Why content sites are getting ripped off

To me, a big part of the problem is that very few consumers actually care who gets the credit for the things they buy...the content generators and the sellers can work out as many systems and deals as they like, but in the end, if the consumers don't have an advantage to helping make it work...there's little chance it will actually 'fix' the problem...

No 'shift in revenues' of a 'purchasing funnel'. Pipe dream. Content expanded much. Marketing budget not so much.

Amazon is the obvious example of a site that does both (user reviews + sales) and profits from the synergy.

Chris, this is a very well said example.

Fifth popular post reviewed, and absolutely a hit on something vital to monetary rewards- intent generation, 10/10.

We are challenged as curators, content generators, information specialists, and entrepreneurs to create a system which optimally and fairly rewards value creation. An unbiased editorial review is worth so much more to a rational reader than a shill. But how can we best construct an information trail that connects that content generator to the action of a purchase?

Advanced tracking tools that allow browsers to remember a users history across systems would enable a user to select a source that most influenced their buying decision. But even this questionable spy network requires conscious recognition of motive forces.

My latest project has some relation to this problem as it connects a users actively shared social communication to personalized ads. At least a content generator could host an ad widget like this and allow for the connection between interest and purchase action on their site. This doesn't help down the road purchase actions.

How can we socially support unmonetized editorials/reviews to enable this valuable service. Micro payments could cover the cost of an opinion independent of marketing. But somewhere along the information value curve we as users have to either pay this cost, or accept product shifted opinions.

ps: the comments to this post are incredible, marking off time to review them later.

Agreed - great discussion here, glad to see it's still going strong.

I'd like to throw a fascinating tangent into the conversation - a recent Cornell study called 'Meme-tracking and the Dynamics of the News Cycle'
http://www.cs.cornell.edu/home/kleinber/kdd09-q...

Can't help thinking there's a content attribution/monitization system somewhere in here.

I've been thinking about this problem for a while also. My startup is working on addressing this using what seems like an obvious solution. At first I was baffled why we couldn't find a vendor utilizing this "obvious" technique, but after a while we realized this is probably due to functional fixation. Vendors are using new variations on old themes to address this: banner ads for facebook apps, extending text ads to real-time search (like twitter), etc.

As I see it, the problem is broader than content sites or publishers. It can be generalized to popular figures and social authorities who cause referrals by their actions or recommendations but aren't compensated. These can be popular tweeters like Ashton Kutcher, popular bloggers like Seth Godin, or popular journalists like Leo Laporte. The solution to the problem Chris Dixon mentioned should also be general enough to cover these other cases as well.

And an excellent post by the way.

There are a few ways of harvesting intent emerging that are less reliant on old advertising models. Have a look at Skimlinks - one good and elegant example. I am also seeing a lot of entrepreneurs approaching this problem in different ways and I expect there to be a lot of startups tackling this problem in the near future.

For many digital publishers, advertising has not been the major contributor to revenue for some time. I was surprised to hear from one B2B publisher recently that they have 7 ways of making revenue from visitors to their site and advertising was the least important. http://thebln.com/2009/10/why-free-is-no-big-de...

This is a topic close to my heart - I would love to fix the problem so publishers are sufficiently rewarded for the part they play in creating intent to buy, not just through display ads but through editorial content.

I think perhaps a solution is that merchants pay a much higher bounty to editorial sites when they do lead to direct sales, or there is a first cookie check alongside a last cookie to see what publisher first sparked the customers interest in their products. Every solution will have its pros and cons, but completely agree a solution must be found, or the web will become populated purely by comparison and cashback sites, and content sites will start to wither and die. No one wants that, surely.

I for one, would love to actively explore solutions - Chris, do you know if this is being looked at by PMA or other bodies?

Fred Wilson had a good post on the defense of using both search and display advertising: http://www.avc.com/a_vc/2008/12/display-adverti...

Great post - I completely agree with distinguishing the functions of display advertising and performance (click/affiliate) advertising fill in the buying process. This is also why:
1. Video is not good for display advertising - http://www.yanivnizan.com/2009/08/why-online-vi...
2. Video is perfect for performance advertising - http://www.yanivnizan.com/2009/10/why-online-vi...

What if there was a standard way of attributing mentions that could some how be tracked and aggregated? It's done for banner ads, why not for body content mentions?

I commented on this idea a little earlier as I was catching up on Chris Dixon's popular posts.

A transparent protocol must be communally accepted to allow transparent and equitable utility of intent driven data.

Great idea.

even within Google, you are buying a mix of stage of intents but all for the same auction based price. The person searching for Nikon D700 in Google might be interested in buying a camera this afternoon or may never buy a camera yet regardless of which user it is underlying the query, the search marketer pays the same amount. We are going to try and help marketers on our site figure out where the user is in their cycle and tranche up the different phases into different prices.

This is a really great post. I'd be curious to learn if there are any 3rd-parties currently working on this problem? Publishers of all sizes would pay good money to crack the code...

Even Google is coming close to advocating getting your content scraped, indicating that sharing is not only inevitable but welcomed.

Of course not. This implies we understand the emotions of branding and how we feel about the act of labeling ourselves, as well as involvement it takes to convince someone that they are a particular kind of human being that would do XYZ.

Content is broken because when you start to actually have more and more data about likelyhood of buying, yet it still doesn't seem to equal actual buying all the time, it frustrates people about the meaning of branding on the web.

In some ways, I hate to break it to you all, branding on the web works fairly similarly to the way it did it print; You just have to deal with real time repercussions and user feedback, plus the possibility of inserting an ad that users perhaps want.

In other words, for the first time, you can do even more reflective mirrors of advertising. At some point though, you don't know how effective the ad is, because it is purely correlary. You can never fully defog the mirror that is brand and desire.

Great analogy to defogging the mirror. We are learnig about motivational psychology as much as we are working to construct an evolving monetary value sharing content stream.

It's no easy problem. But additional measures from independent sources help us track historical patterns for an individual user. Their usage pattern is what counts, not the median of all users. Statistical clumping is part of the fog.

Chris-
I have a very different view on the issue but tend to agree with Rick Burnes.
The vendors are becoming the main content generators and site advertising is dying. The content is the ad and people are growing less tolerant of anything extraneous.
The marketing folks will hopefully realize they need to start making the news and stop simply riding it. Gone are the days of the quick win.
Thanks for a great post.
Cheers,
Jon
@fogcitymedia

The issue is not so much mechanical as fundamental. Haggling over CPMs is what's left in the current framework but that's missing the point and wasing a valuable opportunity. The idea is to change the framework.

The solution is to build vertically focused platforms which are at the same time granular (many participants) and unified (far reaching). Targeted communications become possible, relevance rises across the landscape. Brands gain - they are able to pick the precise strata they wish to engage with, members gain - they are no longer bombarded with irrelevant rubbish from companies they're not interested in, the community gains - a profitable platform contributes all manner of value within the broader vertical.

The interactions can be tracked, measured and acted on if the platform owner has designed it from the ground up so as to make sure that the members' privacy is respected - gradula opt-in etc. this doesn't require any new technology. It requires a redesign of the thinking behind the platform and is a step or two beyond the currently fashionable social network.

It's a new way of doing things so what's needed is a change in thinking - ie. what's needed are more entrepreneurs. An example : as a photographer I mourn the passing of my profession as it has unfolded before our very eyes over the last ten years. As an entrepreneur, I rejoice in the fact that I can build a company which is an enabler (in the positive sense of the word :) for thousands of aspiring professionals who otherwise wouldn't get a look in. McLuhan was right about everybody being a publisher in the future (meaning now.) He should have gone further - everyone is now an entrepreneur. Actually, Drucker has said that everyone is a designer so that's not bad :)

Once the platform is built and the environment designed so as to create value for all the actors, the rules of the game change and content creators all of a sudden participate in the earning, albeit at various points in the long tail.

I'll ping you a Tweet about this - I'd love to get your take on the idea.

Yes all over to this comments points.

I'd like to add the horizontal value of social media into an opt in advertising system. Let the tools I use to share and communicate benefit both serendipitous search for me, and relevant offerings from advertisers. Check http://victusmedia.com for some examples we're building.

Large portals are able to harvest intent because their users tend to be authenticated when researching and browsing. This allows these large sites to build a profile of intent which can later be harvested.

If users are authenticated while viewing content, it would be possible to build a rich profile of user intent, interests, and preferences. This interest/intent profile could then be used to deliver highly targeted ads.

If the user's profile can be used across sites, then a very rich, cross-site profile could be collected. .

The devil is in the details. A behaviorial protocol of information would be required to be an opt in user system, and a transparent information pipeline to advertisors and intent/content generators.

True, Yahoo doing search retargeting is a good example of this.

Great post! I wonder how often it takes a complex sequence of 'events' to generate intent. Part of which is not online (friends, family, radio, tv). Tough problem.

Interestingly, Google is probably best positioned today to close the loop between intent generation and harvesting. Through AdSense, Analytics, and DoubleClick, Google tracks a large portion of the sites we visit further up the funnel from search. For the advertisers using Analytics or AdWords conversion tracking, Google also tracks purchase transactions. Presumably, they could provide advertisers with some reports that indicate which content sites ultimately drove conversions and revenue, though there are some obvious privacy issues to work out. If I knew that people who visited the NY Times where later likely to buy on my site, I would increase my ad spend on that site. Google has most of the data to make that happen.

True. Tracking this would require an almost panoramic view of the web, and Google is indeed the best positioned.

But who has reaped the most reward from the current model? Google. Do they have the incentive to take money away from their own "last click oriented" products to grow the industry at large? You could easily argue that Google's products wouldn't take a hit and they would be enabling the industry at large to add more money in "intent generation" to the pool and by helping big publishers steady the ship in the waves a little, it only helps them down the line.

True - plus they own DoubleClick and YouTube, both of which would benefit immensely from increased online upper funnel spending.

There are quite a few tools today for conversion attribution. They attempt to allocate the purchase value across your ad spend. Naturally, it is not an exact science and putting digital in a silo creates limitations. http://tiny.cc/xlEHX

Don't they all assume that a click took place at each event?

Not necessarily - you can track engagement metrics like mouse-overs on Rich Media banners and can also apply user behavioral info (via third party cookie data) to help approximate the upper funnel 'intent generation' path.

You can also attribute actions back to the ad impression itself via media stacking.

The problem is that buying inventory between CPM, TV, and Search are still in silos within agencies. When you pair them together and track the sales on top of impressions & last clicks that lead to a sale - you get metrics that are worthwhile.

Furthermore, when you use specific and identifiable ad copy in print\tv\banners etc.. then use them in SEM to track and bring accountability to those advertising methods, you further track the efficiency of those channels you could not otherwise attribute with sales in the past.

I agree this needs to get fixed, but the problem is within the execution, buying, and ultimate business intelligence of the advertising that takes place. Last years campaign metrics have a place in this years planning - but you it is hard to convince 25 years of ad agency businesses that this is true.

This is spot-on and at the heart of the problem.

That said, there are signs that things are changing. While most big agencies have silos, Havas manages them together. Similarly, folks like Matt Greitzer at Razorfish are bringing search expertise to the emerging ad exchange buying practices. There is too much money being left on the table through the siloed approach. The agencies will either evolve or clients will find new agencies.

Chris - thanks for the shout-out & another interesting post! Have been chewing on the first one for a few days now...

Smart marketers are already tracking multiple click attribution paths in order to see which upstream ad impressions are strongly correlated with downstream conversions. They know you need to feed the top of the funnel to guarantee sufficient lower funnel volume and multiclick attribution is great for spotting these 'assists'. That said though - I can't see many marketers sharing this data with the publisher just so they can pay higher CPMs for these ads.

"That said though - I can't see many marketers sharing this data with the publisher just so they can pay higher CPMs for these ads."

I think you've nailed a key point in this discussion. The buy side has access to a significant amount more data about the halo effect of various media types. But there is no short term incentive for them to work with publishers when last click metrics allow them to be aggressive in negotiations and hold their hands up when asked about the holistic impact of a campaign. I've been on both the buy and sell side of the business and until both sides have more common ground, the industry will have a difficult time solving for this.

Chris, you are on a roll with your posts...keep 'em coming.

You are exactly correct with this observation. The track-ability of each and every interaction with a page or ad unit has rewarded marketers with an "easy" way to measure and show results, and diminished the value of the other effects of a digital campaign.

The other effects of a digital campaign, that ones that can't be measured have value. Some people (myself included) think that there is more value in what hasn't been measured. Without an "easy" way to calculate that value, marketers (who control the digital media spending) will continue to spend accordingly.

That is the world that we live in...right or wrong...

In some ways, this is analogous to a very old "inequity" in business. It's not at all uncommon for the people in R&D to make middling salaries as they develop or invent the products sold by a given business. But who makes the big money inside of those organizations? Sales guys.

Why? Sales guys don't make anything (they don't "generate intent") and they are generally have no say in the product decisions that ultimately lead to profitability. But because they go out there and actually capture whatever willingness to buy is around a given product or set of products ("harvest intent") they are compensated substantially.

Ultimately, a similar dynamic is at play here - content sites may give me the data, expert commentary or backup I need to make a purchase, but the aggregator who puts me into the sales funnel will capture more of the economic bounty associated with my purchase.

This is an amazingly prescient comment and a look into how society operates and what led us into the current downturn. Unfortunately, the socio-economic DNA in the U.S. revolves around sales vs. value add productivity. If everyone focuses on salesmanship, you are ultimately creating a giant ponzi-scheme environment made up of snake oil salesmen. Thank god there are content creators that are willing to innovate even though they are underpaid. Hopefully they will be redeemed soon.

R&D guys not getting paid enough -> I guess that's why startups exist.

Hells yeah Chris.

Amen.

I think Microsoft was working on a project somewhat similar to this. Technology that used behavioral tracking to justify display ads that were never clicked but were seen and played a key role in building intent to buy before the user finally went to Google, typed "Nikon X40 camera" and bought the thing through AdWords.

It seems like, privacy implications notwithstanding, the same technology (that may or may not have ever been developed, I haven't heard much since I read that article a year or two ago) could be used to track content's role in building purchase intent.

Great post.

Great tie in Aaron, appreciate the opinion/question.

This isn't a MS invention - it's an existing practice used by some of the more sophisticated digital marketers. Look into 'multiple click attribution' or 'partial attribution' for more info.

Thanks Marshall for the heads up. Learning at top speed depends on valuable shares/links like this. Much appreciated.

Thanks for the tip.

But there's a bigger change taking place: Content sites are doing less intent generation. (What kind of intent can plausibly be generated when I'm reading about Healthcare on NYTimes.com.)

Instead of buying difficult-to-deliver intent from media companies, the vendors themselves are doing the intent generation. See here: http://ow.ly/rBhl http://ow.ly/rBiw http://ow.ly/rBjk

This leaves media companies w/o much of a business model, but that's the world we live in.

I don't know if I agree. To take your Whole Foods example, a lot of people learn about the benefits of organic food from blogs, nytimes, magazines etc.

Interesting. I assumed that when you were talking about intent generation on media sites, you were saying that the display/cpm ads are doing the intent generation.

In fact, you're saying that it's the content.

In that case, I think you're exactly right. Media sites have no good way to harvest the intent they generate through their content. I don't think media co's will ever find a robust solution to this problem. I think the only folks who can harvest the type of intent generation you're talking about are the vendors themselves -- which is why it increasingly makes sense for the vendors to publish the content.

well, i was talking about both the content and the ads. i agree there is probably no solution to the problem, especially the content side.

I think there is a solution to the problem; as the core data in both content and ads can be "structured" to be made portable, sharable and branded by the originator for monetization. For example, clicking on the save button on any recipe on Unilever's Knorr site (http://www.letsmakeknorr.com/Recipes/8240/1/cru...) actually "gives" the recipe to the user - all the while keeping the brand and links back to Knorr for related content present.

(full disclosure, I'm a springpad co-founder that's powering the capability)

Interesting connection Jeff. Controlling access at the content generation end then is your solution, but not necessarily by direct monetization, but by attribution or an API.

Hi Mark; it's not so much controlling access but rather attributing the orginator as the source and subseuently sharing monetization revenue generated [off-site]

I'd much prefer attribution, over content control. I just don't see how this works in practice without the generators at least monitoring access. How can a store owner get paid unless the users somehow pay for their product.

Good stuff!

Theoretically speaking, do you think it's possible for a site to generate and harvest intent? Or can it only be accomplished via a partnership? (Perfect example to illustrate your point, Sugar (successful fashion/style content but making little RPM) bought ShopStyle (fashion search which didn't have users, but had a RPA model). It was a brilliant move.

Yes, I think that's a common move. Its why most of shopping.com's revenues are via partners using their api.

Can a site do both though? Can you generate intent and capitalize on it? Or is it too much? You would think that if anyone would attempt to tackle this issue, it would be Google (connecting some of their content entities more with their search). Or Amazon even. Is the only way to marry intent generation with intent harvesting by partnering? (I'm thinking if you are very targeted... perhaps the answer is yes. Perhaps.) You can license your search, or you can have APIs to access your content, but it might be too hard to do both. I can't think of anyone who is...

The key problem that we have to today is that the Internet made content worthless. What Chris proposed sounds logical except in the end it strengthens the camp that is making money today. And that camp is in business or scraping and black SEO. In other words this is the anti content camp.

Essentially all Internet problems flow from that single reality of worthless content, when aggregation is more profitable (naturally) compared to generation. Solve this fundamental problem and the rest is commentary. And yes content should not be valued by the clicks.