I read an interesting post about “Free” as a component of a business model. I encourage you to go read the whole post on Techdirt, but the bit I’m interested in is here, where Mike Masnick presents, and then disagrees with, an argument against the use of “free” –
Unfortunately, both Pogue and Poole then use this to bash the entire concept of free-based business models, with Poole getting unnecessarily offensive in his response:
“I’ll call it, for short, “the Slashdot argument”. It says that books, music, films, software and so on ought to be freely distributed to anyone who wants them, simply because they can be freely distributed. What is the writer or musician to do, though, if she can’t earn money from her art? Simple, says the Slashdotter: earn your money playing live (if you’re one of those musicians who plays live), or selling T-shirts or merchandise, or providing some other kind of “value-added” service. Many such arguments seem to me to be simple greed disguised in high-falutin’ idealism about how “information wants to be free”. Perhaps it’s not empty pedantry to point out that “information” doesn’t want anything in and for itself. The information in which humans traffic is created by humans. And most information-creating humans need to earn dollars or yuan to survive.
While I’m sure there may be some Slashdot-types who may make this argument, it doesn’t mean that it’s an accurate representation of the more important discussion of these business models. The main problem is his use of “ought,” as in people saying things “ought” to be free. It’s not that things ought to be free because they can be free — but that things will be free because that’s just basic economics. Price gets driven to marginal cost in a competitive market, and the reason it happens is because others do learn to put in place business models that work, and then if you’re the lone holdout, people start to ignore you.
When you unpack this, I think there are really two claims that Mike is making:
- When the marginal cost of something becomes close to zero, economic forces will make it impossible to charge money for it.
- Digital goods qualify as “something close enough to zero” that economic forces will make them free.
I’d agree with those statements if humans were “homo economicus” (rational beings that always behave in their best interest), but we’re not. The value of something is what people will pay for it, which is based largely on what people feel it is worth.
People will pay for digital goods. What’s the marginal cost for Flickr to add a new Pro account? Zero. What do people pay? $25/year.
Almost free and free are totally different things, and if you take the cost of almost nothing and multiply it by millions, you end up with a big cost. You need to be able to sell a lot of other stuff to make up for the cost of giving things away free.
You’re especially in trouble if your product goes rotten (old news), and it requires scarce goods (work) to create the abundant goods (digital news).
If that’s your production method, the real cost of your “free” product isn’t all that close to zero after all, which throws a huge wrench into the idea that market forces will force the market price to $0. It won’t do that if at a $0 price point, all news companies go out of business.*
Again, it’s not a matter of “should be free,” it’s a matter of economics; if the real cost of producing it turns out to be noticeably higher than zero (and I’d argue that it is), then it won’t be free.
The really scary thought is that even if the news business can make good money selling people’s attention to advertisers, partners, and data companies, then the business isn’t about selling news anymore.
And, well, there are much cheaper ways to build an audience that will support the real business (selling ads, events, who knows what else), and then the need for news as a method for building an audience goes away.
Worse than that, those organizations that try to use news as an audience-building method will find themselves at a terminal competitive disadvantage to those organizations that use cheaper audience-building techniques.
In a sense, the news industry has been using “free,” for decades, since subscription fees barely factor into print revenue – the real killer is that the cost of a newspaper’s true product (reader attention) has plummeted as the supply of attention-grabbing-material has blossomed.
What’s the answer? You change tactics and build a business model selling digital goods. If you really care about the news and not just the ads, then forget large audiences. Build a product that appeals to a smaller number of people in such that they will pay for it.
“Give it away and pray” is not a working business model. But for the news business, neither is “give it away for free and make money on the ads.”
To get more opinions on the future of the news business, subscribe to the Eat Sleep Publish RSS feed – always free!
—————-
* I’m well aware that many content producers on the internet are making money by selling ads against free-to-the-consumer content. That seems quite possible for small operations, but I don’t think it will ever scale well, and I do think that a significant portion of the news necessary for democracy requires a certain amount of scale to produce. ( return )
Eat Sleep Publish is a Seattle-based blog that explores the difficulties and opportunities currently facing the publishing industry. 
{ 2 trackbacks }
{ 5 comments… read them below or add one }
First… that post is pretty old, but thanks for the link and the interesting discussion.
I’d argue a few points. Value is not what people will pay for it. That’s the market price. Value is what makes up the demand side of the demand/supply curve, but that’s only a part of the equation. If the supply is effectively infinite, value doesn’t matter.
Air? Incredibly valuable. But also abundant. So you don’t pay for it.
Second, you confuse marginal costs with average costs in your discussion on the cost of producing news. You are absolutely right that economics doesn’t care what something “should” cost — and I’m quite careful to make my point clear: I’m talking about what the market decides, not what should happen.
But you lump in the fixed costs and assume that said fixed costs preclude a price of zero. That’s incorrect. It’s the marginal cost — not the average cost — that determines where price will end up under competition. The mistake you’re making is looking too narrowly at the market and not realizing the power of bundling.
I agree that subscriptions haven’t really been providing the revenue for newspapers for ages, so I’m not sure why you think that will suddenly work in a MORE competitive environment. There are plenty of ways to make money. Trying to sell something that is abundant is not a good one.
Finally, you claim that selling ads on free to consumer content can’t scale. Tell that to… Google. Or broadcast TV stations. Or radio stations. It’s scaled for ages. I’m not sure why you suddenly think it no longer works.
Mike – I know it’s an old post – Shirky pointed to it on Twitter a day or so ago. Thanks for dropping by to comment!
I’m aware that my econ is based on a course or two in college, but yes, I’m sort of intentionally confusing marginal cost and average cost. In a sense it’s irrelevant, because the marginal cost of serving another news article isn’t zero…it’s almost zero, so the price shouldn’t be zero anyway.
Two other points:
Why will subscriptions work now? – They won’t work as they are currently imagined. They need to be priced so that they pay for the cost content creation, and content should be distributed much more cheaply via the internet.
I’m actually agreeing with you that newspapers cant compete in the “google” environment when they have to pay for content creation, so they should enter a new businesses. Which leads me to point #2…
Scale -
Google works at scale because they don’t have to pay the cost of creating the content against which they serve ads. This is an excellent model, but one that does not apply to a professional publication that wants curated writing, pictures, video, etc. If you have to pay for content creation, then no scale, because every new page you can serve ads against costs you more to product than you can make from it in ad revenue.
Broadcast TV works because they’re still selling a scarce resource (air time) in a situation where people pay more attention (TV on a TV). Same for radio.
Scale doesn’t work when online ad space is essentially unlimited (so you can’t sell it), and you still have to pay to produce the content.
Hey Jason…
Thanks for the response…
I’m not sure I agree that the marginal cost is “almost zero.” Well… it really depends on the bandwidth contracts, but I’d argue in many cases it absolutely is zero. You visiting Techdirt doesn’t cost us anything extra (normally).
Definitely agree that subs won’t work as imagined now, but I think you’re still making an econ mistake when you say “They need to be priced so that they pay for the cost content creation,” That’s accounting issues, not economics. You’re too focused on covering the costs directly, rather than providing value combined with a complete business model that makes sense. Subsidized/bundled content production is a model that ’s worked for ages if you know how to subsidize and bundle properly.
On the scale question, you still seem to be confusing fixed and marginal costs. Once the content is created it can be reused at an incredibly low (sometimes zero) marginal cost.
Mike – I think my confusion re: scale could be better explained with numbers, so let me give it a go. Let’s say my business model requires that I create content and give it away for free online at scale as best I can.
Here’s how my fictional costs break down by month:
$100/mo for 500GB bandwidth, $0.25/GB beyond that (Rackspace)
Staff of 10 people at average of $50k/year, $500k/year, ~$41k/month
The bandwidth becomes a rounding error at this stage, but I’m going to assume I’m paying at least $300-$400/mo for the servers.
Now let’s assume I’m smart and I sell my ads CPM instead of CPC for an average of $12 CPM, which means a page view earns me 1.2 cents per ad. I get four ads on a page so I make 4.8 cents per page view.
Let’s say I’m doing really well and I serve 800,000 pageviews every month. At those numbers, I earn $38,400. I’m losing money, despite the fact that every extra pageview is “free to me.”
The other un-accounted for cost in this whole setup is that *generating* pageviews is not free either; this assumes I do no marketing whatsoever, pay no office rental, no equipment costs, no software costs, no taxes, and that the work of ten people doing 40 hour work weeks would generate enough content to earn 800k pageviews (which I’d be damn surprised if they could do that consistently).
I just don’t see how the math works. Marginal costs are great, except that the numbers going in and out of your bank account are based on average costs, so it seems dumb to build a business model that assumes the rest of the costs just don’t exist…
Again, you’re focusing on avg costs, not marginal costs.
I agree that the business numbers you put forth is a bad business. No one said that pricing at marginal cost makes a bad business into a good one. You are confusing two separate issues.
My point is that trying to charge for the content doesn’t magically make your bad business into a good one.
But trying to ignore the marginal cost in a competitive market makes a bad business into a dreadful one. If you *understand* the marginal costs, and the nature of the market, then you can try to put in place a business model that makes sense.
So, in the above example, you recognize the return on investment that you need… let’s say $750k based on your numbers. Then, you need to come up with a model that will let you get that amount. But that doesn’t mean you have to charge for content or that ads won’t work. It just means you know what target you have to hit.
But that’s a different discussion than whether or not ads scale or whether or not you can charge for content.
I’m not saying you shouldn’t understand your fixed costs. You absolutely need to. But the price of an individual good is NOT determined by the fixed or average costs in a competitive market. That’s because if you try to price based on your fixed costs, someone else will undercut you, and you’ll be in trouble.