Just why did you choose 10 data entry people?
It was merely a guess on my part, and I thought I made it clear that I was just guessing. Granted, it's a somewhat educated guess, but one nonetheless. (I've got 12 years developing in large-data applications, I've been intimately invovled in a few start ups, I have 3 patents to my name. I've seen where the money goes in a business.) 10 people, more or less, sounded like a reasonable number, costing about $250,000 a year in salaries. Lets throw in some perks, such as insurance and stuff, and just make another guess that data entry costs $300K a year. If Apple's adding files as fast and furious as they claim, this seems fairly reasonable. How many developers do you think are working on their software and website? 3-5? That's a pretty small number, but if so, we're looking at another 450K a year or so for them...developers command larger salaries than data entry personnel, that's for sure. How about tech support people? 1, a dozen? a score? a thousand? Let's say a dozen devoted to this issue, though that number is probably very small. Again, that's about $300K a year...so we're already looking at a million a year in overhead just in labor costs.
Granted, these are just wild ass guesses, but seem pretty reasonable to me, and account for overhead costs you didn't account for.
In fact, in their first 18 hours they sold 275,000 tracks, so right from the start they're on track to sell 100 million tracks a year.
Oh what utter crap. If they sold a quarter-million tracks a day, sure, they'd do a hundred million in a year. But an AAC file is not a big mac. Everyone who bought a track today isn't going to buy one tomorrow. I'd expect there to be flurry of initial interest, due to the newness and the hype. But, much like a box office opening, things will taper off. After a few weeks or months, we'll have a more realistic expectation of how much business they'll do. But I'm willing to bet it's not as the hyped up grand opening first day. I'd imagine 10-20 million tracks to be more reasonable than to expect the first day frenzy to continue for every day of the year until this time next year. Your estimation of one billion tracks per year is just plain rediculous. It's those kind of 'reasonable' (heh) predicitons that made dot coms fail.
The comparison between bits and furniture is not as absurd as you may think. Then again, a lot of dot-coms who didn't realize that failed because of it. Once you factor out the raw cost of the item (in furniture's case, it's what the manufacturer charges you..in Apple's case, it's what royalties they pay), the costs of overhead are more similar than not. And if you think 34 ipods could act as servers and storage for their entire music service, you're just plain ignorant about how data moves around the 'Net. But, lets use your WAG bandwidth figures. At $300K a month, that's 3.6 million a year, adding THIRTY SIX CENTS A TRACK to the cost of overhead if they sell 10 million tracks a year. If they instead move your much inflated guess of 100 million tracks, that's still nearly 4 cents a track in overhead that you didn't account for at all in our "2 cents a track" guess. If we split the difference and guess 50 million, it's still nearly 8 cents a track in overhead, 400% larger than your initial guess.
what do these 10 data entry people of yours do exactly?
Uhm, enter data. Or do you think Apple is going to keep only the tracks they have and never add another song to the service? It's my understanding that they're expanding the service rapidly. Or perhaps the Music Fairy will sprinkle their 34 ipods with her pixie dust and the music will magically appear there. You ever rip a CD? 2-3 minutes per track is around what I average, depending on song size, and accounting for time shuffling cds in and out of drive trays, lets say that someone can digitize 150 or so tracks a day. Granted, if they get the files pre-ripped, there's less "rip time" but the data (song names, titles, tracks, etc) still has to be entered into their database to link up on their service. And someone still has to eyeball the stuff and decide prices (since by all indications pricing is not uniform), so ripping or not, spending a couple minutes per track seems reasonable. Now divide those 200,000 tracks by 150 and you get 1300 man-days. Put 10 people on it, and it takes about 5-6 months (guessing about 22-25 work days per month, no time off for vacation or sick days). One of the biggest complaints I've heard is that the database is shallow...not very many complete albums. You think they're not going to keep adding songs at a quick pace? And again, that's not counting tech support, customer support, software developers, maintenance, DBAs and all of the salaries that go with that. These higher-paid, more skilled workers will easily cost much more than the data entry people below them. And then factor in the costs of middle and/or upper management to oversee it all. You think those managers make the same as the data entry people? Even without doing further math, it seems pretty easy to see that salaries will take up about 20%-30% of their sale price. Which puts things exactly in line with a traditional, realisitc business model. Add in costs for the structural overhead (server/database clusters, etc, utilities, cost of rent or purchase of land to house the facilities and employees) and it seems reasonable that the price per track on 20-30 million tracks a year might be as high as 40-50 cents a track. You could add another 10 or 15 cents a track for advertising the service, but that might be somewhat offset by the advertising revenue you're talking about. But maybe not...as the dot-coms found out, advertising on the internet doesn't really bring in all that much money. You seem to be stuck in dot-com math, where nothing digital costs anything, and revenues are 50 times realistic expectations. Of course, we all know what happened to the dot-com bubble.
Now, so far I've been assuming your 5-cents-per-track royalty is accurate. Based upon what I've seen so far about your ability to guess real-world business numbers, I wouldn't count on it though.
[ Parent ]