receives a period of free service of either Sirius or XM. The service is not subsidized by the OEM. After the trial period, if a consumer keeps the service they are counted as a subscriber. If they do not keep the service they are NOT counted as churn. The benefit here is that the churn number is not impacted in any way. The negative is that there are no guaranteed dollars recovered for the subsidy paid for the radio installation. The revenue share on these types of subs is typically much lower than the types of deals above. Both Sirius and XM have OEM deals that fall into this category.

There is no right or wrong way count. However, there are clearly differing methods at play, and understanding them gives an investor some insight as to why certain metrics between Sirius and XM simply can not be realistically compared. Metrics impacted include, churn, SAC, CPGA, subscribers, self paying subscribers, ARPU, and others. Investors need to bear all of this in mind as they draw comparisons.

Some deals have higher subsidies, but also have higher prepayments from the manufacturer. Other deals have a lower subsidy, but a higher revenue share. The key is knowing as much as you can about such deals, and how they impact the bottom line.

All of that being said, it still boils down to the bottom line. That is where we see whether the business models are viable. With a potential merger on the horizon, all of this will become less relevant anyway, but that is another conversation for another day.

Position - Long Sirius, Long XM