Defining success (or failure) for Google+

Literally thousands of commentators have posted their take on Google+ this week.

But I don’t want to join the peanut gallery passing detailed judgement on Google+ this early in the game. Clearly Google is committed to the endeavor and this is just Day One. As with any new initiative, I think it’s most important to define “success” and “tolerable failure” up-front.

Success case: feasible, yet unlikely to be achieved

It seems to me that a reasonable success case would see Google+ achieving results similar to what Microsoft has achieved thus far with Bing search. Through partnerships, Google+ can seek to raise their share of social network “engagement” to about 30% after investing a few billion dollars in the project.

To achieve this 30% market share, they’ll likely need to get about 60% of Facebook’s 500 million active users to also become active on Google+ on the assumption that users will initially devote less of their time to Google+ than Facebook. That’s a considerable uphill climb, although Google already knows a little bit about the “social graph” of each user via e-mail and other services.

Tolerable failure case: social products, without an engaging network

I think the most interesting aspect of Google+ is what the failure case looks like. Previous efforts like Buzz and Wave just dissapeared, or ended up in lawsuits. But I think the downside for Google+, given its deep integration with Google search, is that it just ends up becoming a wrapper for various Google’s non-core search engine services.

Already, we see signs of that with Google’s decision to retire Picasa and Blogger coinciding with the Google+ launch. In the future, I think it’s plausible that Google+ will be a UI from which to access Reader, Video Chat, Android and Shopping “apps.”

While, in this failure case, Google+ will fail to make any meaningful impact on Google’s share of social media attention, it will at least provide Google with enough social graph data to make its advertising targeting competitive with Facebook and provide an attractive API for developers to build upon.

The trillion-dollar question which I’d be trying to find the KPIs to track is: can a social product thrive independent of an engaging social network?

What kind of cookie are you?

There has been much buzz lately about the impact of behaviorally-targeted advertising on consumer’s privacy online. Noting that consumer’s online interests are now tracked and profiled by thousands of pixels and cookies, Senator John Kerry has proposed legislation to force websites to allow consumers to opt-out of tracking. This comes on-top of the FTC’s own “Do Not Track” program and the self-regulation efforts of the Internet Advertising Bureau and the Better Business Bureau.

Most of these efforts are motivated by a desire to retain the potential benefits of targeted advertising while averting any chance of an Orwellian scenario where the mindset of every citizen can be predicted from online data collected without their explicit consent by websites and traded on exchanges like BlueKai and eXelate.

However, I’m not sure if anyone who is presently involved in regulating or self-regulating online advertising has ever tried to answer a question which I just set out to answer: how good is the actual data quality within advertising exchanges?

Data quality in behavioral advertising has significant implications because it may very well be that the most harm to consumers could come from advertisers “knowing” a lot of false facts about consumers and targeting advertising against them. For example, if BlueKai thinks that I’m a Male, 35-44 who lives in a home valued at $100-200K and drives a Domestic SUV, then I’ll see a number of advertisements for Lowe’s and Ford that are almost completely irrelevant to my, actual, Manhattan lifestyle. Meanwhile, a company whose products I actually care about, such as American Express, may use this information to determine NOT to provide me with their best possible terms and promotions. In this scenario, a customer fails to benefit from receiving properly targeted advertising while being harmed by the possibility that information about them will be mis-used.

Thanks to Evidon’s “Online Profile Manager” I was able to catch a glimpse into the information kept about me by a number of major online advertising brokers including BlueKai. The information provided captures my membership in a number of “audience interest groups” which advertisers can target banner ads against. For example, using the data below Capital One might see that I’m a Consultant (i.e. frequent traveler) who is interested in Credit Cards and Airline Reward Miles, so they would send me ads for their Ventures reward card:

Here’s what I learned about data quality on BlueKai:

• BlueKai tracks 285 “facts” about me and my interests… but most of the topics are innocuous: BlueKai tracks my interests in shopping and travel, propensity to purchase a variety of products: from life insurance to cold medication to tissue paper and demographics at the state, age and income bracket level.
• BlueKai is correct about my interests just 35% of the time
• The erroneous information provides a pretty lousy profile of me: they think I watch a TV show called “One Tree Hill” which I’ve never heard of. Apparently it’s popular with teens. Yet, they believe me to be a Male age 35-44 who lives in a home valued at less than $100K (is that possible in New York?). While they recognize that I live in New York City area, they also believe that I live in a “Average to Low Cost of Living” area. I’m simultaneously a member of both the “High Net Worth” and “Lowest Wealth Decile” groups and I am included in the interest group who shops at “The Gap” and “TGI Friday” but can’t recall visiting either chain in the last 13 years.
• Profiles miss most of the important traits which actually define me: I was raised in Ohio but support the Red Sox; I love learning, revere Enlightenment values and studied history at Cornell and business at Chicago; I speak four languages; I don’t use Facebook much but I share interests like dining, world issues and technology innovation with a few close friends. I recently entered my thirties and will be getting married soon so am beginning to take pride in “growing older” while also worrying about the responsibilities that entails. If I could receive a behaviorally targeted ad for something, I’d like it to involve learning to drive race cars or buying something nice for my parents…

In conclusion, this little experiment has been enough to show me that the data available about me on BlueKai hardly instills any Orwellian fear. The central problem is that I generally wouldn’t recognize myself from the profile presented on BlueKai. This means that if I want properly targeted advertising, I’m not getting it.. but if I want to stay off the radar, I may be better off staying “opted in” to BlueKai’s scrambled data — letting them think they are reaching a a potential Chevy truck buyer who also shops for M.A.C. eye liner — than “opting out” of it.

Diseconomies of scale in data services?

Much well-deserved attention has been given recently to “Big Data” and “Utility Computing” as transformational technology developments.

However, the recent data breaches at Sony and Epsilon have now exposed personally identifiable information from hundreds of millions of consumers. Billion-dollar brands have been defaced and this necessitates understanding the drivers of risk in “Big Data” and “Utility Computing”

Were these breaches due to budget-scarce, conflict-averse corporate IT? Or are they the result of inherant risks in “Big Data” and “Utility Computing?”

That the worst breach in history occurred at a media and marketing-centric company like Sony rather than a technology company like EMC might seem to support the “blame corporate IT” response. But Epsilon poses a challenge to this explanation. Founded in 1996, Epsilon can be thought of as one of the original “cloud computing” service providers. They send over 40 billion e-mails annually on behalf of 2200 clients. This is a very highly scalable model in which technology should be central. Clearly size and smarts didn’t keep Epsilon from failing at security.

What are cloud computing providers and purchasers to do?

First off, it’s important to recognize the advantages of cloud computing well-done:

Major cloud computing providers combine two scalable inputs, hardware and talent, in ways that corporations big or small can’t match. Talent is particularly hard to match because the most talented engineers will naturally want to work on the most scalable problems in the least bureaucratic environments. This ups the premium that a large company, like Chase, will need to pay to attract competitive talent and thus increases the necessity of running more services “in the cloud.”

Secondly, cloud computing providers enable businesses to do more by doing less. The failure of Lehman Brothers, and the abject failure of executives to see it coming, helps illustrate the fact that managerial attention suffers from huge “diseconomies of scale.” As businesses get more complex, they increasingly need “Big Data in the Cloud” to do two valuable jobs: enhance visibility giving executives a “sixth sense” while removing the burden of executing the complex work to make it happen.

With that said, achieving economies of scale without jeopardizing brands requires cloud computing companies and clients to have frank conversation about a few key factors:

• How stable is the underlying technology? Utilities are only as stable and dependable as the technologies which underlie them. Anyone evaluating Epsilon as a provider should have been worried to discover that not a single technology executive sits on their Executive Steering Committee.
• How important is the asset you are outsourcing?  This is a Catch-22. On the one hand, a business cannot sustainably outsource the key drivers of its competitive advantage. If you could do that then, axiomatically, so could competitors and you would have no advantage left. On the other hand, the practice of outsourcing “non-core” assets begs the question of how much you care about the stewardship of those assets. It’s up to a company’s CEO and Board to make decisions about what is really “core” but the reality is that many outsourcing decisions are made locally without this kind of thinking.
• What is the impact of a catastrophic failure? To evaluate the risks of a complex system, it’s important to isolate whether a failure will take down the entire system or just a component of it. How you choose to share data with a cloud services provider has a lot to do with mitigating or intensifying the costs of failure. For example, companies who out-sourced their entire customer lists to Epsilon must be kicking themselves when they realize that only a small fraction of customers are e-mailed on any given day. If they had only shared data with Epsilon on a “need to know” basis through something like a RESTful API, they might have limited the impact of the breach to simply the number of customers they e-mail in a day rather than in a lifetime.
• How much is actually budgeted for security services specific to your install? Many clients expect their cloud service providers to implement security as part of the overall application “stack” and cloud services marketing plays along by insisting that security shouldn’t be a worry. But these data breaches clearly show that it should be. Expecting to get optimal security for a low, monthly, bundled price is a bit like hoping that your landlord will bundle optimal security into your office cost per sq. ft. Price and security are competing objectives where price is the feature that gets you business and which a client notices monthly, while security is the one that they rarely notice. Clients need to recognize this and work with cloud service providers to create custom, client-specific security procedures and audits that are budgeted independent of the core product.
• Do you know how Big Data improves customer satisfaction? Customers seem to care very little about privacy under normal circumstances, but they care a lot about it when something goes wrong. Many companies have responded to this by seeking ways to educate customers about how they data is used. While this isn’t a bad idea, I think the approach is likely to fall upon deaf ears. The better approach is to audit uses of customer data to ensure that they are “on brand” and being used for customer’s own benefit. When disaster strikes, companies like Amazon and Netflix will be better positioned to recover because customers already believe that they have derived great benefit from their use of Big Data. But companies like Bank of America may be a bit hard pressed to explain how the end customer benefited from BofA saving a couple of million bucks by outsourcing an e-mail list to Epsilon…
• Can a “Big Data” service provider steal your customers? It’s important not to narrowly concentrate on security risk. Sharing your data also provides others with the ability to analyze and use it to better serve your customers. In most cases, the service provider has a shared interest in selling this insight back to you on fair terms, but they may also sell it to your competitors. Chances are that some of their personnel will be hired by competitors in the future for that very purpose.Management consulting firms have lived with this dynamic for decades. Clients hire them and provide secrets, knowing that secrets will be closely guarded. Yet, they also hire these firms precisely because they have worked for their competitors and can provide “best practices” examples… in other words, code for Other People’s Processes. A high degree of trust and interaction between consulting firms and their clients helps assuage fear, but cloud computing providers have thus far avoided having to invest in consultative relationships in the same way.

  My suggestion to clients would be to talk with your cloud computing vendors and ask for specific consulting investments and points-of-contact which ensure that people you can trust have a genuine stake in your success. If you’re going to ask them to take stewardship over part of your business, then you need to know who you can hold accountable and feel confident that the right incentives are in place for a win-win partnership.

Why quality doesn’t sell itself…

It’s become slightly trendy of late to decree that we live in a post-marketing era. Investor Fred Wilson remarked lately that “marketing is for companies with sucky products” and that idea has some intuitive appeal to it. After all, aren’t good products likely to generate word-of-mouth and thus free promotion in this age of social media?

But the real world is filled with counter-examples to the idea that products sell themselves. Washington Post reported Gene Weingarten provided one of the most poignant examples of late when he had world-famous violin soloist  Joshua Bell perform disguised as a busker in a  Washington DC Metro station to see if anybody would notice.

Stripped of the endorsement of the world famous concert halls where he plays for more than $100 per seat, Bell barely attracted an audience. Out of the thousands who passed by, only a handful stopped to listen and just one person recognized him and his total collection for his 42 minute concert was $32.73.

We can think of other examples:

• Could you spot a painting from a great artist if it was hanging at the neighborhood diner, rather than the Metropolitan Museum of Art?
• I fell asleep watching Citizen Kane. Would I ever have bothered to remember it if it wasn’t widely acclaimed to be the greatest movie of all time?
• How do you feel about the quality of the university that you attended? Let’s assume you think it’s excellent. Now, would you rather have gone to Harvard?If you agree that you’d rather go to Harvard, then you ought to ask whether your answer, rather shocking when you consider the risk of giving up an excellent experience,  is due to any certain knowledge of Harvard’s quality or rather, the influence of marketing.

Therefore, I think rather than adopt a blanket statement that “marketing is only for sucky products” that we ought to perform a more careful evaluation of under what conditions that statement could be true, and which it could be false.

As I’ve thought about this, I think the key test is to understand to whether the product produces customer satisfaction by virtue its specifications, by its immediate outcomes or by its long-term outcomes.Let’s define three cases separately:

• Products judged by their specifications:These products own characteristics are “self-sufficient.” Foods sometimes fall into this category because we have a pretty good idea what spaghetti & meatballs ought to taste like. If a restaurant has good spaghetti, it shouldn’t need to market itself because the taste profile will instantly match our mental model of a good solution pretty well. If it has bad spaghetti, well that’s Olive Garden.
• Products judged by their immediate outcomes:Business strategist Clay Christensen has popularized the notion that customers typically “hire” products to do a specific job — they don’t purchase the product because of its inherent characteristics. To illustrate, Christensen tells the story of a food fast company who sought to improve the taste of its milkshake to appeal to more customers. But better taste, it turned out, didn’t influence sales because the job which consumers imagined for milkshakes wasn’t to delight their taste buds, it was to stem the boredom of a long commute. Thus non-taste attributes like convenience and thickness were more important.While it takes some effort to tease out what job a product is being asked to perform, the performance of this type of product can at least be assessed immediately.
• Products judged by long-term outcomes:Many products require a sustained period of use, a supporting ecosystem and social approval before they pay dividends for the individual user. Take Microsoft Excel as an example. I recall that when I was a wee teen, I opened the program, found a blank worksheet and wondered what the fuss was all about. I suspect that about 90% of Microsoft Office users still fall into the same category. But for those who work in an environment where they regularly engage in data-driven conversations with their peers, have abundance data to work with and are afforded opportunities for repeated trial-and-error, Microsoft Excel has an invaluable tool in the long-term. 

While this may simplify the problem somewhat, I’d suggest that “marketing is for sucky products” is true when it’s known that products can be judged according to their specifications. But indeterminable ex ante (before the consumer trial) when products can be judged according to their immediate outcomes. And false when it comes to products whose outcomes can only be assessed after some period of time and investment.

For example, consider Xobni, a program that enhances Microsoft Outlook e-mail. Every consumer knows what e-mail is as a product, but there are pretty distinct “jobs” that consumers put it towards. Some consumers are using e-mail to keep in touch with a select few family & friends, others are using it for business but even those scenarios vary considerable. The mid-level manager in a company likely uses e-mail quite differently than its outbound sales force does. Due to the initial press, Xobni received a great number of downloads (including myself) but a supermajority of these were uninstalled within days (including my own). As a result, Xobni has struggled to define its market niche despite $30 million in funding.

The Xobni story isn’t all that rare. It nearly happened to Twitter.

Circa 2009, the Internet was full of stories about how users were unsure what to do with Twitter even though they had heard about it. I published a blog post offering some suggestions to help overcome Twitter’s adoption issues and, since then, Twitter has recovered quite well. The answer for them wasn’t to dramatically improve the product but to market it. Most improvements in Twitter user experience came through the adoption of the platform by celebrities and leading brands and by the development of third-party software like TweetDeck to enable individuals to personalize their Twitter experience to suit the “job” they want out of it.

Giving something back: what companies could learn from Amazon

When I recently made the move from Chicago back to New York, I decided to do what I once thought was sacrosanct: selling much of my book collection. If you were once a liberal arts major like me then you can understand how eventually your shelf gets stacked two deep with books you’ll never read again.

I’m convinced that knowledge is always better spread than horded and I can use the cash so I sold my books on Amazon using their EasySell service where you simply ship the books to them by cheap USPS Media Mail and they do the rest including shipping the book to the customer.  The basic result, I’ve made a boatload of money and been able to sell my books in a flash. Here’s my sales report:

That’s over $2,000 from selling about 200 used books, most of which are the definition of long-tail. For instance, I recently sold The Crusades: Islamic Perspectives for $19.99 to a college professor from California.

Of course, Amazon’s making a little coin from their commissions. But the most lasting impact for me is that I’m now an even more delighted customer who wants to talk in detail with others about their service. I’m blogging about Amazon now and talking to my friends all the time about it.

I’m convinced from this experience that Amazon is a company which actually helps me do more of what I enjoy. They recognize that my experience with books doesn’t end when I sign their receipt. As a book lover, I want to be able to read more books so Amazon’s solution helps me to free space on my shelves and in my bank account to buy new ones. Amazon ultimately profits from these transactions, but most importantly enriches the depth of their relationship with me which makes them my book provider of choice. The power reciprocity is profound.

This brings me to my question, what can other companies learn from this?

- Wireless: what if Sprint offered a marketplace where customers could buy and sell their phones and plans. Somebody who wanted to join Sprint could swap their Verizon plan and phone for a Sprint one knowing that they could easily reverse the trade if it doesn’t work out for them. A Sprint customer who is sure that they’re locked into a plan they don’t need could sell it to somebody who wants it and buy a new one.

Does this fit the current lock-in strategy of cell phone providers? Of course not. But this program isn’t incompatible with 2-year contracts either. It simply removes the painful burden of calculating 2-years ahead from the consumer, delighting them and enabling the cell phone companies to profit from their superior information. For a company like Sprint which is advertising customer centricity even as it bleeds customers, this is one way to prove it without re-hashing the  ”our network is the best for everyone” story which Verizon already owns.

- Health Insurers: it’s well known that some people get more coverage from their employers than they need and others don’t get enough. What if United Healthcare enabled you to trade your policy for another within their system? Second, taking a cue from Amazon, what if the health insurers thought of the end goal of their customer as providing health not policies? Amazon realizes that people like to consume the information in books more than they like to horde them. Those who want to consume more books will naturally turn to Amazon because of the liquid marketplace and information which exists within it.

Do any health insurers think this way today? No. They compete to win corporate accounts, letting the end consumer wither on the vine. Meanwhile, the customer perceives that the health companies are doing a great job at selling policies that don’t provide the service of health. They clamor for national health care. This might change if they made a serious commitment to providing the health service rather than just talking about it. A great example is the  Health & Wellness at United Healthcare where you’ll find that you have to have already bought the product to gain access to whatever scarce bits of information are inside. If United Healthcare wants to show that it cares about health, it should offer content from its own databases and third parties like WebMD to uniquely integrated health content with the health policy products it offers. Consumers could shop for policies by starting with information on an important topic, like joint health, and then working backwards into the solutions that United provides.

- Clothing Retailers: people who love brands like JCrew aren’t comfortable visiting the Goodwill shops which often end up housing them or sorting through the stuffed close-out racks. Fashion assortment and merchandising are so complex that it doesn’t pay for Amazon to keep a standard catalog of JCrew outfits. But JCrew has this information and audience themselves, why don’t they use it? The existence of a liquid used market for recent JCrew styles would help many of its devoted fans to afford to shop there more often, buy more of the latest styles each season and feel less worry about taking a chance on buying  something new.

Is this compatible with today’s romantic branding notions of a semi-upscale soft goods brand? No, but I wouldn’t rule it out until we hear what today’s consumers think. Millions of customers on Amazon have no problem differentiating between buying used and new merchandise and love Amazon for giving them the choice. They buy with confidence on Amazon knowing that they can always sell back a movie. This requires a very different, customer-centric way of thinking, but it’s the type of radical idea that has the potential to delight customers who will could just start to think of themselves as fashion merchants rather than victims.

Chipotle does brand right

In a recent post, I gave credit to MyChipotle.com for its excellent execution of a so-called “brand campaign” online. The campaign launched in May and features print, bus stop, taxi top and online spots in select cities like Chicago and Dallas which direct customers to:

This campaign works well for several reasons:

1. Illustration: the first thing a viewer sees actually shows the food. It’s appetizing. Looks fresh. Showing good looking food should be a basic pre-requisite for all fast food advertising. Otherwise you get the Burger King debacle from Crispin, Porter which has won plenty of awards but sold almost no burgers as sales and market share at the “The King” have dropped.

2. Headline: it’s simple and bright enough to be seen walking by the bus stop or as tax cabs pass by.  This principal should be applied to web ads which are glanced at for, at best, 1 second. Here’s an example of a terrible ad which lacks a headline, has tiny print and black reverse copy which is especially difficult for the aging eyes of the older target audience to read:

3. Call to action: it’s actually very well executed and based upon a consumer insight. If you visit MyChipotle.com you’ll find a site which highlights all 60,000 great combinations  that you could make with Chipotle and includes some colorful user generated videos highlighting their favorite burrito recipes. It’s fun, informative and gets you thinking about trying to come up with a new flavor combo yourself. It also gives you a higher sense that being a regular at Chipotle makes you a member of a community.

This campaign is based upon a customer insight gathered from research: consumers don’t feel like Chipotle offers enough variety relative to other fast food competitors. This is despite the fact that Chipotle actually offers almost infinite flavor combinations with every dish fresh made. But most customers actually order the same thing every time.

That said, the campaign does fall short on a few dimensions:

1. Advertising-dependent: if the problem is variety and consumer boredom, Chipotle should also be looking at its product mix, menu design and service process. Perhaps it’s time to introduce a new product or sauce now and then at Chipotle? Chipotle’s menu design resembles a flow where you “pick one” from each of the wrap, meat, filling and sauces categories. But this leads some consumers to feel like a rule is being imposed upon them when it’s not: you can actually pick two meats if you like. Finally, the service process itself could adapt a bit to this. Chipotle could train its employees to prompt stumped customers with friendly suggestions.

2. Lack of follow-through in the store. There’s no tie-in between MyChipotle.com and the stores. You can’t find the nearest Chipotle from the MyChipotle.com site. Nor can you one-click order your favorite user-generated recipes by using Chipotle’s web ordering tool. And finally, no sight of the most innovative implementation which would be to stream MyChipotle.com content into the stores to be enjoyed by customers in the notoriously long lines.

The lack of foll0w-through likely results from an abundance of marketing silos within retailers. The problem is that none of these silos call themselves marketing, but they are crucial to it. The group that does service process doesn’t say that it’s engaged in marketing, but it touches the customer more than anybody else in the company. The people who design the menus and order the signs “aren’t in marketing” but consumers spend more time looking at those than any advertisement. Finally, yet another group manages “technology” and is likely waiting for its marching orders from marketing. But alas, those might not come because the general problem is these companies fail to recognize is that marketing is something that everybody in the company needs to be worried about. In other words, marketing is too important to be left only to the marketers.

That said, Chipotle and its agency Butler, Shine, Stern and Partners seem to be running great campaign which pushes down the barriers between brand and response, bricks and clicks.

Can Twitter be the pulse of the people… and their celebrities?

In a recent post on Twitter, I speculated on their business model by working backwards from their design priorities. Now, TechCrunch has leaked the internal Twitter documents which shed a lot more light on what’s going on in Twitter’s head.

I noted that Twitter seems to want content on the network above all and is obsessively keeping up a hip image. Thus it has de-emphasized search, following, monetization and partnerships. But when it comes time to monetize, Twitter will be torn between becoming a mass messaging platform and a tool for elite influencers to broadcast to the masses.

Looking through these documents, we see a strategic morass where Twitter is really hoping to become both a mass platform and a hub for influencers but these are likely incompatible goals:

• Twitter’s goal is to become “the pulse of the planet” with 1 billion users. Doing this requires being “open to scale” on all devices using the “lowest common denominator” techologies like SMS.
• But Twitter has no idea how to monetize this vision. $1 per user is a purported goal, but it sounds like a total shot in the dark akin to the “if we just get 1% of the market” revenue estimates found in bad business plans. The reality, as Twitters management realizes, is that “most users are not monetizable” and this is made even worse by having an open SMS-like distribution. Becoming the pulse of the planet further dilutes the possibilities: can Twitter monetize 50 million users from Brazil at $1 per user? Unlikely.
• Thus Twitter speaks at great length about monetizing corporations, celebrities and pursuing distribution deals with Google, Microsoft, etc. but they can’t bring themselves to do anything other than “verified celebrity accounts.” Twitter’s management views these as a way to get some revenue “without mobilizing the whole company around something” but that’s troublesome. It means they’re going with a tactic that is incompatible with their strategy. Servicing celebrities, ask anyone in Hollywood, is completely unlike dealing with regular users. More requests will come in from the likes of Diddy who feel that Twitter should be paying THEM for their contributions to the network. If there’s competition from better monetized sources like Google or Facebook, that just may happen.

A final issue which Twitter management seems to be struggling with is “brand”. For instance, it rejects acquisition by Facebook and partnerships with Microsoft and Google largely based upon how its brand might be perceived or developed. These brand concerns are short-hand, I think, for many issues like how do you keep early adopters happy while bringing the masses on-board, how do you keep direct relationships with users while supporting an abundance of third-party apps and how do you fend off future competitors?

The key to finding a successful brand will be making Twitter stand for an aspiration not a feature. Being the “pulse of the planet” is a feature, and indeed one that is hopelessly complicated. How can one reduce 6 billion heartbeats in multiple languages to a true pulse? Look at the case of MySpace whose CEO recently lamented that MySpace users “don’t know if we’re a social portal, a music site or an entertainment hub.” That may be right but the greater problem is that users don’t know why they should come to MySpace instead of Facebook, PEPSI instead of Coke. This is another reason why these online brands need to develop some revenue: you can’t create brands for 1 billion people just by changing “status” to “what’s happening?” You need advertising, people and a lot of expense.

Here’s my first take at an aspiration for Twitter that would emphasize the focus on celebrities: Be Somebody.

Twitter lets you hobnob with celebrities and be the cool kid that people look up to even if they don’t really know them. Where Facebook is about reflecting your static social graph, Twitter is about growing it. The business model for that implies that influencers like celebrities should be treated with kid gloves and seen as the driver of adoption. In reality, they are. Most of the 1 billion users are going to join Twitter to follow the celebrities they already know.

The users who want to become influencers, who aren’t yet celebrities, should be charged for the opportunity to promote themselves on a powerful network. These charges can be as simple as $20/user for tools and tips that help them use Twitter more effectively but they can also include charges for directly embedding video in Twitter posts, etc. There should also be an option to pay for placement in searches by topic: for instance, people who want to find “Advertising” influencers.

Finally, Twitter should provide consulting and analytic services. It’s remarkable that companies are paying consultants hundreds of thousands of dollars to train them in how to use Twitter effectively but Twitter corporate itself provides no services – not even to those consultants.

Similarly, an abundance of third-parties are building businesses which help corporations, recruiters, stock analysts and so forth to find information on Twitter. Twitter should acquire and manage many of these services for the reason that supporting them will require development by Twitter of data structures and APIs to suit them. For example, in recruiting, data structures need to be built which allow resumes and applications to be linked to job hunters and referral networks on Twitter. These changes are not inexpensive to make, market and maintain. It behooves Twitter to try to capture some value from the work it does here.

What does all of this require? More management, more focus (which will upset some employees). Tough things for a company to embrace as it grows from 50 to 500 employees but absolutely essential given Twitter’s ambitious goals.

Will online advertising ever get brand analytics right?

Online advertising’s standardized metrics and ad formats may be the slowest developing web technology of the last 15 years. Doubleclick was founded in 1994 and over that time, ads have tuned their targeting, become visually richer and interactive and, of course, been embedded into search results. But overall, not much has changed in advertising’s strategy, styling or approach.

Compare this with the web where the pace of disruption has been much faster. Technologies like AJAX and Flash have made sites more interactive, application APIs have made them more open, CDN’s have made them much faster and cheaper. In general, the web has become more social, open and interactive. This fast pace of innovation is one reason why the top websites of 1997 like Geocities and Excite are unrecognizable today.

Conservatism would be warranted if online advertising was living up to its potential, but it’s not. One statistic is that 2/3rds of the $180B spent on advertising fits into the “branding” category but just $7 billion of is spent online. Those who have attempted to defend the status quo haven’t been doing a good job of it.  Take this recent study by the Online Publishers Association which touts the success of brand advertising:

Even a cursory look at this study shows that it’s little more than evangelism under the guise of science. Let’s look at a few slides:

Slide 11: purporting to show that people exposed to brand campaigns visit the advertiser’s site most often:

Diagnosis: this slide is meaningless because it only shows the results of the experimental group. An experimental finding is only interesting if the results of an experimental group, exposed to advertising, were different than control group who was not. But they don’t show the control group here. Since the “experimental sample” in this study is 86 million, it seems like the control group is amazingly small.

Slide 17: purports to show that users exposed to ads spent significantly more on e-commerce

Diagnosis: aha, we have a control group with with little information about it.

So we should be asking some question like:

Does the methodology make any sense? I’d say no. The chart indicates that the statistic represents spending on all of the 53 brands sites if the customer was exposed to any of their ads. So, a visitor might have been exposed to the Allstate campaign but have spent nothing on Allstate.com and they will still be counted in the “exposed” sample. Chances are if you look at this data at an advertiser-specific level, you get much less convincing data. Yet this is exactly what you want to do since obviously the quality of execution and a brand’s product should make a great deal of difference. You can’t just invent a brand, throw up a lousy ad, take the customer to an uninteresting site and expect to get something out of the advertising.

Is the control sample otherwise identical to the experimental? If the groups were targeted ads because they are of different incomes or have different online behaviors, that’s the cause of their differential e-commerce spending not the ads themselves.

Are the results statistically significant? When I see a study that says “significant” without the statistical before it, I wonder if they’re trying to talk around something. The sample size is fine, but we need to know the standard deviation of consumer purchases to determine whether $242 is “significantly different” than $227 statistically speaking.

Do the numbers make any sense? The average individual in this study spends about $234 per month on e-commerce. That’s $2800 per year. Multiplied across the 220MM US Internet population, that’s a market size of $617 Billion or 4.5% of total US GDP. E-Commerce is not nearly that big of a deal, it’s probably about $150 Billion.

Now don’t get me wrong, I’m a huge advocate for online advertising, analytics and branding. I just think that using data mining and Powerpoints to prove your point isn’t going to do anybody much good.

The work that people like David Blum of Butler, Shine, Stern and Partners have done for Chipotle on the MyChipotle.com campaign presents a much better example of how to meld brand with response, online and offline to delight customers. It’s integrated marketing with a mission that stands out to anyone who sees it. It’s the product of the unity of creativity and technology with a call-to-action that’s right on brand.

Is it an online branding success? I’ll speak to how I might evaluate that later. One thing’s for certain: any analysis should map to basic marketing principles and avoid generalizations based entirely upon undifferentiated panels like ComScore which only track online behavior like looks, clicks and buys.

Can “paid content” save journalism?

The hype cycle on “paid content” as a way to save journalism is picking up steam. Gordon Crovitz and his Online Journalism start-up are touting that their technology platform will enable 10% of readers to pay for content. The Atlantic has micro-payments up as their “Idea of the Day” and the Dow Jones is purportedly developing its own platform for paid content.

But the problems with paid content online, as I see it, are manifold:

First, paid content is historically a modest percentage of print media profitability so it’s not the “culprit”. For newspapers and magazines, circulation revenues are about 35% of the total and we can presume that they contribute almost zero profit since the objective until recently of both was to maximize readership.

Second, consumers don’t pay much for printed media. According to the Census 2007 Survey of Consumer Expenditures, the average consumer spends a mere $118 per year on Reading products, including books, and this data is very socioeconomically skewed. Most consumers spend less than $50/yr. except for those over $75K in income who spend $200-$300. Table 4 of the same also shows that persons under 44 spend half as much money as older people on reading. Journalism Online, the aforementioned paid start-up, conducted a survey in which it was estimated that the average subscriber would pay $300 per year for their service. This datapoint is flatly contradicted by the Census’ own findings.

Third, following Becker and Posner’s excellent analysis of this same question I’d point out that the decline of readership of newspapers has followed the increasing proliferation of information in our society. The more information we encounter in our daily lives, the faster news spreads and opinions can be heard and shared, the less we need to wait to hear them from a journalist. Of course, some have called this “information overload” and argued there’s a role for journalists and technology to help with this. Surely there is, but that doesn’t mean that solving this problem will point consumers towards professional journalism content. In many cases, the work of journalists is being increasingly done by subject-matter experts who publish blogs online. Whether the subject is what’s happening on your local block or what’s happening in the economy, experts who don’t get paid primarily for their writing are able to answer these questions far more effectively than journalists. Journalists have been disintermediated.

Finally, this brings us to the fundamental problem with paid content: too many substitutes. If information is all around us and most of it is not very valuable (after all can you remember what you read about in yesterday’s paper?), then why do we need to pay for it?

While education correlates with spending on reading materials, our increasing levels of education as a society bode poorly for news/magazines. The average newspaper story is written at the 8th grade reading level. I think that’s one reason why news analysis looks increasingly shoddy to us: we’re more educated than ever. The NY Times taking its content back behind closed doors won’t change this dynamic: they can’t impact the price of a commodity that borders on free. And if they did, new entrants would expand supply and push it back down. That’s exactly what Politico, PoliticsDaily, HuffPost, TMZ, Gawker, the Daily Beast and so forth do quite successfully because they aren’t burdened with the overhead of major news organizations.

There are a few conditions where paid content works:

• You serve a relatively small group of people. With a small group, say less than 200,000, the potential advertising market is very small so entirely ad-supported market entrants are less likely. Of course, as advertising becomes more efficient and production costs decrease, market entry becomes easier. This helps explain why local newspapers are still a profitable business in small towns, but in New York City the free subway papers are the lowest common denominator of American newspaper publishing.
• The information you provide is proprietary and really important to your audience. As in, they couldn’t live or work without it. Bloomberg was once a good example of this sort of business but it’s already seen its nadir. Reuters entered with a solution that’s a fraction of the price. The information is no longer proprietary and its arguably less important when a) everybody else has it b) automation and computer algorithms.
• You can solve information access and organization problems. In other words, you help people access structured data, archived data and hard-to-find aggregated data. This is what Lexis-Nexis, Westlaw and so forth do for lawyers. This is what CapitalIQ and FactSet do for bankers. Note this is different from what “Internet aggregators” like Google or TechMeme.com do because they typically highlight easy-to-find information that’s current and unstructured. They are great services, but not solving this problem.
• You sell to businesses: the expense can be written off against taxes and amortized as part of the infrastructure of a large, complex organization. This is important because it creates cultural support and switching costs. In contrast, the consumer bears the full cost of their expenditures and switch providers on an annual basis.

None of these conditions apply to the current content of magazine and newspaper publishers. I do think it’s plausible that they can morph some of their content to solve deeper problems. The Economist has, for instance, opened up an expert’s bureau that offers high-priced subscription products, studies and consulting. But profound changes would be needed for say Better Homes & Gardens to provide an information service of that magnitude.

I believe that in this lies more of the solution for publishers than “printed content.” With print readership declining, companies need to restructure to cut costs and focus on what’s of most value to readers. The core of most media will be advertising supported because consumers have too many substitutes to pay much for it, but if you have experts, sources and data then you might be able to carve out a small, focused audience for paid products. If you leverage your local brand and make good partnerships, you might be a useful distribution channel for other providers like classifieds, employment and personals.  But as much as I’d like to see it, the economics of “paid content” don’t make much sense to me.

Why marketing is broken: the SEO/SEM craze

I suspect that marketing is broken and the craze for search engine optimization and management is a good example. Search for “SEO” on a job site like TheLadders.com and you’ll find hundreds of positions paying 100-250K for someone to “define and lead a SEM strategy.”

But SEM, search engine management, isn’t strategic. It’s essential but completely tactical. Strategic issues require creative and analytic thinking to synthesize ambiguous data on internal and external factors in a very dynamic game. Intuition, which is developed only through experience, education and diligent practice, is essential to strategy.

But you can learn 95% of what you need to know about SEM from reading two webpages. Or you can pay $200-$1000 for an online seminar from the Sempo Institute. The most valuable source of info is a panel of 37 expert’s aggregated wisdom on the importance of 200+ factors to Google’s results. The study is of 2005 but it’s unlikely to change dramatically. The key factors are:

1. Having the search keyword appear in the Title of the page
2. Having the search keyword appear in the URL of each page
3. Having lots of other popular sites link to your site
4. Having some category-specific sites and communities link to your site

With that set, you need to understand what keywords people are searching most often. Your best sources for this are your own log files/analytics program followed by the search engines themselves, as Yahoo or AdWords will provide estimates of traffic per search. Finally, you can also consult WordTracker.

There is, of course, plenty of work involved in implementing SEM: keywords need to be bought, page titles need to be set-up and so forth. There are very dedicated professionals who excel at doing it as well as software tools which can speed the process. But thinking of SEM as the heart of “online marketing” is misguided. Take some learnings from the leaders. I’ve heard about one company which is likely the world’s #1 search engine advertiser. If anyone has incentives to do things “on the margin” to improve search engine placement it is them. But they hire more product managers annually than all of the folks they have on their entire SEM/SEO team. The reason for their success is that they are in love with their customers, and vice-versa.

This takes us back to why marketing is broken: no amount of advertising can fix a bad product, marketed to the wrong customer who is being served poorly. Worse, SEO/SEM doesn’t even offer the control of advertising. Google controls the search engine and how it ranks you. Your customers and influencers are the ones deciding to promote you and increase your link popularity. Random traffic can’t take you very far and that’s one reason why Amazon will actually show advertisements for competitors to in-bound traffic that comes from Google.

I’ve said that “marketing is broken” but by that, I mean the practice of it. There’s nothing broken about the theory of marketing if you go to a source like Kotler who defines it as the “science and art of finding, retaining and growing profitable customers.” This view of marketing is powerful because it links it to businesses key objective: profits. But the real world often thinks of marketing as “promotion” after all the other P’s have been defined: product, price and placement.

This gives marketing an image problem. One of the best studies of this is provided by Professors Sheth, Sisodia and Barbulescu in the compilation “Does Marketing Need Reform?” They ask colleagues of marketers from other departments like finance and operations to evaluate their performance and the results are indicting: only 38% rate their marketing colleagues as good or excellent and only 34% view marketers as strategic thinkers. To address this, marketing needs to start its reformation by making its objective start with providing benefits to customers. Making SEO central to strategic marketing efforts is probably missing the point of it all.