Dan Heller's Photography Business Blog Industry analysis from www.danheller.com

The photography world -- the business, the culture, the art, the politics, the technology.

Site Feed

Subscribe to
Posts [Atom]

My Photo
Location: Santa Cruz, California, United States
My Books on the
Photography Business

Sunday, February 18, 2007

Two-Phased Approach to photo-sharing/licensing model

Since I posted my last blog entry on where I think the photo-sharing and photo agency markets are going (and that they will inevitably converge), I had more feedback from non-photographers than any of my other postings. Many emails came from computer programmers who are also building AJAX applications (the fundamental building blocks for how Flickr works), and are wondering whether they have a viable business model (not being photography). I had a few emails from people in the investment community (some of whom I've had contact with before from other postings on the subject), but the most surprising emails were those from start-up companies who are also building photo-sharing sites with their eyes on flickr. No less than five "viable" (IMHO) start-ups are doing much the same sort of thing (well, "the same, but different") with hopes of being the first photo-sharing social-network site to double as a licensing agency.

Oddly, the one question that everyone had in common was this one, (paraphrased)

"Will people accept having their photos online to sell?"

The context, of course, is that social-networking sites are not traditionally places where people look to sell their photos. Still, I think people's attitudes about this are similar to that of placing online ads on their own websites. They will not only accept it, but most will embrace it because there's money to be made. If what is demanded of them is very simple and straightforward, I believe they will participate. And as smart people figure out how the optimize their returns, a new economy will evolve much like how online advertising was transformed by Google.

But, this won't come right away; it must evolve incrementally, because there are too many unknowns that have to play out, such as pricing models, liability concerns, and what's the middle ground that will appeal to both buyers and sellers of imagery.

I believe this will come about in much the same kind of two-pronged approach that Google followed, at least in the following summary: Google started rising at a time that didn't look so good for online advertising: the tech bubble had burst, online rates had long peaked at ridiculous levels without a track record to justify them, and consumers were sick and tired of pop-up ads, pop-under ads, and otherwise invasive and unrelated flotsam of ads from the sites they visited. It didn't help that many ads were also the source for spyware and such. And here comes google with their innocuous text-only ads that sit quietly on the side of the page.

At first, their model was to place ads on the search results pages. Then, as their business grew, they exploited the model where other people could put ads on their pages. (The "Adsense" model that is now replicated by everyone in the online ad business.)

I believe that photos have similar opportunities (not necessarily to the same scale of ad revenue generation). Photo licensing is currently done using a traditional model where buyers go to online agencies or directly to photographers. However, as per my previous posting, I believe that it is inevitable for other photo-related sites to get into the licensing business, which will change the fundamental landscape for how licensing is done. Because I have always believed that the photo industry has vastly underestimated how much consumers really participate in both the photo buying and selling economy, the migration of consumer-oriented photo sites toward a licensing model is a perfect fit. (My first $100K of sales was generated in photos before I ever really took photography seriously as a business. And that's what prompted my early articles on the subject and business proposals to various photo companies.)

Phase One of this business will be where a photo-sharing site merely allows visitors to license images directly from the site. Phase Two will involve the distribution of the same photo assets to other sites, much the same way online ad sales are hosted (or "published") on other websites. When and how the second phase happens for photos depend on how these early adopter sites implement this model. As shutterfly's experiment demonstrated with their "event-photographers-only" approach, it's easy to get wrong. For the sake of discussion, I'm going to assume that the approach ultimately adopted is the one I've suggested in the past: make it pure and simple by giving the user a toggle for setting whether his photos are (or aren't) permitted to be "sold" under the general terms and conditions provided by the site. The pricing should be set by the user for a small set of tiers, with a "default set" to be used in the event that the user wants to defer this decision. Obviously, I'm hand-waving a little here, but I'll pick up on it again later.

If provided with a very simple and easy way to opt in or out, the larger majority will opt in, almost for the sole reason that people put online ads on their websites: money. Instantly, the photo-sharing site has morphed. Flickr would be to the licensing business that Google is to the advertising business. Sure, Google looks like a search engine, but they make money through ad sales. Similarly, Flickr looks like a photo-sharing site, but they (could) make their real money through licensing.

Why risk it? As my previous article said, it's inevitable, largely because it involves very little technical expertise or investment, and the financial returns are too interesting to not bother trying. So, what is that "investment?" Oddly, not much. To get a sense of this, one of the start-up companies that emailed me is a two-person team, only one of which is a programmer (whose 18 and never had a real job). You'd never know it by their site, as it looks and acts surprisingly professional, looking as though it were managed by a sizable team. Of course, great credit goes to the duo, but the real underlying "magic" is simply the ever-growing development--and use--of AJAX, the basic underpinnings of what is now more broadly known as "Web 2.0" technologies. (Flickr is what is it because they were early adopters of AJAX.) In other words, creating the technology behind what would become a social-networking photo-sharing photo licensing site is not hard. This means that the technical "barrier to entry" is not technology. It's really about three other things. First, one has to have the vision that this is the future. Second, the ability to attract a critical mass of users and create a living, breathing social network that people talk about. And third, to know and understand the economics of selling a commodity whose value in the individual units, not as a generic "class" of widget. Hint: an auction-based system where buyer and product are combined to establish viable price points.

For example, the current hot-commodity in photo sales are microstock and other agencies that are dumping images for pennies on the dollar. I place no fault in their doing it, as I see their entire businesses as a predictable byproduct of the broader industry not having found any other way for a huge population of people to monetize their photo assets. I consider microstock agencies to be temporary anomalies because they are taking advantage of one particularly important aspect of today's market conditions: their sales do not depend on a good, strong, established base of images. They just dump images. One could argue that "it helps to have better content than the competition," but I will only concede that as a minor incidental. It should be well-known by now that photography is now a commodity with infinite supply, and a growth rate that exceeds the demand. The ability to acquire good, quality images isn't hard. What is hard is mining that drudge and presenting a package that's interesting to buyers, and microstock agencies don't do that. In fact, no one does.

And that's where Phase Two of this business model begins. To address this, let's return to the aforementioned Web 2.0 tools, and the online advertising auction model.

Online advertising depends on an underlying ranking algorithm to determine which sites best meet the criteria of a search pattern. To also place ads on that site, it also depends on the ability to rank the "value" of certain keywords. Ultimately, the system works when the system feeds results to the searcher that best meets his goals. And the ads the searcher sees when he gets to that page should be relevant enough to inspire him to click on it. It took a while to get to where we are now, but it happened because there was a business model that needed it.

During Phase One of photo-sharing sites that make images available for licensing, the same patterns will emerge about who's coming, what they look for, what they ultimately buy, and what the site can do to better meet their searching needs for the purpose of building the same kind of ranking engine. Stock photo sites think they have that information, but their visitors are a much smaller subset of the broader (and more highly trafficked) visitors to social networking sites. So, although the data from agencies is interesting, data from photo-sharing sites is more important for the broader auction based learning system. The objective is to build intelligence about how to rank images in the context of searches and buying, and what kinds of keywords (or other photo attributes) command higher prices.

Search technology has matured to a point where it's easy to find the good stuff first, despite everyone's attempt at getting their website higher in the results page. This isn't true with photos, but if online licensing ever takes hold on social-networking sites (or even general search engines), then you can be sure that such a ranking system will evolve. The mere presence of money demands it.
And whoever does it best will transform from a web-sharing site, to a hub of a world-wide photo distribution system.

So, let's examine what this would look like.

It's already possible (using AJAX and tools made available by flickr) to host flickr-based photos on your personal website. It's as easy as, say, creating a google adsense account, or a yahoo publisher account and hosting ads on your site. The next obvious step for flickr and others is to use exactly the same existing tools that users already use to put photos on their sites, to also display "buy" icon next to those photos. This icon would go back to the hosting site where the buyer would have the chance to license the image for a given use. The price would be based on the results of the auction system, which will have been established by this point.

The attraction for doing this is actually more compelling than advertising, albeit for different reasons (and with different revenue structures). Unlike an "advertisement," which is not "content," photography is. What's more, it's content with value. If a newspaper wants to run a story on a fire downtown, the story isn't better because it has ads for fire extinguishers. The story is better because there are photos of the buildings being burned. If the paper can do a quick search for photos of that building (which is very likely when there is an infinite supply of images), there's a quick, immediate sale. What's the price? Depends on how the auction plays out. In ads, the auction goes one way. In photo licensing, the auction goes two ways. The photographer has to accept the bid. The more unique the photo is, and the more difficult it is to obtain, the auction system will rate it higher. If not, the auction goes to the next photo in line. The supply and demand kicks in, along with the buyer's perception of need for higher or lower quality photos. The real economics of photo pricing will kick in, and people will have far more incentive to participate in that model, than in lesser-performing ones.

For anyone that would want to monetize their images, this model enables them to have their images splashed on as many places as possible. Hence, photo searchers would be able to find and source images from more sites around the net. The underlying middleman is less important, much like how users who click on ads and end up on websites to buy a product without even caring that Google hosted the ad.

It's even possible that some people would make websites that do nothing more than present images FOR SALE. Suddenly, anyone can be a photo licensing agency.

The publisher (licensee) of the image has even more incentive to license it than just to enhance his website's content: he also becomes a publisher with potential income. Just as a website shares in revenue when an ad is clicked, the same publisher (website owner) should share in revenue when a photo he displays is bought or licensed by someone else on his site. (Even more incentive not to steal.)

Users from either inside the photo-sharing site or outside interact with those photos, and their input is aggregated into the same user base, establishing one of many variables for what will be used for pricing purposes (auction).

Pros would naturally be concerned because they may fear a dilution effect of their images, or feel their prices would drop. If the model is adapted well to the same ranking/auctioning type system used to sell advertising, then it would follow that higher quality and more desirable images would sell better. Rather than having the arbitrary and fickle decisions of photo editors choose your photos or present them to buyers, or whether to rank them highly on any given photo agency site, images would be subject to ranking through a more stabilized system weighted by the results millions of people participating in the system.

One more benefit to this model: manging the misappropriation of images online. As all photographers know, images are illegally lifted from photo sites all the time. Yet, this business model would have a net positive effect on image stealing: first, it would provide potential copyright violators with more visual cues and tools to properly license images. "Innocent" stealing is often because there is no obvious or automated mechanism by which one can "buy" the image. Another deterrent to stealing is that the facilities for tracking down violators would be implemented by these sharing sites, because they want to protect their assets and potential income. Currently, photo sharing sites don't have a business interest in tracking down violators, but if their revenue depended on it, that policy would change quickly. So, not only would there be more streamlined and automated tracking, the entities who would pursue violators would be far more effective. (To the individual photographers reading this, how many of you have succeeded in sending a mean email to an infringer and, as a result, gotten payment?)

Now, if you've been focused solely on the online world, remember, most images are licensed for print, or other uses that don't involve the web. Not only is there more opportunity there, but I return to the question of image stealing: there is already a fast-growing sub-industry of photo-tracking services, seeking to jump into this game right now. "Ideeinc" and "photo scout" being two such contenders. Photographers currently have very little ability to monitor the potential use of stolen images in print form. In this new model, more players would be involved in this, which would also bring the price down for doing it.

So, image licensing would fall into more compliance that currently exists today, which not only helps in price stability, but it adds a huge amount of buyers into a system where they would otherwise not exist.

Lastly, final comments on pricing: because this approach is phased, it's impossible to know how pricing would end up. One has to have inherent faith in a model (proven by google's effect on online advertising) that stabilization occurs when market forces are allowed to perform naturally. The auctioning system in online advertising is a model to follow, and I would certainly expect eyes to be looking in that direction. It took the online ad auction system a long time to get to this point, and I would expect nothing less here. Best yet, it is a much more efficient system than the $1/image microstock model.

Initially, the system will have to begin open-ended to serve at least two masters: established and experienced photographers need to set their own prices for what they think the market will bear for their work, both because they have good experience with this, but also to get them engaged in the process. Secondly, to meet the needs of the naive consumer, hoping for whatever comes in, reasonable default prices will have to be established. These seem like competing objectives, but the fact that the photo site itself makes money through a percentage of sales as well, everyone has incentive to optimize return, while also being aware that competition pushes prices the other way.

As the system evolves, photographers who are "ranked highly" would naturally command higher prices, whereas my 2.5 year old son wouldn't really sell that many of his photos. (Unless a museum took it as a work of contemporary modern art.)

Obviously, competitive forces will affect prices across the board. but this is expected. Using my own photo business as a reference, I would expect to make up for any per-image price drops by the much broader distribution of my images. After all, I face the same competition today from microstock photo sites who charge $1 per image, yet I continue to sell pictures for hundreds (sometimes thousands) of dollars each. Why? Partially because my site and my images "rank highly" on websites and in buyers minds. (Indeed, a future blog posting will show some stats on how my latest price revisions have yielded a net increase due to pricing and buying behavioral shifts in the industry.) But, also in part because I'm a sophisticated seller--I know what I'm doing. But, I don't want to oversell myself. It doesn't require a high IQ, just some common business sense (that I discuss in my books). There's no reason why a photo-sharing site can't recruit similar know-how.

Clearly, intelligent pricing has to be involved here, and my experiences prove that the laws of economics for commodities are still sound, and "anyone can do it." If an intelligent pricing engine were introduced into the automated auctioning system, price erosion doesn't have to result. If broadly applied--just like in the ad industry--there should be a tremendously stabilizing effect on prices. As more and more photographers, buyers and publishers participate, the distribution of revenues should go to the content providers that actually have good content to provide.

In summary, any company that wishes to get into the broad-based image licensing model will have to do it incrementally, at least in the two phases I outlined here. You can't jump in with both feet, but you also can't sit back and wait. But you have to start now. Based on the calls I've been receiving, there are many people who feel the same way.