Chief Marketing Technologist by Scott Brinker

Semantic branding and gender politics

2008-08-07T09:01:13-04:00

Just read a fascinating article on ReadWriteWeb, Will the Semantic Web Have a Gender?, that perfectly illustrates the challenges of semantic branding.

The article summarizes arguments made by Corinna Bath, a researcher who studies gender and technology, in a recent interview. She makes a case that the semantic web may develop a gender bias because the categories of relationships that are becoming standardized — e.g., the Dublin Core ontology — subconsciously reflect the worldview of male computer scientists. This will further calcify certain male-centric concepts.

(I think of this as analogous to how "he" became the default pronoun in the English language.)

Whether you agree with Bath or not, the subtlety she's pointing out is brilliant: our choices of models and categories will have tremendous impact on how "knowledge" is represented, perceived, positioned, and ranked in the semantic web.

This goes right to the heart of semantic branding.

In my first article on semantic marketing, I wrote that semantic branding is how a company will categorize its data — and hence its products, services, differentiators, thought leadership, etc. One's choice of metadata will strongly influence how and where its data is used, distributed, and interpreted.

This constitutes a new kind of market positioning and placement via the semantic web.

But this categorization is by no means black-and-white. Making the assumption that the semantic representation of a company can just happen automatically misses the point that associations and characteristics are rarely absolute and are loaded with value judgements. (This may ultimately be the fatal blow to "top-down" approaches to semantifying the web.)

The categories and characteristics that get associated with something in metadata are not a "given" — they can be highly subjective and error-prone. Just because something is presented as true, doesn't mean it is true. We learned this in the visible web and on The Colbert Report, but we may need to relearn it in the semantic web.

In a categorization-centric worldview, the definition of categories — and the categories and models that become dominant, such as the Dublin Core — wield tremendous power in shaping semantic conversions and positioning.

In the semantic web, companies/people/concepts will have a brand that will very much be defined and colored by how they are modeled and represented.

According to Bath, the semantic web is in danger of having a strong "male" brand. She may be right. But the why and how of that branding is going to be a recurring issue across many, many domains.

As semantic web models and categories across most fields of knowledge are currently undefined or in early ferment, there is great opportunity for first movers to frame the playing field in ways that are advantageous to their brand.

Marketer-technologists for agencies

2008-07-08T21:06:27-04:00

Earlier this week, Adweek published An Open Letter to the CEO of WPP that I wrote as a guest columnist.

WPP is one of the top advertising holding companies in the world — up there with Omnicom, Publicis, and Interpublic. Its more well-known subsidiaries include Ogilvy, Young & Rubicam, and J. Walter Thompson.

The reason for my letter was to dissuade WPP from fighting Google and Microsoft directly to get a piece of the current "ad network" gold rush frenzy. My main points were:

Ad networks may not dominate online marketing as much as their early hype would suggest.

Rather than fight over ad networks and technical gadgetry, the major agencies would do better pursuing a strategy that elevates effective "creative" to the top of the value chain.

They can do this by:

expanding the scope of advertising to include post-click marketing;

bringing more marketer-technologists into their teams to leverage technology as part of the creative;

and organizing their industry to force ad networks and technology providers to more open standards and greater transparency.

As I've written about before on this blog, I strongly believe that the future of marketing will thrive on an intersection with computer science and career paths for marketer-technologists — led by chief marketing technologist executives. This will hold true for agencies as well as in-house marketers.

Quoting from my letter in Adweek:

...you must absorb marketing technology savvy into your DNA. Although creative cannot be bottled into a computer program, the nature of creative is changing. More than ever the message and the medium are fused together. You need to wield targeting and tracking software as part of the creative mission. To do this, your companies need to be populated with marketer-technologists — a new career track tightly integrated into your core business.

This is the key to competing with technology companies: You must show that the real value is not in the tools themselves but in the way in which they are employed as part of marketing strategy, creative and execution.

Click here for the full letter.

Debate on The Long Tail

2008-07-04T07:27:20-04:00

Chris Anderson's theory of The Long Tail in Internet commerce and marketing has been one of the more influential concepts in online strategy and business models. Amazon, Google, eBay, iTunes, NetFlix — these are all major Long Tail success stories.

I'd contend that many of the current best strategies in search engine optimization are essentially derived from Long Tail theory. These ideas will therefore be naturally extended into semantic marketing and SEO++ accordingly.

Even MySpace, Facebook, and Twitter groups and followings are another kind of incarnation of The Long Tail — with similar implications for the new field of social media marketing.

In other words, when it comes to modern marketing — and marketing technology — The Long Tail is an underlying force with which to be reckoned.

If you don't know much about The Long Tail, here's the Wikipedia entry and a link to Anderson's own blog. Apparently, a new edition of the book — with an added chapter on Long Tail Marketing — is just now being released. My one sentence description of The Long Tail is this:

The Internet has enabled an "aggregation of niches" to successfully compete with a small number of blockbusters (which used to be the central focus of commerce and marketing).

However, no good theory goes without debate, and a recent article in Harvard Business Review, Should You Invest in the Long Tail? by Anita Elberse, does its best to poke holes in this one. In my opinion, it doesn't succeed particularly well, as many of the arguments presented seem like semantic debates on "what's the tail" and "what's success".

But the blog discussions around the article are fascinating. Read Chris Anderson's response on his blog, as well as two threads on HBR's blog here (starts with repeat of Anderson's response) and here (Elberse's response to Anderson's response). Even TechCrunch has a thread. Some of the comments are the best parts.

I do think that Elberse raises some interesting ideas about social dynamics in The Long Tail. In her response to Anderson, she cites the marketing phenomenon known as "double jeopardy" to suggest that blockbusters in markets have a reinforcing virtuous feedback loop — once a brand dominates a certain space, it's easy for it to keep dominating it and very hard for someone else to break in. I'm not convinced that this pattern will extend as seamlessly into the digital domain, but I certainly agree that it's worth paying attention to. I think some of the dynamics in the social media space — the evolution of social networks such as Facebook and MySpace — will shed fresh light on this.

I also think that ongoing studies of The Long Tail will give us a better, deeper understanding. One of the most obvious courses of research is to look at these effects in a time series: compare the evolution of digital inventories, sales distributions, and the dynamics of how a "hit" becomes a hit (and for how long).

Exponentials in marketing: watch out!

2008-06-29T13:47:13-04:00

Exponential growth in marketing is not just a colloquial way of saying, "wow, there's an awful lot going on in marketing these days". That's true, of course, but I'm specifically fascinated by — and concerned about — actual mathematical exponential growth, where something doubles, doubles again, doubles again, and so on.

Let me start with the legend of the chessboard, as adapted from Wikipedia:

When the creator of the game of chess showed his invention to the ruler of his country, the king was so pleased that he gave the inventor the right to name his reward. The man, who was very clever, asked the king this: for the first square of the chess board, he would receive one grain of rice, two for the second square, four for the third, and so forth, doubling the amount each time for all 64 squares on the board.

The ruler, who was not strong in math, quickly accepted the inventor's offer, even slightly offended by his perceived notion that the inventor was asking for so little. He ordered the treasurer to count and hand over the rice to the inventor.

However, when the treasurer took more than a week to calculate the reward, the king demanded a reason for his tardiness. The treasurer replied that it was impossible to fulfill the inventor's request. By doubling the number of grains of rice 64 times, the inventor would be entitled to 18,446,744,073,709,551,615 grains of rice.

Or, put another way, if all of the Earth's arable land was dedicated to producing rice, it would require more than 100 years of abundant harvests to deliver the reward.

The inventor thereby owned the entire kingdom.

So what does this have to do with marketing?

Online marketing is overflowing with ideas that contain exponential growth. Many of these ideas sound reasonable — and cool — when examined on a small scale. But as they start to ramp up, they quickly become monstrous. These are actually called Second Half of the Chessboard problems because, like our story above, the first few moves seem fine... but about halfway into it, you suddenly realize you're in trouble. ("We're going to need a bigger boat.")

Consider this example from web marketing:

Multivariate testing (MVT) of elements on web pages, a topic I've talked about on ion's post-click marketing blog (Playing Russian roulette with your landing pages?). At first, MVT sounds pretty cool: you load up a bunch of different headlines, images, offers, etc., into a web page, and then let the MVT software try different combinations to pick the best one.

But do the math. If you try 8 different headlines, 6 different subheads, 7 different images, 3 different body copy blocks, 3 different background colors, and 5 different offer buttons, that's 8x6x7x3x3x5 = 15,120 possible combinations. Then, if you then consider 20 different ads and vehicles driving traffic there, that's 20x1520 = 302,400 combinations. And what if you have 4 different audience segments you want to optimize individually for? Now you're up to 302400x4 = 1,209,600 combinations. All of a sudden you might need 100 million or more respondents in order to reach statistical significance to determine what is a winning combination. This quickly becomes an unmanageable muddle.

Other examples include:

The number of distinct marketing vehicles being engaged in Long Tail marketing campaigns. Think about all the keywords you might buy in a search marketing campaign. Then multiply by all the different creative ad variations you might try. Say it's 250x20 = 5,000 today. Now, add a new campaign, and this can double overnight. Which means your reports and scope of analysis double overnight. But the hours in your day and the headcount on your team probably didn't double.

The growth of data analytics and behavioral tracking. Think about analyzing the "path" someone takes when they visit your web site. If your site has ~20 navigational choices on each page — some sites have hundreds! — then tracking the first 3 pages visitors go to has 20x20x20 = 800 possible paths. Maybe you think that's not too bad, so you decide to look at the first 5 pages... which is suddenly 20x20x20x20x20 = 3,200,000 paths! Whoa. If you're now trying to determine someone's intention or segment from this huge pool of data, it's nearly impossible to achieve statistical significance.

The number of combinations in a chained sequence in marketing automation. A prospect goes from seeing an ad, to visiting a landing page, to receiving a white paper, to visiting your site, to getting a follow-up newsletter, to reading your blog, to being given a special offer, etc. The dream of marketing automation is to have this process flow with minimal intervention by the marketer's staff, while each prospect feels like they're getting individual and personalized attention. But like the web site paths in the previous example, every step added to the chain multiplies the number of possible combinations — quickly millions and billions and trillions — again making it essentially impossible to understand the exact dynamics of every interaction history. Closed-loop marketing is great in concept, but hard (NP-hard!) in practice.

The viral explosion of a hot idea — or reputation crisis — in the social media sphere (i.e., they tell two friends, who each tell two friends, who each tell two friends, and so on).

For you as a marketer, responsible for managing your company's brand and channels in the world, the sheer number of things competing for your attention (the attention economy). Just when you thought you were getting search marketing under control, social media marketing is exploding. Do you have a Facebook strategy? Great. Now what about a Twitter strategy? Friend Feed strategy? Widget strategy? And now, whoops, Google has just changed their policy on landing pages, so you have to go back there. Your boss calls: what's the current update on any one of these things?

Dizzying, isn't it?

However, the first step to recovery is acknowledging the problem. Exponentials are most disruptive when they catch people by surprise. Don't be the king with the chessboard problem.

Avoiding these areas where exponentials lurk isn't an option either. You're going to have to deal with campaign optimization, Long Tail marketing, large-scale web analytics, marketing automation, and viral effects in the marketplace.

So if you can't ignore them, and you can't avoid them, what do you do?

Here are some practical strategies — many adapted from computer science — for managing exponential growth in marketing problems:

Know your ceiling and strategically stick to it. Like those signs on overpasses telling trucks how much clearance they have: low car, no problem; tall truck, big mess. Most exponentials are fine for the first few steps, so as long as you know in advance where the numbers go from being reasonable to unreasonable — and explicitly draw the line before they become unreasonable — you can work with combinatorial ideas. Don't expect (or promise!) those techniques to scale beyond a focused, narrow window.

Limit the number of multiplicative steps where you have control over them. Instead of engaging in multivariate testing of a web page with thousands of combinations, do a simple A/B split test to experiment with a new idea — it's fast, easy, and clear to understand the outcome. Instead of using landing pages with full navigation — dozens of options on each page — temporarily narrow the choices presented to users to 2-5 relevant choices, at least for the first few pages. From the user's perspective, this can be good context-relevant navigation. From your perspective as the marketer, this can reduce the number of "paths" being analyzed from an unmanageable exponential to a relatively tame one.

Reduce the granularity from previous steps. For instance, you might have hundreds of ads driving to dozens of landing pages, all of which eventually generate leads that then need subsequent re-marketing. At the point you convert a respondent to a lead, you can grade them (A, B, C) and segment them into maybe 5 buckets. Now instead of having to directly deal with the thousands or millions of combinations in which a lead was created, the marketing automation logic at the next stage of the funnel only has to deal with 3 grades x 5 segments = 15 possibilities. This "reduction" process can happen again and again further down the pipeline. But be careful! This is a lossy one-way function: a B-grade respondent in a particular segment may have gotten there many, many different ways — and perhaps some incorrectly — so you have to balance your assumptions for how to deal with them accordingly.

Only examine one thread of past choices at a time. For instance, looking at the choices a single person has made — which ads they responded to, what they did on the landing pages, what subsequent interactions they've had with your company — is reasonable. However, this means specifically looking at that one person's history and not trying to generalize it into rules or predictions — because as soon as you do that, you're back in the fire with explosive combinatorics.

Intentionally drop things off the back end. If you keep giving a juggler more and more balls to coordinate in the air, eventually some are going to be dropped. But in marketing, not all balls are equal. Instead of dropping balls randomly — where, by Murphy's Law, the most valuable ball is the one most likely to get dropped — consciously choose which balls are low priority, expendable. And let them go. For example, once certain previous campaigns have passed, remove them from your reports and analysis. I'm not saying that there might not be relevant data there — there very well may be — but if it's not as relevant as other factors, take them out of your field of vision. Simplify to clarify.

Techniques for managing continuous partial attention. This particularly applies to exponentials around the number of things you're responsible for running well. It's a different way of working, and new software and processes can facilitate or leverage it.

Apply probabilistic and approximation algorithms to the problem. This last approach is arguably the most powerful, but it can also be easily misused — without evening knowing that it's being misused. In this approach, the software doesn't guarantee completely accurate results, but it takes its "best guess" by working with a random sample or subset of the data and extrapolating the results. This can be an acceptable trade-off, as long as you know what you're trading off. What exactly is being randomized? How much margin of error is there? What are the consequences when a wrong guess occurs? I believe that one of the key advantages of bringing more computer scientists into the marketing domain will be incredible innovation with algorithms of this kind and better, safer intuition for understanding when and how to apply them.

So now, as the ruler of your marketing kingdom, hopefully you're primed to ask the right questions — and use the right judgment for practical outcomes — when clever exponentials try to sneak by you. Be on the look out! In online marketing, they're everywhere.

P.S. To keep exponentials in perspective: if you were to start with 1, double it to 2, double it again to 4, and keep doubling only 267 times, i.e. (2)^267, the result would be equal to the total number of atoms believed to be in the observable universe.

BrandTags.Net -- A Great Experiment

2008-05-12T20:17:10-04:00

Noah Brier put together a fantastic little web application, BrandTags.net.

The site shows you brand logos and invites you to respond with a one-word or one-phrase "tag" that pops into your mind when you see it. After you submit your tag, you can then see how other people have responded by viewing a "tag cloud" that shows all the words and phrases that other people have submitted, where the size is larger the more people who submitted the same thing.

So, for example, you might see the CNN logo. I entered "larry king" as my phrase. When you drill down to the tag cloud, not many people have answered Larry King, as it's very small. The word "news" is humongous. "Biased", "liberal", "lies", "propaganda", and "tv" are also rather large. Well, two out of five ain't bad.

As Noah remarks in his introduction, "The basic idea of this site is that a brand exists entirely in people's heads. Therefore, whatever it is they say a brand is, is what it is." Somewhere out there, a CNN brand manager is spitting up blood.

Obviously, there's sampling bias and probably more than a few people who have maliciously spammed this experiment by now, so it's taken with a grain of salt. But it's fascinating nonetheless. Breathes new life into the tag metaphor, as an easy way for people to categorize and remark on things — even subtle and complex things — in a way that can then be collectively culled for insight.

Marketing in the cloud

2008-04-14T00:33:29-04:00

Marketers should have their heads in the cloud.

If you haven't read The Big Switch by Nicholas Carr yet, you need to add it to your list. Cloud computing — that is, computing infrastructure that is based somewhere out on the Internet, rather than installed on hardware locked in your company's IT center — is becoming real. Fast.

Combined with the maturity of web-based software-as-a-service offerings, the strong gravitational pull of social media sites where marketers now work beyond the borders of their company's sandbox, and the widespread proliferation of web services and mash-up APIs, the cloud has become marketing's new IT platform.

For marketers, this is a terrific opportunity (a) to re-calibrate the relationship between marketing and technology and (b) to expand your capabilities in the "new marketing" environment, where the pace of innovation in online marketing channels and methods continues to accelerate.

I think of "computing in the cloud" for marketing in fairly broad terms — more broadly than the technical definition of cloud computing — and divide it into four buckets:

1. Web sites where you are a participant or sponsor, particularly social media communities. These are sites such as Facebook, Twitter, LinkedIn, Xing, MySpace, Yahoo! Groups, Digg, del.icio.us, Twine, etc. This isn't what most tech people think of when they talk about cloud computing, but from a marketer's perspective, these are important services out on the Internet that you must plug into to do your job. Your audience is in the cloud, and you have to go in there after them. The IT aspects, however, reside outside of your control.

2. Software-as-a-Service (SaaS) applications. These are applications that you access through your web browser, are hosted on a server farm by a third-party, for which you have zero operational involvement. Some are free, such as Google Analytics, but many are provided on a subscription basis, such as Salesforce.com. Pay only for as much as you need for as long as you need. There are a plethora of SaaS applications for marketers available today, with more coming online every month — if you have a favorite that I've missed in my "for instance" below, please add it in the comments:

customer relationship management (ex: Salesforce.com);
landing page management (ex: LiveBall — disclaimer, my firm);
surveys and questionnaires (ex: SurveyMonkey);
email marketing management (ex: ExactTarget);
web site analytics (ex: Google Analytics);
marketing resource management (ex: MarketingCentral);
blogging platforms (ex: Typepad);
lead nurturing (ex: Marketo);
digital asset management (ex: Celum Imagine);
search marketing management (ex: SearchIgnite);
project management (ex: AtTask);
web conferencing and webinars (ex: WebEx);

3. Web services and on-demand IT infrastructure. If there are specialized applications that you still want or need to build yourself — including online applications that are built into the ecosystem of your product offerings — you can take advantage of components, services, and infrastructure layers out in the cloud. Your programmers will develop your app, which will nominally live inside your IT environment, but certain components or layers of it will be dynamically accessed in the cloud. These include:

data storage services (ex: Amazon S3);
semantic processing (ex: Reuters Calais);
mapping and geotargeting (ex: Google Maps);
APIs for Twitter, Facebook;
payment processing services;
feed aggregators (ex: Yahoo! Pipes);

4. Platform-as-a-Service (PaaS). The furthest down the continuum, platform-as-a-service means that you develop your own custom applications, but they live entirely in the cloud. Salesforce.com was one of the pioneers of this idea, by opening up their platform — force.com — to let any developer create application on their infrastructure. Google recently announced their Google App Engine with a similar vision.

The value proposition for developing on a cloud-based platform is that you get all the scalability, reliability, and high-speed performance of these environments without having to build up all that infrastructure yourself. Put in terms that marketing can appreciate: you can create a custom app on the web without necessarily having to engage the overhead or ongoing involvement of your IT department or corporate data center.

Advantages of cloud computing and SaaS/PaaS in general:

you can try a new idea with less up-front investment, which facilitates and encourages a more experimental culture — this is a great way to quickly engage the leading edge of new marketing channels and methods;
if you end up with an overnight sensation, you have "instant elasticity" for matching capacity as demand requires it;
economies of scale, since cloud providers are aggregating infrastructure for a certain type of service across hundreds or thousands of firms;
time from concept-to-launch of a new application is shortened, since you're not reinventing the wheel or waiting on physical installations;
more direct relationship between expense and benefit, as cloud computing is usually a variable cost;
direct IT maintenance overhead is almost completely eliminated;
for SaaS applications, new features come online automatically without having to engage in the delay, cost, or frustration of "upgrades";

There are, however, objections, caveats, and trade-offs as well:

you end up with dependencies on third parties for these applications, rather than having the safety (or the illusion of safety) of being self-reliant for uptime, availability, continuity, etc.;
your data can become fractured in many different silos by working with many different providers in the cloud;
security and privacy risks: what happens if someone breaks into your account or hacks the provider's database?
business continuation risk: what happens if the provider in the cloud goes out of business?
terms of service risk: what if the rules of how you can use the service change unexpectedly?
price change risk: what happens if usage costs rise unexpectedly?

While these are legitimate concerns that should be weighed when adopting any cloud application, the reality is that many of these risks already exist with your internal IT operations. Uptime, security, privacy, sustainability, data integrity, etc., are all issues that plague IT departments.

This is the "fear of flying" phenomenon. The odds of a fatal car accident far exceed the odds of dying in a plane crash. However, many people who are afraid of flying aren't afraid of driving because in a car their hands are on the wheel. This is a false sense of security. (Sorry, hope I didn't spoil your commute.)

Failures of major cloud computing applications and platforms get a lot of attention, but the number of internal IT "crashes" are actually much higher, the far majority of which receive little or no publicity outside a company's four walls. (And sometimes not even that.) Because cloud computing providers are under such intense scrutiny for their performance — their reputation, and hence their business, is on the line — they often take far more precautions and invest in heavier and more redundant infrastructure than IT departments hosting the same application for their own firm.

The other concerns can be mitigated by taking a few proactive steps and common sense precautions:

have a technical gatekeeper in marketing who vets cloud applications (e.g., a chief marketing technologist or some such role);
define and defend "core data" structures, while being open and flexible with more ancillary data feeds;
understand the service level agreements (SLA) and track record of the provider;
read the fine print of the agreement for terms of service;
how much control do you have to customize via configuration and/or APIs?
determine how easy is it to get your data out;
think through "switching costs": if your concept works, but this provider doesn't, how easy is will it be to migrate to a different solution?
follow security best practices with your cloud applications: good passwords, changed frequently, not distributed across multiple users, always used over secure connections, etc.;
keep tabs on your provider, an ear to the ground on their business, more closely than you would a software vendor whose product you've installed internally;

Overall, working in the cloud enables marketing departments to react much faster to new ideas and new opportunities, without necessarily having to drag along IT. But, as has been said, "with great power comes great responsibility".

As marketing becomes more in control of its technological destiny, it must absorb technical leadership into its management and ranks. The future of the chief marketing technologist awaits in the cloud above.

A new S curve for search engine ads?

2008-04-06T07:32:08-04:00

Are we on the verge of a new S curve for search engine advertising?

S curves are a phenomenon of technology adoption. They begin where a particular technology — in this case, ads on search engine results pages (SERPs) — arrive on the scene with chaotic, slow growth while early adopters figure it out. Then a dominant model emerges — e.g., Google AdWords text ads — where the winning approach and the benefits to the market become obvious, powering mainstream adoption at an exponential rate. But eventually growth levels off, adoption saturates, and incremental improvements and benefits plateau.

At this point, new technologies, approaches, and business models vie in the same space to be the "the next new thing" — to trigger a new S curve that will overtake the old one.

The beginning of this new S curve is again chaotic and slow: lots of ideas, many of which will fail, many of which are initially not as effective as the dominant winner in the old S curve. There's usually resistance from people on the old S curve, who have mastered that game and are reluctant to have the rules change. When you're used to something, especially something on which your livelihood is based, it can be hard to imagine something different.

But the forces of entrepreneurial ferment and early adopter incentives conspire to push past that skepticism and uncover the winners of the new S curve. Once they crystallize, and the new value proposition comes into focus for everyone else, people jump from the old curve to the new curve, and the new dominant model takes off at exponential speed.

And the cycle of disruptive innovation continues.

Recent events make me wonder if a new S curve is soon to emerge for search engine ads:

1. Growth in US online ad spending is slowing. A recent eMarketer article reported that, in particular, "comScore data shows a second straight month of slower growth in paid clicks for Google's main ad-serving business." Now growth is still growth, and given the talk of recession, a slowing of search ad budgets may be a temporary blip. But it's also consistent with the hypothesis that search engine ads are nearing the top of their S curve. Since "paid click measurement is critical for Google" financially, they have strong motivation to find the next catalyst of growth.

2. The Google Operating System and Digital Inspiration blogs recently captured screen shots of tests Google is running to include video ads alongside select search results. The interface is cautious about maintaining Google's minimalist design standards: ads with accompanying videos have a small button with a plus sign that can be expanded to watch a commercial or testimonial.

Although this isn't a huge surprise — Google had previously announced its intention to try this, with commentary from Danny Sullivan's SearchEngineLand and the Mashable blog back in February — it sets the stage for a major shift in the concept of search engine marketing. There's a big difference between 130 characters of stand-alone plain text and 130 characters of plain text as an introduction to a 30-second video spot.

Whether this particular ad format takes hold or not, it signals Google's intention to experiment with new advertising ideas in their previously sacrosanct SERP interface. Marissa Mayer of Google told the New York Times that the company would explore adding small thumbnail photos to the video ads as well as considering other formats that may include ads with images. That's a momentous shift.

3. Yahoo's plan to render structured semantic web information in its organic search results. This will no doubt have a major impact on search engine optimization (SEO) practices, a movement I describe as SEO + Semantic Web = SEO++. However, once you set the example for how organic search results with a structured interface are more attractive and useful than plain old text summary results, it's not too much of a stretch to visualize the same for paid search ads.

4. Google's completed acquisition of DoubleClick. As a story on BetaNews pointed out, up until now, Yahoo! and Microsoft have left Google in the dust with regard to display advertising. Now, with DoubleClick officially in the fold, analysts expect Google's display ad sales to be going up exponentially. That's going to put pressure on them to grow display ad opportunities — as well as to find synergies between their businesses that extend beyond being complementors in independent areas.

5. At last month's SES NY show — a gathering of professional search engine marketers — I was struck by how much mileage search marketers have squeezed out of a 25 character headline, two 35 character lines of text, and one 35 character display URL. People are running whole careers based around a canvas of 130 characters of plain text. Yet while I was impressed by how much nuance they've extracted from so little, it also seemed that the curve of innovation with text ads is nearing its upper bound. After listening to a panel of senior executives debate the use of the word "free" versus "complimentary" for 20 minutes, it's hard not to conclude that the bottom of the barrel is near.

6. Recognition of the value of brand marketing — including exposing an audience to your brand before they're ready for a direct interaction — is on the rise (again). Although search marketers and brand advertisers have come up with novel ways to engage brand development in search engine text ads, it's like trying to cut a steak with a spoon. Brand marketing may still represent a huge market of "non-consumption" for search engine advertising as it exists today. Incorporating images and video in search ads could be the catalyst for a flood of brand spending in this channel.

7. Entrepreneurial creativity continues to push the boundaries of alternate search engines. Under the shadow of the Google, Yahoo!, and Microsoft search mountains, a plethora of new search sites continue to blossom. Take a look at Alt Search Engines and their Top 100 list. Vertical specialization, social search, and new visual interfaces distinguish these upstarts. But innovation in the search experience also opens the door for innovation in search advertising. This is a Petri dish from which new search ad formats may likely evolve — and eventually be adopted by the giants.

Add these pieces together:

the search engine text ad S curve starting to level off;
strong financial incentives for unleashing the next S curve;
a cauldron of new search ad formats beginning to bubble;

...and it sure seems that conditions are right for a new S curve.

The implication for marketers — and marketing technologists in particular &mdash is to be ready, alert, and open-minded. When a new S curve surfaces in an industry, reaction speed and flexibility are competitive advantages. As the environment changes, it can be a valuable edge to experiment, learn, and adapt before your competitors do. This is what it means to view marketing as a science.

If image-based ads and/or video are predictably an integral part of this next wave, you can take a few steps to prepare that are either low-cost or valuable in their own right:

put your digital asset management house in order;
shore up your internal or external design resources;
start to talk with new vendors in the micro-video space;
don't let your search marketing processes get too rigid;
if purchasing or building search management software, make sure that you have contingency options in case significant changes are required quickly — or if even an alternate solution may need to be substituted over the next 12-24 months.

This will certainly offer a new round of upheaval for software vendors and agencies in the search marketing space. New entrepreneurs: take heed. And to the degree that this next S curve has more visual canvas than plain text ads, it may offer one more chance for traditional (or "tradigital") advertising agencies to play a leading role in the search marketing space.

S curves do look a little like roller coasters, don't they? Should be fun.

Marketing as a science, but science is a creative endeavor

2008-03-30T09:42:26-04:00

Let me state right up front: I am an advocate of scientific marketing. Just in case there's any doubt as you read on.

Marketing ideas should be tested and one should apply the scientific method to those tests.

Most of marketing is now measurable, directly or indirectly. With digitally produced and distributed marketing — particularly channels such as search engine marketing, online advertising, email marketing, post-click marketing, and web site optimization — it's often practical to test dozens or hundreds of ideas within a short time. So if you have more than one competing suggestion for "what will work best", the solution is "test them and let's find out". The answer is quantifiable, not emotional.

One of the missions of marketing technology leadership should be to provide the tools and the processes for an organization to approach marketing scientifically.

This has inspired more and more people to declare "marketing is a science, not an art".

However, that declaration always raises the hairs on the back of my neck — particularly when it is promoted by software vendors — because it is often used to imply that analytics dominates creative, that quantities are more valuable than qualities.

In my opinion, it's the inverse: scientific marketing is at its best the other way around, when creative harnesses analytics. A subtle but important shift in priority. The right fuzzy qualities have far more value than the wrong precise quantities.

True scientists understand this because science is a creative endeavor.

To be sure, with the scientific method, testing must be rigorous. But the process of determining what to test and how to measure it and framing the right hypothesis in the first place takes imagination.

Science is about discovery, and it's one of the most creative ventures in the universe.

In this way, the divide between science and art is largely artificial. Show a mathematician an elegant and original proof, and he is just as likely to find artistic beauty in it as any painting at a gallery. Vice versa, an artist who isn't continually experimenting in their medium is labeled commercial or derivative. The geniuses of science have more in common with the geniuses of art than either have with the hacks in their respective disciplines. (Enter Leonardo DaVinci stage left.)

That's why I love marketing as a science. It is both creative and analytic, and at the end of the day, it is about discovery. Just as the capacity to conduct more tests with more accuracy has been a boon in every scientific field, the new wave of marketing technologies is rocket fuel for proactive exploration of the marketing domain on an unprecedented scale.

What I don't agree with is the warping of "marketing as a science" to mean "marketing as accounting". This is what happens when the structure of tests and the categorization of customers — the underlying model — gets set in stone, and where all "testing" from that point forward is rigidly boxed into existing assumptions. When marketing becomes more about optimization than discovery, you end up with an organization on rails: streamlined perhaps, but it's only going where the track has been laid.

If you don't see the danger in this, read The Black Swan by Nassim Nicholas Taleb. Its examples are more from the financial industry, but the fallacies of "quants" in that domain are extremely relevant to the risks in analytical marketing that mistakes the map for the terrain.

A couple of years ago, I attended a lunch with the physicist and novelist Alan Lightman, who gave a talk on the power of bridging multiple disciplines. His main point was that the most important aspect of different fields and disciplines — chemistry, physics, biology, literature, sociology, architecture, etc. — wasn't so much their different bodies of knowledge, but rather their different "ways of thinking". Each brings a certain set of perspectives and approaches for solving problems.

The power of cross-pollination is unleashed by applying the perspectives of one to help solve a problem in another. In the scientific community, this has gained recognition as the genesis of many breakthrough solutions.

In the business world, it's known as "thinking outside the box" or lateral thinking. (Or sometimes just "huh?")

Lightman himself embodies such cross-discipline thinking. A theoretical physicist and a MIT professor, he made fundamental contributions to the theory of astrophysics under extreme temperatures and densities. (And I thought AdWords management was tricky.) Yet he also wrote the novel Einstein's Dreams, an international bestseller, and has contributed essays and short fiction to The Atlantic Monthly, Harper's Magazine, and The New York Times.

In his talk, Lightman made three points that every scientific marketer should take to heart:

1. The scientific method — start with a hypothesis, prepare an experiment to test a prediction of the hypothesis, measure the results, draw a conclusion, iterate — is based on the creative spark of the hypothesis. There is no cookbook recipe for robotically generating hypotheses. (Side note: the combinatorial probability of a room full of monkeys with typewriters eventually writing Hamlet may not be zero, but the span of time required exceeds the age of the universe.)

2. The process of discovery involves four stages: (a) mastering the craft of the domain, doing the homework to have a "prepared mind"; (b) getting stuck on a problem, what becomes the creative catalyst in your subconscious; (c) having a shift of perspective, a new viewpoint, which is where the power of different ways of thinking comes into play; and (d) finally having a breakthrough, a creative synthesis of disparate elements into a solution.

3. The irony of very sophisticated models is that they are almost always based on arbitrary assumptions. There's nothing wrong with that as long as those assumptions aren't mistaken for absolutes. Sometimes the most powerful tests are those that test the validity of the models one has been using for testing.

Marketing as a science is — or should be — a very creative endeavor. And that's pretty inspiring.

Semantic Marketing, SEO++ Feedback

2008-03-27T22:09:29-04:00

Thanks to everyone who's shared or responded to semantic marketing and the SEO++ idea. It's been great to connect with other people who are also intrigued by the possibilities that are starting to appear at the intersection of marketing and the semantic web.

Jennifer Zaino at SemanticWeb.com interviewed me and published an article To Market, To Market, The Semantic Way.

Richard MacManus at ReadWriteWeb awarded me Comment of the Day: Semantic Marketing for my remarks in response to Alex Iskold's terrific article Semantic Web Patterns: A Guide to Semantic Technologies.

Valeria Maltoni of Conversation Agent also gave a nod to SEO++ in her post on Web 3.0 Search Aligning with Data.

Kevin Newcomb at SearchEngineWatch shared a little SEO++ link love in his Highlights from the SEW Blog: March 17-21, 2008.

A brief exchange of comments with Paul Miller, who writes The Semantic Web blog on ZDNet, was the catalyst for putting these preliminary thoughts together.

Greatly appreciate the encouragement, feedback, and sharing of these thoughts. It's one of the more exciting areas of ferment in the (not too distant?) future of marketing technology, and I look forward to more discussions around it.

SEO + Semantic Web = SEO++

2008-03-23T18:10:34-04:00

Two weeks ago, Amit Kumar announced on the Yahoo! Search Blog that Yahoo! will be supporting semantic web standards in their new Yahoo! search open platform.

This is a game-changing moment in online marketing.

Essentially, Yahoo! is proposing that search will be the "killer app" for sparking the explosion of the semantic web into the mainstream. With this release, when Yahoo! recognizes semantically tagged data in the pages it crawls — structured data such as contacts, calendar events, reviews, blog/news feeds, etc. — it will present far more compelling summaries of those pages in its search results.

As their example shows with a LinkedIn profile, results no longer have to be bland snippets of text that all blur together on the page. Different types of results can pop, conveying far more information — and far more relevant information — to searchers. The quality and meaning of the data shines through, which reflects extremely well on the site represented by that result.

This creates a tremendous incentive for marketers to adopt semantic standards in their web sites.

According to a presentation at SES NY by Andrew Tomkins, Chief Scientist at Yahoo! Search, this semantic data will initially only effect the presentation of the results — officially it will have no impact on a listing's placement in search engine result pages (SERPs), to prevent this from becoming a way for people to game the system.

Over time, I believe this position will change as the search engines develop better ways to police good vs. bad semantic data, since great semantic results make the search engine look good. And even in the semantic web, people aren't going to page through more than the first couple of pages of results. Exposed semantic data adds to the quality of the results. (This is not dissimilar to how people have been leveraging YouTube videos and Flickr photos to appear in advantageous locations in Google's "universal search".)

In the meantime, however, there will still be a positive feedback loop to incentivize marketers: great SERP summaries that stand out with semantic data will no doubt win a higher click-through rate. The more people who click-through and view your site, the more of them will end up creating links back to your site. This in turn will increase your organic search engine rankings, and the virtuous circle continues.

Simply put by Kumar, the benefit will be "increased traffic quality and quantity".

What marketer doesn't want that?

But to get in on this new game, companies need to get their semantic web act together. Since that virtuous cycle favors first movers — who will clearly stand out more in the early days before everyone is doing this as a matter of course — there is value to moving fast.

Many companies already have search engine optimization (SEO) initiatives underway, either with their own internal web team or through an engagement with a search agency. It probably makes the most sense for those teams to lead the initial charge into semantic web optimization (SWO) because they're already familiar with process of tagging pages and subtly alternating the structure of the site to improve its search engine rankings.

However, the semantic web is qualitatively different than SEO. It requires a deeper understanding of the business being represented online — combined with a strong mastery of semantic web standards such as RDF and microformats — so as to strategically identify the best ways to position a company in the data web. That is going to require collaboration higher up the chain of the organization.

Sure, there will be some low hanging fruit in the early days, with reviews, blog feeds, and event calendars. But to go beyond that superficial level of semantifying will require companies to start to shape their business around producing competitively advantageous semantic data on the web — and where necessary, advocating in the semantic web community to shape the evolution of new semantic web standards.

Companies who know what they're doing here will be active participants in places such as microformats.org and the Yahoo! developer community, taking a proactive role in moving the semantic web forward in ways that benefits them and their customers. (Word of advice: don't jump into this fray half-cocked.)

This sort of new way of looking at marketing and the semantic web is what will be the core of semantic marketing. It's as strategic as it is tactical.

In short, this next generation of web optimization isn't your father's SEO. It's a significant enough advance that I think it would be great to have a different badge to identify it, so as to not fall into the lull of thinking this is just a linear extension of existing search engine diplomacy. I propose:

SEO + Semantic Web = SEO++

(Pronounced: so-plus-plus.) This is a nod to the sea-change that happened when software developers embraced object-oriented programming, moving from C to C++. This is an analogous paradigm shift that will require people to think differently about SEO. And after all, the semantic web is very much about representing and working with "objects" in web space.

Arguably, search engines have been the primary driver of Internet marketing. Yet for 10 years, the interface of search engines has pretty much stayed the same. Google's universal search was a step in the right direction, but this new move from Yahoo! promises to be a major leap. No doubt, it will be followed by Google and will inspire a spectacular round of search engine innovation — one that will present incredible new opportunities for technology-savvy marketers.

This is the wake-up call: marketing must understand the semantic web and start to think strategically about it.