Jonathan Mendez's Blog

The Two Modes of Delivering Relevance: Value Ad & Net-New

Yesterday with much excitement I dove headfirst into two new tools. Glue from AdaptiveBlue and LinkedIn Apps. Both provide for the distribution of relevant content into social media -- or if you prefer, the distribution of relevant social media into content.

Glue is a FireFox extension that provides social context to content while LI Apps easily creates apps from a number of content sources to put on your LI profile (assuming the early glitches have been worked out... How ya like me now?).

Installing them one right after the other, I could not help myself in comparing them since ostensibly they both have the same purpose (as all syndicated content) – optimize the user experience by providing value in the discovery and delivery of information. What is interesting is that while the ultimate goal is the same for both services, they are using opposite ends of the content/user spectrum to achieve it.

LinkedIn Apps uses content I created that should be relevant to others based on their context. (Note: there are a few collaborative apps as well)

Glue uses content others created that should be relevant to me based on my context.

Another way to look at this -- with LinkedIn Apps I become something of a social advertiser. As a Glue user (and of course as a content contributor) I become something of a social publisher. The irony is that Glue is playing the transactional web and LI lives in the social web. I find that juxtaposition fascinating.

What I realized today and what drew me to even blog this is that the services highlight ideas I’ve started to explore in recent work. Namely, that there are two modes of delivering relevance and each needs to be optimized toward different user goals for performance.

Two Modes of Delivering Relevance

1) Value Added Relevance: The content is a helpful & useful part of an experience or primary goal

2) Net-New Relevance: The content is interesting enough to change the behavior or cause latent goals to emerge

Let’s look at Glue and LI Apps in the context of these modes.

Value Added: My linked in profile page is there for the purpose of discovery -- learning about me. Apps provide added content that makes that experience much more informative. Another cool thing is that when I’m on my LI and my goal is to learn about other people, apps is there for me as well providing updates on my network. LI Apps provides Value Added Relevance.

Net-New: In the transactional web consideration stage is aligned with discovery goals. Glue sits on the transactional web to “connect with friends around the things you visit” blending social navigation with semantic navigation to deliver discovery via contextual relevance. Semantic web is all about correlations so it makes sense to incorporate social into these taxonomies. While is Glue is adding relevance to the primary goal of the user, the goal of Glue is for latent goals to emerge via social connections. Thus, Glue seeks to provide Net-New Relevance.

Further Definition

Net-New Relevance is about creating awareness. To be successful at it timing is critical. The more down-funnel you deliver Net-New the more you are fighting against the goals of the user so it is at top of funnel experiences where you can optimize (being helpful and useful to the user) by presenting this information.

Value Added Relevance can also create awareness. However, the primary benefit is to help someone complete a goal. Ideally this information presents itself between the current context & the user and increases in value the closer someone is towards completing their goal.

Heads up

One product I like very much in this “semantic discovery” space is SemantiNet’s headup . CEO Tal Keinan was on a panel I moderated at Jupiter’s Web 3.0 conference a few weeks ago and I had the opportunity to walk through headup with him.  Like Glue, headup is a FF extension only it runs on top of Microsoft’s Silverlight. Headup is keyword driven. The tool works more like a faceted search result to provide Value Added Relevance and uses a built-in social utility to deliver Net-New Relevance as well. 

Both Glue and Headup are worth checking out and likely each will find its audience depending on how you like to use the web. LinkedIn Apps is essential to use. 

Keep in mind these awesome services are only baby steps in the move from a web of pages to a web of objects. Still, they provide a window on how useful and interesting that web will become through optimizing the delivery of information about the things that are relevant to us.

We are just getting started here folks. I promise you, one day the web will be so fucking relevant you will crave irrelevance! And it will be there for you to click on.

The Road to Nowhere: Behavioral Targeting Still Sucks

This post is inspired by and repurposed from my recent comments on the great blog of Jim Novo.  

From my experience profile creation by cookie, also known as chasing people around the web, is not an effective practice for performance minded advertisers (and isn’t that everyone now?). It is a bad strategy to guess where I am headed by looking at the tracks I have left behind.

This is because one of the key components of using targeting for persuasion is timing. We have to be intelligent enough to know when the user is in a context for the ad to generate attention from them. To me this issue is best solved by content matching – ad content to site content.

Previous browsing “behavior” is a bad indicator of real-time relevance and the web is real-time media. There are so many flaws with BT (cookie deletion, shared computers, work/home) but the basic fact is the behavior of people is simply unpredictable.

Behavioral segments always run the risk of becoming sliced too thin to scale but the more you slice the segments the more predictable behavior becomes. Conversely, the larger your profile base the more variance you find in behavior. Ironically, BT has been sold as a panacea for Brand advertising. Makes no sense. If anything, it would work well for DR except for the fact that economically it makes no sense there either.

One new way BT is looking to improve is by inputting search keyword history into profiles. Google & Yahoo are doing this now. However, until people fundamental change how they search (generically) success will be limited and certainly not solve the issues of scale and demand-gen.

Another giant problem I would like to amplify is the role of creative. BT or no BT traditional forms of display creative are just never going to perform well. What’s the performance of a billboard? Publicis and WPP have set up offshore production facilities to create thousands of banners for campaigns so they could be optimized. Throwing (away) money at this problem is not going to work either.

Let us move on to new ways of targeting people with display ads. While we are at it why don't we eliminate the fat cats. There is huge value creation potential for businesses, advertisers, publishers and the audience if we disintermediate the low performance/high cost Planning>Creative>BT systems.  Businesses and marketers need to stop putting up billboards and start paving roads!

Previous posts kicking BT in the ass:

Behavioral Targeting Standards - RIP BT 1.0

Behavioral Targeting is Not Just Banners

7 Facts About On-Site Behavioral Targeting

While I am opening my cans of whoop ass, previous posts on Agencies:

The Next Generation of Digital Marketing Agencies

Brand Marketing in the Digital Age

Platforms, Applications and the Future of Digital Marketing

Think Like a Search Engine – APIs & User Control

There are two powers I feel are essential to unlocking the future of relevance in digital marketing.  

1) The power of user control

2) The power of information correlations

This is not new. You may have noticed the past few months that I keep referencing Google in my writing and presentations on this subject. This is because search has always best exemplified the relevance that can be delivered with a marriage of content and control. We can learn so much from what has made search successful.

Google Search is after all a product. Maybe the greatest product ever created because it’s free to users yet its value to publishers and advertisers is of epic and ever growing importance. It’s also no accident that the success is due to the fact that Google is an information product. Or that delivery of its information is controlled by user inputs and self-selection.

Information also happens to be the most valuable asset in a market driven economy. Information powers decisions. Information powers persuasion. Information, as Ogilvy said, is what advertising should be about.

One challenge in digital media the past decade has been creating alternate information discovery and recovery solutions in channels outside of search. Summize (now Twitter Search) is a great example of a recent product that got it right. The value of information is not in publishing it but in the distribution and discovery of it.

In my opinion Twitter Search is a far more powerful app than even Twitter itself -- though one wouldn’t be possible without the other -- and that’s the point. Site search platforms like Endeca and targeted content delivery systems like richrelevance also push the needle ahead. Surely YouTube and Facebook are as much about searching as anything else. The last vestige seems to be display advertising. I know we're making progress there as well.

Herbert Simon said (back in 1971!!!) that “a wealth of information creates a poverty of attention,” and that we “need to allocate that attention efficiently among the overabundance of information sources that might consume it.”

APIs can do exactly that! They provide us the tools to improve how we allocate information and attention efficiently. APIs allow us to rethink when, where and how we discover and recover information. But in doing so APIs also cry out for us to start thinking like a search engine -- if for no other reason than because no one has solved Simon’s poverty of attention problem better than Search.

Search, because of its user controls and high degrees of relevance delivery has conditioned us in how we want and need to have information delivered. APIs can benefit from the fact that even when people are not actively doing a search they always have primary, secondary and latent goals in their mind as they surf the web.

Aristotle understood that no matter what he said, relevance was the divine provenance of the listener. As marketers and advertisers we would be wise to heed his wisdom. Now we have the technology to do it – anywhere on the web.

Spitting in the Face of Pessimism

If you are an entrepreneur in online advertising and you don’t see this economic downturn as good news, then your business model is fundamentally flawed.

All the pessimism I am reading seems to be coming from investors and entrepreneurs involved in businesses that have poor models. Did they learn nothing from the first .com bubble? The value creation for publishers was not about content, but the discovery and distribution of it. The value creation for advertisers was not about spend, but return. The keystone for both continues to be technology & relevance.

Been There Done That
I lived through the first .com bubble. I founded my first company in 1998 selling nutritional supplements and by 2003 we had won the Nutrition Business Journal Award for growth. While the “smart money” companies went out of business we bootstrapped our way to 30% YOY. We thrived as companies offering little real value to consumers died. I’m fired up to do it all over again.

We did it with a daily commitment to analytics, testing and optimization. We did it by focusing on user experience and convenience. And mostly, we did it by using a new marketing strategy (SEO) with an emergent technology (Google) that delivered relevant content to people. We got better at all of it every friggin' day and guess what...the Internet never went away. Neither did people's thirst for relevance in their use of the medium. I know that formula will work again.

Opportunity O’Plenty
Excuse me if I’m not shedding a tear about the current economic climate.

Some perspective:

• Between early 2000 and 2002, the Nasdaq index fell almost 80%

A different perspective (all deals 2007):

• Quigo Founded 2000 sold for $363m
• Tacoda, Founded 2001 sold for $275m
• RightMedia Founded in 2003 sold for $680m
• BlueLithium, Founded in 2003 sold for $300m
• Optimost 2001, Offermatica, 2003 combined price of $125m

Those exits above were all companies founded during the down cycle that were insanely focused on building new technologies that demanded a different, more relevant way of thinking about advertising. When the market turned positive they were ready to capitalize on it -- and they did very quickly!

The Relevance Phoenix
What the markets are doing now is taking a giant laxative and flushing out the crap. Online advertising will not be immune nor should it. The downturn is exactly what the web needs to focus attention on disruption. Many of us, especially in display, have been comfortably numb for the past few years riding the gravy train. Now is the time to build more advanced and relevant systems. Systems outside the inefficient and antiquated Client>Creative>Media>Network>Site>Analytics hierarchy. Systems outside the false prophets…or is it profits.

If you’re building widget distribution platforms that rely on pay-to-install and CPM arbitration then you are in trouble. Impressions & CPM is a bad place to be. If your homepage has the word “behavioral” on it, also, not good. High cost, low return, little insight just isn’t worth it. If you are promoting the size of your network is or how many eyeballs you reach guess what, no one cares. Get relevant or go home. Be different or become deceased.

That’s why I’m spitting in the face of all this pessimism. Relevance and its child, Performance are something every business needs to be investing in over the next few years. Especially in advertising and especially in online display. So hallelujah, performance is what (finally!) matters. Measure it anyway you want. Call it anything you please. Just deliver it!

Critical Mass
Most of all, I don't give a shit about the gloom and doom because things are different this time. Let’s recall, the three-year period from 2001-2003. Online ad spending was flat at around $7 billion per year. In the four ensuing years from 2003-2007 online ad spending tripled to $21 billion. This year it should be around $28B. Even zero growth the next few years ad spend is now 4X what we had in 2003! That’s more than enough dollars for innovative start-ups to prove their mettle. Believe me, clients will be looking and listening to anyone that can improve their bottom line ad spend ROI.

Last point. Online has as much attention as TV but TV has 68% of ad spending vs. 7% for online! This is a big nut that’s going to crack. One Internet CEO I spoke with recently told me he’s starting to see media planners put TV dollars into his online platform because it is much more measurable and can be optimized. This may be a ripple now, but watch, the next few years as it becomes a wave and the companies just getting started now will become the big exits of 2012. So stop crying in your beer and get to work.

Display Becomes Us

Scott Rafer gets it. I don'€™t know Scott, though I think we both spoke at CommunityNext and WidgetWorld Expo over the past year. In any case, his thoughts on the Lookery blog capture an important point that deserves an exclamation.

"The increasing marginal returns of search advertising are now doing more than taking market share from display advertising, they are en route to becoming display advertising's operating principle."

I drove home the exact same point during Advertising Week!

I always placed my bets with search. For a decade I focused my career on it. First with SEO, then UX, then the buy side with campaign & bid management and finally the sell side with landing pages, multivariate testing & content targeting.

However, if you are to believe Marissa Mayer of Google, search is 90% solved. Working in Search all those years, I believe her. That is why I started RAMP and I am placing my bets now that what Scott sees will indeed become reality. Search will continue to morph into a Display OS.

Contrary to Search, I estimate Display is only 10% solved. The opportunity dwarfs search. Besides all the hurting publishers and advertisers the changes will drive the emerging channels of online video and social media not to mention already creeping into TV, Mobile and Outdoor.

Display will rise again but it will look nothing like it did before. There will be plenty of road kill. Good riddance.

Bring relevance. Get performance.

Here are a couple of posts from January & June 2007 focusing on search practices emerging into display (or vice versa depending on your point of view):

Real Behavioral Targeting Focuses on Intent

Behavioral Targeting is Not Just Banners

Lessons in Ad Persuasion from Aristotle

In digital advertising persuasion is becoming more difficult, complex and valuable. Often, I find it helpful to strategize persuasive techniques for clients by going back to the originator of this thought, Aristotle.

Two and a half years ago an early post on this blog discussed Artistotle’s concept of Kairos (timing) and how it could persuade behavior in the emergent real-time web. While real-time web has not evolved as fast as I may have hoped the past 2 ½ years the way humans are persuaded has remarkably not evolved at all in 2400 years.

Aristotle saw three means of persuasion. Below are my best CliffsNotes version and some real world examples.

1) Character of the speaker
In Aristotle’s world there is no brand equity, only credibility earned from the speech through intelligence, character and good will. In the technical means of persuasion, what the speaker says and how they say it is ultimately what matters in determining their character.

Great examples of the value in using intelligence (technology) in a way that creates good will (User Experience) are the iPhone, Google, Firefox.

2) Emotional state of the listener
Emotions have the power to modify our judgment. What’s important to understand here is Aristotle’s aim of persuasion was to arouse emotions in order to persuade a certain judgment, not an action.

Search ads are a great example because the action (intent) already exists. These ads then, like so many others, have to be clicking with people on a purely emotional level (hello brand marketers!).

3) The argument
We persuade when we prove something is the case. Aristotle defined two modes of argument. Deductions and Inductions. Induction proceeds from particulars to a universal in order to give an example. Deductive is the example of proof or a demonstration with premises and a conclusion.

Apple exemplifies both tactics well with their advertising. The “I’m Mac” ads use induction and the iPhone/iPod Touch ads use deduction.

What's next for persuasion?

There’s no question we will continue to see advertising determine character, create emotions and make arguments. The sea change is that technology will underpin persuasion. Creative, while still the king, will have to adapt so it leverages and amplifies both its role in persuasion and the benefits of these new technologies.