Nov 27, 2011

The Golden Age of Internet Marketing?

"Let the products sell themselves... fuck advertising, commercial psychology ... psychological methods to sell should be destroyed." - The Minutemen, Shit From An Old Notebook
We have enough data now to realize that display advertising on the Internet doesn't work. Some suggest click through rates are as low .09%, a shockingly low number. It seems like click through rates are trending towards zero. Users don't notice the ads, they don't click on them. Ultimately, will we eventually see Google giving impressions away for free?

No amount of targeting, behavioral or otherwise, will solve this problem.
There are a number of reasons why, but the main one is that display ads online are the wrong metaphor. They come from a construct where web services were viewed as "pages" - magazine pages. They were invented by applying an old model (magazines) onto a new medium (web services) and assuming that the user is a "reader" and will accept being interrupted. Over time, the web has proven both these paradigms to be untrue in a truly profound way.

In short, web display ads are not web native; therefore they do not and will not work.

However, it also feels like we are about to enter a new, maybe a golden, age of Internet advertising and monetization. Even the word “advertising” in this new golden age is not accurate. Josh Stylman said to me that “advertising as we've known it is dead. Marketing on the other hand, may be entering a golden age with the ability to spread ideas in a way that pundits only dreamed about 10 years ago.”

Web services now exist at a scale that dwarfs the old “web page” model, and the value that many of these services deliver derives from users as contributors, not simply viewers. They then lend themselves to native business, or advertising models (again, the word advertising hardly applies here because this is not like advertising as any of us currently imagine it).

These new emerging revenues streams will be native monetization models that are consistent with the fabric of the product, that run with the grain of how users interact with and use the service. Google ads are the perfect, and prototypical, example, because they deliver a unit in a manner consistent with the way the user is using the product to search for information. These units generally work because they align the interest of the three interested parties in a search: users, marketers and publishers. Users get what they are looking for; marketers' get performance on their spend because they buy against the search action itself; and finally publishers generate traffic from the content.

Other examples - new marketing products - now emerging that are beginning to solve the user behavior/marketing experience include StumbleUpon Paid Discovery, Twitter Promoted Tweets, Buzzfeed Social Content, Facebook Sponsored Stories, Percolate Brand Curators, foursquare for business. Those are just six examples where the “ad” unit is consistent with, and integrated into, the very fabric of those social services themselves. Six examples developed only in the past few years.

I imagine Tumblr, Instagram, Soundcloud and other services will introduce similar initiatives, again that are consistent with the way their services natively work. These platforms, as James Gross of Percolate likes to say, are brokering interest across vast information networks, so in order for a brand to succeed they must broker interest in a native way that makes people enjoy them.

These are all new, and as a result will cause some confusion amongst the “buyers” of the products (as Fred says, The Fragmentation of Online Marketing), in the same way that Google ads originally did. They will take a while to be adopted, maybe even years. But I believe they will work and scale. Because when they do, hardly anyone will even notice them. “You want to not look like an ad at the first glance, but to look like an ad on the second glance.”

The products will sell themselves.


Local Markets Team said...
This comment has been removed by the author.
Jon Steinberg said...

Agree on the "Golden Age" - I call it a "coming creative revolution"

RickWaghorn said...

We have enough data now to realize that display advertising on the Internet doesn't work...

Someone go tell, Eric. He foresaw a $200bn business last summer...

dherman76 said...

Andy, as someone who is involved with advertising on a daily basis, I totally agree. However, we do have evidence where "display" ads do work, when planned and measured accordingly. It's not easy.

I wrote a recent post that I think you'd really like... it's called Helping Demand Find Supply. Enjoy:

Anonymous said...

What is the CTR of the magazine ads?

Adam Schwartz said...

I don't view platforms like Facebook, Twitter, Tumblr as advertising platforms. Instead, I view the Internet as the platform. And you activate that platform by creating original, relevant and compelling content. The platforms are simply distribution channels to drive to your content.

Over time, technology commoditizes. Content doesn't.
Invest in original content creation and user experience. They are always the best advertising.

Nicholas Marx said...

I wonder how recommendations will/can fit into advertising:

Hashem Bajwa said...

This is a great perspective Andy.

Good marketers would much rather have their content be experienced in a way that is authentic within the environment they are trying to tap into as they will be more relevant to the audiences or users of those products or places.

I always liked the intent behind "Diggable Ads" on Digg where a brand's content would be treated like all other Digg content -- digged up if it was good, buried if it sucks.

There's a challenge in terms of resources for a marketer and maybe a publisher in terms one size fits all monetization vs more tailored approaches for each, ie taking one banner and blasting it across a dozen sites is easier than building 12 unique pieces of content for 12 different apps or sites. But that's also a laziness left over from traditional media channels and an evolution that is fast happening. Mobile will help force this change as display doesn't have a legacy in mobile to undo.

What I hope changes with that is the "ad tech" world that is huge and highly fragmented, with hundreds (thousands?) of players trying to make display more efficient or automate or optimize old models.

I hope smart products and publishers can keep leading the way.

Jonathan Mendez said...

Indeed. As you know this is right out of our Yieldbot playbook. One small tweak:

I don't think it's as much... "models that are consistent with the fabric of the product, that run with the grain of *how* users interact with and use the service." it is running with the grain of *why* users interact with and use the service.

I highlighted that exact disconnect a week ago in a blog post about Promoted Tweets. Twitter's ad product is built for "how" - it should be built for "why"

Andreas said...

It is a nice thought, but advertising is there to create demand ... and without the demand there is a lot less search ... so less action on google adwords. And if that's the case, google will change their business model.

anyways - the ideas are nice, but once these new forms of revenue streams are in place, people will earn a lot less money with content, and this will also bring down online publishing and innovation.

Unknown said...

brilliant post, andy.