Sirius XM Holdings (NASDAQ:SIRI) reports a variety of non-GAAP measurements called "Key Operating Metrics" that are published in its annual and quarterly SEC filings. These reports state:
In this section, we present certain financial and operating performance measures that are not calculated and presented in accordance with generally accepted accounting principles in the United States ("Non-GAAP"). These metrics include: average monthly revenue per subscriber, or ARPU; customer service and billing expenses, per average subscriber; subscriber acquisition cost, or SAC, per installation; free cash flow; and adjusted EBITDA. These measures exclude ... [a list of adjustments used in these calculations has been omitted for brevity] ... We use these Non-GAAP financial measures to manage our business, to set operational goals and as a basis for determining performance-based compensation for our employees.
We believe these Non-GAAP financial measures provide useful information to investors regarding our financial condition and results of operations. We believe investors find these Non-GAAP financial performance measures useful in evaluating our core trends because it provides a direct view of our underlying contractual costs. ... By providing these Non-GAAP financial measures, together with the reconciliations to the most directly comparable GAAP measure, we believe we are enhancing investors' understanding of our business and our results of operations.
Sounds great, but unless an investor carefully reads the definitions and tracks changes to these measures, they can be very misleading and give investors a false impression of the performance of the company. This article will focus on one of these metrics, subscriber acquisition cost, or SAC, per installation. It is, in my opinion, not only poorly named, but also one of the least understood metrics published by the company. In a recent article titled OK, I'll Say It: Sirius XM Had A Great Quarter, the following item was highlighted as one of the reasons Sirius had a great quarter.
SAC Per Installation
• 2014: $33
• 2015: $32
The cost to acquire a subscriber is an important number, and it is good to see it continue to decline even if that amount is only around 3%, or $1 per subscriber.
As ARPU inches higher and SAC inches lower, the time it takes for a new subscriber to pay for itself becomes less. That's 'good.'
It's true that the cost to acquire a subscriber is an important number. I'll go one step further. It is a critically important number. Unfortunately, that's not even close to what this number measures. Here is how Sirius calculates this metric, and describes the term, in the most recent 10Q:
Subscriber acquisition cost, per installation - or SAC, per installation, is derived from subscriber acquisition costs and margins from the sale of radios and accessories, divided by the number of satellite radio installations in new vehicles and shipments of aftermarket radios for the period. SAC, per installation, is calculated as follows:
Tap to see the table
"Subscriber acquisition costs (GAAP)" is a line item in the income statement, and it is described as including and/or excluding the following:
...hardware subsidies paid to radio manufacturers, distributors and automakers; subsidies paid for chipsets and certain other components used in manufacturing radios; device royalties for certain radios and chipsets; commissions paid to automakers and retailers; product warranty obligations; freight; and provisions for inventory allowances attributable to inventory consumed in our OEM and retail distribution channels. The majority of subscriber acquisition costs are incurred and expensed in advance of, or concurrent with, acquiring a subscriber. Subscriber acquisition costs do not include advertising costs, marketing, loyalty payments to distributors and dealers of satellite radios or revenue share payments to automakers and retailers of satellite radios.
Note that this line item is mostly about the subsidies paid to radio and auto manufacturers prior to acquiring a subscriber, and excludes many of the costs one might typically think of as the cost to acquire a self-pay subscriber, such as advertising, sales, marketing and telemarketing.
An examination of the income statement indicates that the line item Subscriber acquisition cost is the single largest expense for Sirius, and it continues to increase. We can see that the main reason that SAC, per installation, fell is because the total costs in the subscriber acquisition cost line item for Q2 increased by 9.7% from $124,407,000 to $136,504,000 while the number of installations increased at an even faster rate, 11.4% from 3,280,000 to 3,655,000.
More importantly, the $32 or $33 is clearly not a measure of the cost to acquire a subscriber. It is mostly the cost or subsidy to induce an OEM to install a satellite radio in a new vehicle or help a radio manufacturer acquire parts. It doesn't matter whether or not that installation results in acquiring a new self-pay subscriber.
We also know, based on the new vehicle conversion rate, that three out of five of these OEM installations won't result in the acquisition of a self-pay subscriber. One might choose to argue that a paid promotional trial is considered a subscriber by Sirius, but that still ignores the fact that there are many OEM installations that become unpaid trials and never become any type of subscribers.
I recently wrote an article about the churn statistic which describes another scenario that investors should consider. When a current self-pay subscriber trades in their vehicle for a new one that comes from an OEM participating in the paid promotional trial program (as compared to an OEM that does not pay a fee for the trial), they are not considered part of the churn statistic. Still, that self-pay subscriber's new vehicle is an installation, it increases the denominator, and helps to reduce the SAC, per installation. Does it make sense that any current self-pay subscriber that purchases a new vehicle be considered part of the calculation to acquire a subscriber? Each of these purchases by someone that is already a self-pay subscriber is included in the denominator for the calculation of the statistic and serves to reduce the average of SAC.
And it's not just that the denominator is overstated (which reduces the average cost) by including all installations without regard to whether someone new becomes a subscriber, but it is understated when one considers what occurs with the used car programs. Any time Sirius acquires a subscriber through ServiceLane, or a trial that comes with the purchase of a used car, or in any manner other than a new vehicle purchase, that acquisition is ignored.
On the Q1 conference call the company talked about running approximately 16 million trials, and based on 17 million new vehicle sales and a 72% penetration rate, that indicates that there will be about 12 million new vehicle installations and trials. It also means that there will be close to 4 million used car trials. Since we also know that used car trials convert in the low 30% range, that means that there will be more than a million subscribers acquired via the reactivation of used vehicles. Counting these acquired subscribers would, in a sense, bring down the average cost, but are ignored.
Summary
The name - Subscriber acquisition cost, or SAC, per installation - leads an investor to believe that it only costs the company about $32 to acquire a subscriber. When that investor then looks at the ARPU, or average revenue per user, of more than $12 per month, it seems that the company recovers the cost to add an incremental subscriber in just a few months. Sounds great, doesn't it?
Subscriber acquisition cost, or SAC, per installation, may be a key operating metric, but it is wrong to think that "as ARPU inches higher and SAC inches lower, the time it takes for a new subscriber to pay for itself becomes less." It's wrong because it simply isn't the way the metric is defined. SAC, per installation, never measured the cost to acquire a self-pay subscriber or even what Sirius considered a total subscriber. At best, it is an indication of the per vehicle net subsidy paid to the OEMs to install a satellite radio in a vehicle.
The subscriber acquisition line item expense on the income statement is allocated to all installations, and not just those that become subscribers. The calculation doesn't factor in acquiring a subscriber from the used car channel. The calculation excludes many expenses that one normally considers part of the cost to acquire subscribers.
While management may believe they "are enhancing investors' understanding of our business and our results of operations", I think they are only confusing and misleading investors. A more meaningful number for investors might be to divide the subscriber acquisition cost line item by the self-pay net adds. That would pick up the net additions from used car channel, even if it didn't include all the marketing and sales expenses. Of course that would also show a payback that takes more than a year.
Management also stated "We use these Non-GAAP financial measures to manage our business, to set operational goals and as a basis for determining performance-based compensation for our employees." Maybe giving bonuses to employees helps explain why management included this particular "Key Operating Metric." It certainly can't be to help investors understand the business.
Disclosure: I am/we are long SIRI.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: In addition to my long positions, I have January 2016 $4 covered calls written against a portion of my long positions and January 2016 $4.50 covered calls written against a majority of my long positions in Sirius XM. I also continue to make frequent short term trades on large blocks of Sirius XM on a regular basis. I may close the current call positions, open new call positions or buy or sell large blocks of Sirius at any time. I hold no positions, nor do I have any plans open any positions, in any other company mentioned in this article.
