Brian O’Kelley of AppNexus wrote an interesting article last week in ClickZ affirming the ad networks’ rightful place on the agency media plan. He encouraged ad agencies to worry less about about the ad networks’ business — i.e., those tantalizing 70% margins reported in SEC filings — and more about their client’s business. After all, if you are able to do a better job for your client, who cares if it causes someone else to make more profit? This is good rational advice, but its not the way humans ordinarily think.

Take the Ultimatum Game, a psychology experiment that examines decision-making processes. Here’s a quick summary that I found on Ars Technica:

The basic rules of the Ultimatum Game are simple. One person is given a stack of cash, and told to divide it between themselves and a second party. That second party is then given the chance to accept or reject the offer; if it’s rejected, neither of them get any money. Clearly, any of this free money should be better than nothing, so under assumptions of strictly rational behavior, you might expect all offers to be accepted.

They’re not. Things in the neighborhood of a 50/50 split are accepted, but as the proportions shift to where the person issuing the ultimatum tries to keep seventy percent of the total, rejections increase. By the time they hit an 80/20 split, nearly 70 percent of the offers are rejected, even though that 20 percent of the total cash would leave the recipient better off than where they started.

Some agencies fall into the same trap, leaving 20 cents of client value on the table because they can’t stomach another supplier taking 80 cents.

If you’re focused on buying marketing inputs (e.g. impressions) as cheaply as possible and you demonstrate value by referencing the savings you’ve generated, then you’re a procurement officer and you create value by hammering away on your suppliers’ margins. If you’re focused on delivering marketing outcomes (e.g. consumer engagements, a few points in brand lift, sales) by properly pricing marketing inputs, then you’re a strategic partner and you create value by building your clients business. I know this seems a bit polarizing, but I do believe these are distinct mindsets and people are usually operating in one or the other.

We’re at a transformative moment in the media business. The advertising value chain is playing a game of musical chairs and, for better or worse, you’re going to be sitting in a different chair a few years from now. The question is: what chair do you want to land in? The mindset you take today will decide your role in the future.

I’m attending the Right Media Open in Chicago and, no surprise, change is in the air. Although there is a general consensus on where the industry is headed, I am seeing a healthy debate around the timeline for that change.

While discussing the importance of indirect, bid-based sales to publishers, Dave Zinnman from Yahoo pumped on the brakes, saying that if you believe exchange-based inventory will become dominant in a “career-relevant timeframe”, you need to “step back from the punch bowl.” For me, “career-relevant timeframe” is the most important phrase I’ve heard today.

No matter what your business, its important to have a realistic understanding of how fast your market is changing. Just today, VMM founder Darren Herman retweeted his 2008 post comparing the rate of innovation with the rate of adoption, and reminding entrepreneurs to build for today’s market. That’s the relevant timeframe for a venture backed startup between rounds.

Here in Chicago, the question of the day is: what is the relevant timeframe for advertising-related companies evaluating the momentous shift toward automation?

Up until now, I think media decisionmakers have been very confident in their ability to influence the rate and direction of change. At the 2009 24/7 Real Media Summit, I was struck by GroupM CEO Irwin Gotlieb’s remark that he felt it was, in some part, his responsibility to manage change in this new media landscape on behalf of various stakeholders. Consolidated media buying firms exist for the sake of exerting this type of influence and the comment made me think a lot about how and when the industry would change.

Now that Google has turned its focus to display, they will radically shorten the relevant timeframe for considering change. Google has more clout than any single company and they have built their business on automation and bid-based buying. Whether or not you believe that audience is more important than content in valuing an impression, it is impossible to deny that a lot of client money is lost to the transaction costs of directly buying standardized display inventory. I began my career as a media planner and, like every planner, I spent plenty of time “guesstimating” and doing menial tasks in the current advertising operating system: phone, fax, Excel, email.

Its not necessary for agencies and publishers to surrender to Google — in fact, I think that’s a terrible idea. But it is necessary for agencies and publishers to recognize the economic imperatives that drive Google’s strategy. Without Google, media companies might have been able to defy gravity for a career-relevant timeframe, but that’s no longer the case.

[Thanks to AdExchanger for publishing the original of this post, which I'm now reblogging. There was a lively comment thread over on the original post.]

My work focuses on the economics of advertising, but recently I’ve been thinking about the political economy of advertising. After all, advertising dollars don’t have a mind of their own. They need industry professionals to push them around from one company to another. Trusted personal relationships have historically been the conduits through which ad dollars flow.

This relationship driven world of advertising is now being replaced by the data driven world of advertising. RTB, DSP, SSP….these acronyms have got sales leaders thinking. How do you staff up for this alphabet soup of new business models? Who do you call on and what do you tell them? What does the advertiser need and how do you win their business as a publisher?

When I was a media planner, I had the answer to the last question. I could tell you what the advertiser needed and I would decide whether or not you met that need. This is the type of arbitrary power that brings a recent college graduate towering seafood platters, custom sneakers, and a taste for fine scotch.

Working on the new demand side, I still meet with major publishers and tell them what my clients need. The needs themselves don’t change that much, in fact. Brands still want effectively priced advertising units in safe environments that will get consumers to engage with and buy their products. The key shift is that I no longer decide whether any specific publisher or web page fits that need. The algorithm is the new decision maker.

The algorithm is a better decision maker. It bids rationally by fully incorporating learnings from past performance. It can value tens of thousands of individual impressions per second based on multiple data points.

Algorithms also remove the physical constraints that limits agencies to a small list of publisher partners.  An actual media planner using phone, fax and email can evaluate proposals from a couple dozen properties at most. From a time management standpoint, it is inefficient to actually go through with buying and optimizing more than a dozen sites/networks. But with algorithmic buying, the agency can buy and optimize in real time across thousands of sites.

Advertising is the art of persuasion but personal persuasion has now been taken out of the media buying process. Is it fair to make publishers compete on raw performance and not give them any appeals process when they lose? I would argue yes. It’s certainly a better deal for the brand whose marketing dollar is now working harder. I’d also argue that it makes things more fair for publishers, since they are no longer competing on the basis of access to decision makers. The algorithm is the decisionmaker, it will evaluate all publishers in the secondary channel, and its unbiased.

The traditional sales conversation — scheduling a conversation, determining if the product is a fit, then negotiating price — still happens but it occurs between agencies and technology vendors, not between agencies and publishers.

Now, the agency-publisher conversation is less of a sales conversation, and more of a collaborative problem solving conversation. Both agency and publisher are solving for the same thing: getting as much inventory as possible in front of the true decision maker, the buyside’s bidding algorithm.  Below is the complex equation – the now infamous ecosystem slide:

Ideally, this chart would be much simpler. There would be one big pool inventory that everyone plugged into, and bidding optimization would be entirely automated. This is not the case, unfortunately. Between the brand and the publisher, there are lots of different TradingDesk+DSP+Exchange+PubOptimizer+Publisher permutations, some of which may lead buy-side actors and sell-side actors to be disconnected. Coarse-grained optimization, like eliminating entire contextual channels or entire exchanges, also removes individual publishers from consideration.  So the conversation becomes about managing the supply chain to minimize these disconnects.

To use the old media buying paradigm as a metaphor, its as though agencies and publishers are administrative assistants, working together on logistics so that the the publisher’s inventory can get in front of the ultimate decision maker, the algorithm. That doesn’t sound glamorous, but getting the supply chain right offers much greater rewards than even the biggest direct deal.

“Brand is magic… there is no computer that can figure out magic,” according to Jim Heckman, CEO and founder of 5to1.com, explaining why advertising and marketing will always require a human element — meaning the good services of media planners and buyers.”

“Digital Hollywood: Media Agencies Are Here to Stay — Ad Networks, Not So Much.” MediaPost. May 6, 2010

“Any sufficiently advanced technology is indistinguishable from magic.”

Arthur C. Clarke, Profiles Of the Future, 1961.

Well, do you? Which type of marketing magician do you prefer?

[Thanks to @adexchanger for publishing the original of this post, which I'm now reblogging.]

As our industry continues to rationalize the way brands buy advertising, we’ve seen plenty of new companies and products pop up. Some provide solutions to old advertising problems, like universal frequency capping. Others deal with fresh challenges, like how to handle tens of thousands of real time bidding requests per second.

Despite the rapid pace of innovation, I think its possible to identify 10 larger trends that will continue to operate for years. Taken together they represent not just a bunch of complementary technologies and organizational challenges, but rather a new school of thought — a new way to to think about, plan, and execute marketing campaigns.

When it comes to providing relevance, advertisers are handicapped by industry privacy regulations — our codified, monolithic interpretation of consumer desires. For non-advertisers like Twitter and Facebook, consumer privacy is considered more from a product design standpoint, rather than a regulatory standpoint. In this blog post, I’d like to look at ways in which non-advertisers make consumers comfortable sharing the personal data that drives relevance in web experiences.

1) Offering Transparency — When someone I log into a new web service through Twitter and all my Twitter data is populated, the data flow is pretty intuitive. Compare that to the data flows for online advertising. As a consumer, its impossible to understand where the data enters our labyrinth of redirects and market mechanisms and where it exits to target an actual ad.

2) Offering Choice & Control —  It’s my choice whether I want to have my data ported into a specific new service. I, Greg Hills, tend to consent. There is of course a tradeoff between anonymity and relevance. But within specific use cases, like signing up for Plancast.com, the cost to my privacy is clearly defined and the benefit of volunteering information is immediately apparent.

Advertisers fail to convince consumers of the benefits of relevance since they argue in the abstract rather than in the context of specific use cases. Urging someone to sacrifice privacy for the sake of Relevance is like urging someone to embrace hedge funds for the sake of efficiency in the capital markets. The benefit is too abstract to seem compelling.

Also, Privacy is considered practically sancrosanct when presented in the abstract. Going against Privacy in general is like going against Motherhood in general. It’s a difficult position to argue.

But when you present the trade-off in the context of a circumscribed web experience, people are willing to sacrifice anonymity for relevance. They’ll put in their ZIP code to get the weather. They’ll volunteer information about the high school they attended so that long lost friends can get back in touch.

When the benefit is made clear, and the privacy cost is limited to a specific web domain or service, consumers will consistently opt for relevance. The challenge for advertisers is to make the cost and benefits clear so that consumers can make informed choices.

3) Creating Trust — Consumers trust what they know. Social networks and content providers are consumer-facing brands, which is a big advantage. The plethora of companies in the ad tech landscape are barely recognizable to people in the industry, forget about consumers. Why would you trust someone you’ve never heard of?

———————————————————–

So, what lessons can we take as advertisers?

1) We need to create a trusted, recognizable brand entity that represents data-hungry advertisers. I think we’re making great progress with the development of universal behavioral targeting icons.

2a) Users are more likely to sacrifice anonymity when presented with clear benefits. Simplifying data flows would make the whole process less scary and allow consumers to associate privacy costs with corresponding relevancy benefits. For this reason, I’m very bullish on log-in’s becoming the building block for the data landscape of the future.

2b) We’re seeing an infusion of venture capital into data exchanges and it seems like a foregone conclusion that data exchanges will play a foundational role in the ad ecosystem. Let’s think hard about the fact the publisher’s interest in not having the value of their audience diluted through data exchanges is diametrically opposed to the consumer’s interest in knowing when they are being observed by advertisers on publisher sites, as well as how and when that data is later being used to target ads. Is this sustainable?

4) A culture of distributed opt-in leads to more innovation than a culture of centralized opt-out.

5) Consumer preferences vary greatly but privacy guidelines tend to set the baseline at the level of the most conservative consumer. A culture of opt-in where individuals express preferences for specific experiences empowers consumers to define their own web experience. They can then make their own decisions regarding the anonymity vs. relevance trade off. Giving consumers power through opt-in culture creates freer flowing data than an opt-out culture.

Relevance is the key to winning on the internet. As advertisers, we’re always striving for relevance, but we’re not really competing with each other. Most often we’re competing with the adjacent content. And average click through rates of around 0.05% makes me think we’re not winning the game.

As I move around the internet, I’m seeing amazing increases in relevance. I can sign up for a new web service like Plancast via my Twitter account and OAuth, and have my first site experience informed by information about me and my social graph. I can read e-mails in GMail and have Rapportive pull in data from Rapleaf showing my correspondents’ Flickr account, Twitter accounts, and LinkedIn account. When I visit LinkedIn, they apply an uncanny intelligence to my social graph, suggesting “People You May Know” that I either do know or would truly like to meet.

Across the web, new levels of relevance are driven by data that is:

1) Persistant — The data doesn’t get erased, so it continues to get smarter

2) Portable — The data goes everywhere I go

3) Personally Identifiable — The website is speaking to me, Greg Hills, not some approximation of who I am

The data driving these new levels of relevance is organized around user names, email addresses, and other persistent, portable, personal identifiers. It is transferred largely through API’s.

Online advertising data is stuck in the ghetto of cookie based storage. How could a cookie-targeted ad compete with content that’s informed by open API’s? Its like bringing a knife to a gun fight — and their guns are only getting bigger! With the accumulation of time, user data is becoming richer. With the growing pervasiveness of API’s like OAuth and Facebook Connect, the data is becoming more ubiquitous at the same time. Advertising relevance is growing incrementally while the relevance of the content I consume is experiencing hockey stick growth.

When will we move past the cookie?

Last week I attended a NextNY panel called “How to Make Advertising Not Suck” in a great space provided by Mike Dudda of Deutsch.

If you attend industry events and haven’t been to a NextNY meetup, I certainly recommend that you give it a try. You might need to buy your own drink, but it’s a great crowd and the community spirit along with the open conversational format make for a great experience.

I’ve listed the problems in descending order of importance to last weeks’ discussion.  Relevance and ROI, two deeply intertwined concerns, dominated the evening. Relevance is defined as the proper answer to the age-old advertising strategy question: “Who, What, When?” In the past 12 months, we as an industry have made tremendous strides in providing better answers to this question. Tech companies have built great new products, agencies have reworked their organizational structures to adopt these new technologies, and clients have literally “bought in.”

We still have a ways to go, however.

Event moderator and avid kayaker Charlie O’Donnell mentioned how recently he noticed a wetsuit sale on  fashion retail site Gilt Groupe. When he tried to buy the suit, it was sold out. Charlie has probably seen hundreds of display ads since his failed purchase attempt and not one of them had anything to do with wet suits.

We’re going to achieve amazing levels of relevance that will benefit consumers and companies throughout the value chain .  But should we, as digital marketers, be so bold as to aspire to achieve more than entirely unprecedented relevance in advertising?  Yes.

Brands demand that we go beyond relevance. Online advertising doesn’t suck in general, but it does receive disproportionately small budgets relative to other mediums such as television.  One reason is that digital is very good at harvesting purchase intent for products sold online, but other mediums, like TV, are better suited for purchase intent generation.  Online is a great complement to other mediums.  Online advertising is good where TV fails (relevance, for example) and struggles where TV excels (emotional connection and narrative, for example).

Since online ads exist in a non-linear, non-interruptive, consumer-controlled medium, it is difficult for them to pass what I call the “Nick Drake test,” in honor of my favorite TV ad of all time, below. You can rename the test after the ad which speaks to you personally with the deepest level of meaning.

This ad is not harvesting purchase intent.  I wouldn’t even say that it directly generates purchase intent.  Instead, it accesses involuntary memory, aka Proustian memory, to create an emotional experience that I still remember 10 years after first seeing this ad as a teenager.  This ad certainly does not suck for the consumer, brand, or publisher.

To be sure, digital advertising, specifically conversational media, is entirely capable of creating emotional connections. To take a personal example, I was impressed when I tweeted about wanting a Kindle and Amazon replied with extremely relevant sales information.  But I was truly amazed when a human responded thoughtfully to my sarcastic reply to what I assumed was a Twitter-bot.  This advertising interaction delivered both relevance and a personal connection.  Once TV ads are digitally served, this fusion of digital relevance and analog emotion will become even more scalable.

Many companies and individuals are now focused on numbers 1-3, and the winners will win big. But that is just one battle. The ability to pass the “the Nick Drake test” at scale will decide who wins the war.

I had a thought-provoking client conversation earlier this week. As the client increased spend for their display campaign with my company, we mentioned that the eCPM for ad exchange inventory would increase slightly. They asked an interesting question: “Shouldn’t it be cheaper if I buy more?”

We’re all used to getting volume discounts. When I supersize my meal at McDonald’s, buy toilet paper at Costco, or buy two bars of soap to get one free at CVS, I’m decreasing my unit cost by buying more. Retailers include volume discounts in their pricing structures when they can increase profits by selling more units, despite the lower unit price.

Volume discounts are prevalent in media buying. Publishers often lower the CPM, or unit price, below the “rate card” for big agencies that spend a lot. Big agencies turbocharge this volume discount. They leverage the collective spend of their entire client base in upfront publisher negotiations, achieving lower rates which are then passed on to the individual clients. Big brands, in turn, leverage their own scale in fee negotiations with agencies. Everywhere you go in the traditional media world, bigger spend equals cheaper rates. So it seems totally backwards to pay higher CPMs when buying more media. What kind of incentive is that after all?

Here’s the important point revealed by my client conversation: A huge, underrecognized shift occurs when advertisers and agencies begin buying media from markets, such as ad exchanges, as opposed to media retailers, i.e. publishers. In a market, prices are determined by the intersection of supply and demand and, all else being equal, price goes up as you buy more. Those familiar volume discounts go away.

These pricing dynamics are familiar to search buyers, since search prices are determined by market mechanisms. Google designs the market for selling their search listings, but they don’t set the CPC price themselves. Nonetheless for anyone used to buying media from publishers instead of markets, paying a higher rate when buying more still feels kind of weird.

I’d like to make two high level observations on how media markets change the agency business:

1) The shift to media markets and away from media retailers is a democratizing force for the buy side. If the big guy doesn’t get a price break, it makes it easier for the little guy to beat the big guy at serving advertisers. When the market paradigm is dominant, like in search, talent and technology will win every time regardless of who is the biggest.

2) The media market paradigm advances the client-agency conversation from a discussion about media to a discussion about the client’s true business goals. When the media agency stops talking about delivering good media prices ( “Through volume discounts, our agency delivered $20mm in cost savings to your business this year.”) they have more time to talk about delivering customers to the client (”Through smart targeting and bidding on the $100mm of media we bought, we delivered new customers representing an estimated $400mm in lifetime value to your business). Once this more strategic conversation becomes the norm, and advertising is discussed as a business driver instead of a cost center, agencies and clients will both be better off.