I polished off Chris Anderson’s, Editor of Wired Magazine, new book Free the other week. Releasing a book in the current media climate, with papers folding everyday and magazines (including Wired) suffering from an ad drought, arguing that everything should be ‘free-er’ still is heresy. To be sure, it was a good read, but it was incomplete.
The biggest critique on Anderson is that he doesn’t have the ‘Freeimum’ answer for the dying publishing models of today, he only points to a few examples like Google as (imperfect) case studies. Most notably, Malcolm Galdwell made just this point in his review of Free for the New Yorker. While Google offers YouTube for free, it doesn’t actually produce any profit. The server cost alone would swamp any lesser company. So if Google, Anderson’s poster child for Free, can not find profits from a freeimum model, doesn’t this cast doubt on the whole Free argument?
Just to be clear Anderson is not really saying that everything should, or ought, to be free. He is actually very specific to say that data and information want to be free. His view is that future business models will have to seek revenues from a smaller core group of individuals who are willing to pay higher premiums for ‘value-added’ components. Think of a free game that asks you to pay for the 2nd level. The base is free, or cheap, but the extras are pricey. (Over simplified)
So that got me to thinking: if I accept Anderson’s argument on face, and I do, what would be my ‘Freeimum’ solution to replace the old publishing model?
At first I focused only on Wired, but each idea I came up with applied to any other lifestyle publication. Although the audiences change, the economics do not. All lifestyle publications seek to create a brand which will attract a specific audience to whom they can sell to advertisers.
The publishing business’ problems are structural. In a nutshell, print display adds are priced 4-5X higher than an equivalent web ad. Even though paper, printing, and delivery are ½ a publishers costs, with the loss in revenue from print to digital ad sales, a firm faces a 30% structural deficit. Anderson makes the argument that as technology expands shelf-space and reproduction costs approach zero. Unfortunately, this misses a key distinction, marginal cost approach zero, but the fixed costs of discovery are as real today as they ever were. And those costs are not being met in the current digital environment. More to the point, on the internet high-traffic loads carry a disincentive to the publisher because of the exploding server costs. It’s choking off the NYT’s and bleeding the balance sheets of every major publisher.
My Freeimum solution would involve five different planks. First and foremost, new methods need to explored which can demand a higher price-per-impression for digital display ads. Related to that would be an offering a pay-per click option to have your company and products linked throughout the site. Next, I would move to diversify the revenue streams by building an affiliate retail operation. Most importantly, I propose a new ‘freeimum’ membership community that will be centered on regularly sponsored events in locations across the country. With these four things first implemented, I would then create a set of bundled ad packages, comprising print pages, web ads, web links, event participation ect.
Better Faster Stronger
Magazines need to earn more money from digital ads; in particular it needs to extract a bigger premium from their display based ads, because they have the highest operating margins.
Today’s display advertising on the web is ugly, small, and cheap. It does not make a strong impression, if it makes an impression at all. Between Ad Blockers, conditioned blindness, and irritation with ad severer times the actual impressions made, as opposed to served, are a fraction of the page view number.
The best parts of print display ads in magazines are the serendipity of finding something new and beauty of a full-bleed 4C pictorial, on the web that gets lost. In an effort to recreate that type of interaction I propose ‘pre-loaded transition ads’.
Delays in load-time resulting from the ad-serving process can infuriate people, almost as much as the sight of a blank page. What I propose is that on any given page a script is run, after all the main/editorial content has loaded, that loads a ‘full page ad’ eg 800X600. However, the ad isn’t shown until a new link is clicked, when that happens, the picture is displayed for a time, while the new page is loaded behind it. Once completely loaded, the ad disappears. While clickable, there is no skip, no count down, if you click back the ad slides away, otherwise the ad stays up until the page is ready. The key is in the ‘pre-load’. We don’t add any additional time to the transition; and since we are not showing the ad on the page itself, the load should be quick.
The result is a ‘piece of eye-candy’ that enhances the user experience, specifically the transition, without degrading the primary functionality of the site or the browser. Implementation of this is of course key to its effectiveness, it could easily be abused and thus create a worse user experience. Done correctly, and you have a large format digital display space, with little to no loss in site traffic.
But what if a person doesn’t click on any link, and simply closes the page, you haven’t served an ad?!
The flip to transition ads is site skinning. This was brought into the fore by Denton’s Gawker collective, where you allow advertisers leeway to control major design aspects of the site, including sidebars, backgrounds, and more. This is an infant of an idea, but one that I think will have legs going forward.
Skinning will take on a new dimension when, IMHO, over the next 3 years wide screen web-design will start to become the norm. With that added real estate, skinning takes on new possibilities regarding large pictorial background images, moving ads, and more. This is especially true if other browsers start adopting Google Chrome’s JavaScript engine in whole or part. (Safari and Firefox will adopt it in the year or so)
My vision of this plan would incorporate the skinning with the transition ads together. The feel should be that the ads are a seemless part of the UI. Not annoyance but a reason to view the site in and of themselves. Both approaches give the advertisers a better canvas and the power of exclusivity. These benefits justify a higher CPM. Of course this means not having the banners, skyscrapers, and footer ads they have now, but the higher price point is to their benefit in the long run. In this case what’s best for the user, is best for the bottom line, because it’s better for the advertisers.
Contextual Linking: aggregating the clicks
I’ve stated before that clicks are fungible commodities. A click from Cosmo is the same as a click from National Geographic, or Google, from the advertiser’s perspective. This attribute gives advantages to those with size. The obvious example is Google, but there are many ad networks that work under a similar premise. Using a behavioral algorithm, ad networks serve ads targeted to individuals, but across many sites. An example would be getting a Travel ad on a page talking about baseball, that was chosen because you visited Expedia earlier.
This mathematical targeting requires a wide network, and millions of data points to function effectively. No classic publisher is going to reach the level of views needed to make this work on their own, and allowing an ad network to serve the ads cuts deeply into an already thin margin. And it doesn’t matter…
Typically a publication has a core group of a few dozen advertisers. And while publishers charge for banner ads on a CPC basis, the editorial staff gives away free links everyday to the same companies. Not only do they give away the exposure, they give it away in the most prime real estate, in the context of an article.
My solution would be to offer advertizing companies the option of ‘fully integrated contextual linking’ throughout the site. What I mean is that companies can buy the right to have brand key words in copy linked to whatever site they choose. From the publishing end we would take that link and place it behind a redirect short-url. An updated fair-use policy would be imposed stating that anyone can excerpt or aggregate original works, with attribution and link carryover. The rational is that if anyone wants to use our original material they have to ‘compensate’ the originator by also carrying their advertizing.
By using the short-url redirect we can charge for the click regardless of where the click came from. Furthermore, you also make it easier to spread said links across the social universe of like-minded people by using the short-url in the original content, which is then monitizable regardless of where the link ends up be clicked.
The goal is to allow your content-contextual advertizing to travel along the same viral path as your content, and tracking that path to ensure that resulting click revenue flows back to the content originator. Furthermore, rather than fighting the massive new distribution channels, this uses it to the publishers advantage. The more your work is quoted, the further its spread, the more your links appear, the more they are clicked, the more you make, at a marginal cost of close to nothing.
Though this does involve an extra step companies could make it easier for the bloggers and public to comply with an excerpt tool that is pre-coded with links. When faced with the choice of an extra step, or the prospect of paying for ‘online subscriptions’ I’m willing to bet that the vast majority of people will pick the links, over the wallet.
Prying Coin From Purse
Now that I have purged the web of all banners, skyscrapers, sidebars and every other traditional Web Ad, people will expect me to replace those revenues.
In their place I propose a full-on retail program tied to the product review pages. This can done in a few ways, and would be dependent on negotiations, but let’s sketch the outlines of two variations.
One is the traditional affiliate model, where an Amazon or Buy facilitates the order processing, warehousing, and fulfillment. In exchange for sending them orders, the publication gets to keep a percentage. In order for this to work, the affiliate percentage would need to be substantial, and the products, electronics to books, needs to large and wide. Here’s the hook, Subscribers get X% off across the board. The savings are enough to warrant a subscription, which because of the additional revenue one gets from growing the rate base, and the X% from gross sales, should more than supplement the lost Web Revenues from the banners.
Two, is the Fashion Model and this one is largely dependent on specialty merchandise. In this model it’s the advertisers themselves that fulfill the orders, the magazine provides the retail space and facilitates the transaction (So no discounts). The key to making this version work is the inclusion of exclusive merchandise that is only offered on your site. It can have branding, it can be designed in-house, it can be many things, but it must be exclusive and sell at a premium. Those premiums are your bulk profits. In this variation mostly what you are offering is a curating service to your customer base. You hand select the very best, and develop the rest yourself.
Both methods have advantages and disadvantages. But they both offer a new revenue stream, which ties directly into the publications core competencies and stakeholder interests.
Step Away from The Internet
Lifestyle magazines exist to bring together people with a similar set of interest. Before one page of advertizing can be sold this audience must come together… but magazines almost never really bring these people together. They may all read the same articles, but let’s face it, when big magazines throw a party it ain’t the readers they are looking to invite. No they want A & B list celebrities, some bullshit with Jerry Seinfeld, and Phyllis Diller. Or at least it’s been that way.
I think this is the A-#1 thing that needs to change and this where you get your ‘freeimum’.
Whether it’s fashion, beauty, literature, politics, interior design, sports, computers, or anything else like-minded people want to be around other like-minded people. Sponsoring events for these audiences to interact in real life would be a very valuable thing to your most loyal and interested base.
The math is obvious. Take a rate base of 750K, convert 1% into premium members, $250 a person, that’s $1.8 million revenue. Minus costs, but you also have a new ad driver. Before I delve too deeply into the cost revenue breakdown, let me explain exactly what these events entail.
What do I get for $250?
It depends on your interest and the advertiser base. The most obvious thing for business and political magazines are networking events: speakers, live interviews, happy hours, ect. Fashion and beauty are perfectly suited to shows and demo’s. Sales and sneek peeks could also be attractive. Computer and science speakers, gadget labs, gaming parties, networking events with the nerdiest of the Nerd-core for a Wired subscriber.
The idea is to have a regular series of sponsored events open to club members only. These small group of highly interested people are your best advertizing resource. They are the trend setters, they know before anyone else knows, they are your guerrilla marketing army, geographically diverse, they prime the well spring your advertisers are looking to pump.
From a business perspective you only need to charge the people marginal cost to breakeven (you can charge more), but most of the revenue comes from advertizing sponsorship of each event. From the advertiser perspective they get a captivated audience to whom you can introduce your newest lines to the public in a highly visible, yet exclusive manner.
Selling this is tricky, but potentially very profitable.
The ‘freeimum’ subscription revenue must meet your marginal costs. The reason being that even if an event goes unsponsored it still has to be held. At a minimum you need a space, a draw, and a planner. One planner can handle multiple locations, budget six people for 20 locals, $35K a year, $210K total. Event space varies widely, from a low of $5K in smaller cities to highs of $100K in NYC. Sponsored events can be held at larger nicer venues, since someone else if paying for it, and non-sponsored events can held at more low-key locations. Therefore the budget needs to premise on a baseline cost. Assume an average of $7.5K, 20 cities, 12 months, that’s $1.8MM (Just about what I project for sub revenue). Draws should typically be provided by sponsors, since that is largely the point of the events. But assuming that no sponsor comes forward a publisher could always send the editor, or a writer who has a new book, or you can invite a group of people whom are experts in some sense to hold a round table. These are very cost effective options. My point is that given the ‘freeimum’ sub revenue before you ever sell one sponsorship slot, you are close to even. So taking a low ball estimate of $10K per-event, you break even after one month of sales, and from there it’s straight profit from February forward. That’s $4MM in revenue, $2.2MM in profit.
Granted these are broad strokes, and I will be the first to admit that details matter. Yet, as a structural profit center, this has real potential. You could run the entire editorial staff at cost off web revenues just to pull the audience together for these events. Add in single event ticketing, up the sponsorship ante for select cities and seasons, and you have an $8MM revenue stream.
Which brings up the biggest question I have dealt with yet, how to sell this new ad package.
Pre-packaged bundles: Like Revenue Crack
Now that I have laid out the new digital and interactive advertising platforms its best to explain exactly how I expect to sell them; the short answer is all together.
The long answer is by combining the different elements into ad-packages. If you were to sell everything al-carte everyone would go for the exact same things, and other less popular aspects would go unsold month after month. Those low demand products are the ones with the highest profit margins.
Instead, you need to sell in bundles.
For example if you are going with the affiliate model on the retail side, you should also offer an exclusive deal to own all of the contextualized links. In this case the exclusivity allows you to negotiate a higher price-per click, or higher retail sale split.
In the fashion model, naming rights would be included in the price of a paging bundle. So in exchange for a minimum buy of X, you will also get brand placement links.
Site skinning and transitions are as two sides of the same coin, like I said my vision is that the two ad mediums would work together. So the ‘package’ here is not to sell impressions, you sell people by session length. So if Ford buys a person, they get a 1st view skin and transition. If they want a longer chain, they can buy that at a per-page view cost, which would be lower than the 1st view. (The site setup will make big difference in how well this works. 1st page views are expensive, but consecutive views are cheap once the person is on the site. The initial ad setup is the expensive part because of the fixed costs, but the extension of an ad run is almost costless. Therefore, any extension is a pure profit, which we should be pushing to the fore)
The event schedule is the prize. As such it requires a minimum buy somewhere. Whether, its paging, web ads, retail, or event locations these are your strongest leverage, because of the medium and the audience. Imagine splitting the city list into 2 levels, NYC, LA, Chicago, SF are level 1, Philly, DC, Boston, Miami, Houston, Seattle, Phoenix, Atlanta are 2. You would sell packages of 1 level one city, with two or three level 2 cities. If a client wants two level 1 cites, they also buy 2 more level two cites. It’s imperative to sell the lesser cites with the cites in greater demand, because even if no one buys the event for a location, and event must still be held to maintain engagement. (For this reason sub revenue must justify the marginal per-person cost) These events are for the most engaged, they are the ones who will justify for the client spending all the previous money, because these are the crème-dela-crème customers.
Its about leveraging high-demand products to yield higher ad buys across the board. Bundling is a proven economic strategy that has seen success in every area of media, from CD’s to Newspapers. Refashioned for a digital strategy, bundling is how I plan to bolster the sales of low-demand-high-margin inventory by tethering them to the high-demand-low-margin products.
The Forest from 50,000 Ft.
Free, for better or worse, is here to stay. The Genie has been released, and will not be put back into the lamp. While Anderson does a good job of explaining how we got here; he offers precious few thoughts on what it all means and how to adapt.
I think that’s the point Gladwell was making, and I agree with. Anderson, like many technologist writers, can point out excellent examples of tall trees in the forest, but not the shape and size of this new ecosphere. Google is doing great; it gives away email, space, results, maps, web browsers, movies, and now an OS for free. Yet, it makes billions by skimming off the algorithmically chosen premium advertisers, which subsidizes everything else. Minus that search revenue and Google is a black-hole-of-money.
This should all sound familiar, because it’s not too far off the old newspaper model. News rooms were funded by localized classified ads, with a dollop of national ad buys in the more consumer oriented style, travel, and sports sections. In that respect ‘the news’ has always been ‘free’, typically more of an ego trip for the papers owners. But now the classified are centralized on the web, benefiting from local and national exposure, and the national buy isn’t enough to run the papers at the level that they had been.
So where’s the beef? What’s the next search? How do we replace classified revenue? How do we grow the value of our digital ad space? To fight aggregation? Embrace? Can we monetize anything? These are questions that get begged, and left unanswered in Free.
These are questions I seek to answer in small part here. And my solutions on any point might be wrong and off base. But given how small most web revenue currently is for most publishing intuitions, there isn’t much risk to experimentation either. It’s like global warming, or the frog in the pot, if you jump you may escape or get burned, but if you do nothing death is assured.
….Hip – hop
