Name drop commences:

Attendees included June Li, Maciek Adwent, Jenn Fong-Adwent, Mark Dykeman, Dave Hamel, Glinski, Alex Brasil, Mike Fiorillo, Jose Davilla, Romy Klaus (in from London!), Lida and Mike Sukmanowski, Gar, and the whole Syncapse Measurement Science team – among many others. I counted some 45 attendees at the apogee. Thank you all for making it an excellent evening.

Name drop ends.

There are no stated agendas at these gatherings, only hidden ones. And indeed, I had a chance to talk extensively with Mister Glinski (@glinskiii) about our recent transformations from being such extreme analysts to becoming rounder strategists. I’ve written in this space before about the combination of evidence and product development – specifically about using analytics to make product better. Last night, we expanded on that notion of going beyond a source of proof to becoming a collaborator in the formulation of strategy – to using that skillset to make better products from the outset. That’s a subtle bit of word play – but an important one.

We’ve identified how the path from execution excellence after-the-launch and strategic collaboration before the launch is an ill-charted one. Patrick cut out his own path and is succeeding. I have too.

Such stories will be worth sharing.

This was one of possibly 500 discussions that happened over the course of the night. There was genuine overlap. Thanks all, and let’s continue to talk to one another.

The invite:

“Developers make it possible to measure anything, statisticians and dataminers work models, IAs finesse interfaces, analysts mash and managers action. Effective Analytics takes an orchestra. Lets talk to each other and see whats possible.”

Historically, WAW’s attract a strong contingent of web analysts, social analysts (many from Syncapse), IA’s, a few dev’s, recruiters, vendors, and yes, two dataminers. And it’s a great mix. Let’s keep that mix and expand it. Additional invites to business strategists, eScientists, Marketing Scientists, and specialized developers.

As promised:

Link to the White Paper:

syncapse-sentiment-analysis

Link to the Data Set:

The Geurilla Analytics Project _ Sentiment

The paper will speak for itself.

We can discuss it here and on Twitter.

Time for a story. It’ll be fun and egregiously self-deprecating.

My first Real Time Simulation (RTS) game was Age of Empires I, back in 1998 or so. And I loved playing it online. Problem was – the online experience really sucked because most of the players were jerks. The experience sucked and the game lagged like hell.

By 2000 I had joined my first gaming community. They were referred to as gaming clans, and you could identify its members by having a telltale tag at the front of a name. MNPE_username, JCV_username…and so on. There was also this entire notion of being anonymous. It was highly prized. Back to this in a bit.

Gaming clans were never really intended to be part of the multiplayer gaming experience. At least, I don’t think Ensemble Studios ever really foresaw them. People, in a way, spontaneously formed them. The most dedicated ones would then register a domain name and host Internet forums. I don’t think we even knew the term social media at the time.

I regret not keeping a long enough list or better records – but there were many, many, many clans. The vast majority had very short half-lives. They would be founded, exist for 3 weeks, and then die a cold death. Some would persist. They’d grow and thrive. And then later, they’d fragment and explode. They died a hot death.

A very few would be self-sustaining. In effect, they’d be founded, they’d get 7 members, and from there they would be successful…continuously refreshing the membership over time and maintaining just the right size to insure familiarity and community.

Different clans had different utilities for their members. Some of them were Elite Clans. In effect, a small group of 7 to 15 people would get together and trade insider skills and knowledge. You had to have a certain rank to even apply to be part of them.

Other clans were friends only. You were invited in.

Some clans were open. Anybody could join. So long as you were fun to play with, you were pretty much guaranteed to be in.

Clans served a few purposes. There was a reason why so many people joined them back then.

For one – they provided a source of gaming quality control. The online experience sucked back in then – with people flamming and dropping, cheating and yelling. More often than not it was unpleasant. If you knew a group of people who behaved reliably well, then the experience of the game would be all the more better.

Secondly – they provided a source of quick games. If a group of people routinely logged in at the same time, chances are they’d play together and have a good game. The tag provided the boundary for friendship. Or, it guaranteed a certain level of competitive quality in team play – which was important for people’s ladders rankings and tournaments. Gaming was (and probably still is) serious business.

Thirdly – many evolved to form actual communities over time, and evolved a set of social norms and commonality. For a very long period of time, pre-Skype-Facebook-Twitter-YouTube-MySpace, they formed a type of safe-place. You could remain anonymous and still be part of a community. And it was as though somebody in Korea was just next door.

After a game had become stale – the community frequently remained.

You might ask – “Why didn’t people play with their friends?” (Like they do now – like on Xbox Live…?)

Not everybody had broadband connections at the time. It was still rare in countries other than Canada, South Korea, and Netherlands. Secondly, not many people played RTS games, either. For sure, some people knew other people ‘in real life’, but this was the exception, not the rule. Finally, Xbox live didn’t exist. Console games were consoles. Online games were social in a different way. The distinction, I suppose, in the transition from kids in front of Nintendo to young adults drinking beer in the basement playing a console game – and the notion of virtual community anytime/anywhere gaming: was a distinctive split. You could only get an online social experience through the PC. This unto itself caused a different dynamic.

You could have an authentic community experience through a clan while being anonymous. Many people back then wouldn’t imagine doing anything online unless it was through a pseudonym. (This was pre-Facebook).

Many of these communities gradually faded. For one, Ensemble Studios began incorporating clans directly into the gaming experience. While this was nice to a certain extent, they removed a few sources of quality control, and an important concept of continuity of the community in case the leader leaves. Ensemble Studios, much to their credit, actually did what we now refer to as ‘community outreach’. And they were actually quite proactive about it at the time.

Clans continue to thrive in the First Person Shooter (FPS) genre, where the quality of the game is heavily dependent on the quality of those you play with, and where dedicated servers are highly desirable. Sony, through it’s massive FPS game ‘MAG’, might still accidentally cause a renaissance of the early-era clan.

There are quite a few takeaways for social media.

People sort other people. Not always. But often. In the instance of clans, people sorted themselves into groups based on the experience they wanted but couldn’t get otherwise. I suppose if the parts of what remain of Ensemble wanted to innovate, they could introduce an jerk-sort into the experience. Those who have propensity to ruin an experience for others could be grouped together and be miserable with one another in random match-ups.

Communities are much more than joint-utility seeking entities. To see them only in that light would be too simplistic of a model. But communities do, at least, seek joint-utility. Even 4Chan has its bouts of concern about the cancer that is killing /b/.

Communities might also self-organize so as to compensate for a deficiency in a corporate offering. Perhaps some companies can understand which deficiencies exist.

Finally, communities can outlast their original purpose. For how long they can survive I’m unclear about. But they do.

Community Seeking and Online Gaming in the Early 2000’s were really interrelated. I’m curious to watch how this next wave of social gaming will change the landscape that much more.

This is an interesting line of thought Christopher, perhaps I can help with a bit of a framework. And you’re right, product is the root of Marketing decision making. I hope my attmept at a chart below makes it through the CMS without breaking…

Brand for any product is a continuum between Product-centric and Image-centric, example:

……….Product Centric………..Image Centric
Beer…….Sam Adams………………Budweiser

Image-Centric Brands tend to have commodity status, which begs the need to differentiate by creating some kind of unique Image. Product-Centric Brands differentiate on hard Features and Benefits.

If you think about the Marketing for Sam Adams, it’s all about ingredients and customization. If you think about the Marketing for Budweiser, it’s all about wanting to be like or associating yourself with the people or images in the spot – “Yea, that’s me!”.

Now, if you think about Social success stories, you find that they really gravitate towards Product stories, and not Image stories. Image stories are too easy to destroy in the social fabric; product stories bubble up *from* the social fabric.

So the success of social will largely be determined by where your Brand is on the continuum between Product-centric and Image-centric.

And here we arrive at a bit of irony.

Many of the most successful Social “Campaigns” happen when the company does absolutely nothing overt in the social space – see Apple, and many other Product-centric Brands.

And some of the lamest and most clueless Social campaigns have been from commodity Image-centric products that tried to do something overt in the social space – see various packaged goods.

Meaning, you don’t really have to *do anything* to get ROI from social if you have a successful Product-Centric Brand – the ROI is infinite because there is no spend. And the ROI for an Image-Centric Brand is likely infinitely negative – any spend will never generate enough incremental sales to pay for the spend.

As far as Marketing discplines go:

……….Product Centric………..Image Centric
……….Direct Marketing……….Mass Marketing

Direct has always been a Product-Centric approach; it has to be or the Math doesn’t work; it’s Feature / Benefit driven.

That’s not to say companies employing Direct do not have “Brands”, they most certainly do. But the Brand is very tightly tied to product, not so much with “me too” Imagery.

Make sense? Help in your quest?

This both makes sense and helps.

The nature of the product is certainly a dimension in this problem, and your model rings true.

So much of what we really consume is in our heads. Certainly, there’s often a physical aspect of most products. For certain products, it’s about how we feel about them. And to a certain extent, how we feel about a product is what we are and what our friends are like and how much our friends really influence ourselves. I can point to product diffusion studies in marketing science around artist popularity and DVD sales to back that up. The nature of the social graph is as important as the social impressionability of that social graph.

If you’ve listened to any commercial music since the advent of the Bit Torrent, you know that there isn’t really a qualitative explanation as to the variation in popularity of a given DVD. What we consume is sometimes, though not always, in our heads.

For a commoditized product that really isn’t differentiated based on the general attributes or quality of it – I can see where a company will want to compete in your head and for your friends. Of course, not everybody is as susceptible to social effects as others. Beer choice amongst friends versus beer choice amongst professional associates might be one.

I think we have to consider where a product is in terms of adoption is important too. Just because something on its own might have better features than another – it doesn’t guarantee successful diffusion.

I’ll need to think longer and harder whether or not the ROI curves asymptotic.

But, as of now, we have ourselves at least three dimensions.

Roger Martin, in “The Opposable Mind”, appears to de-emphasize the importance trade-offs.

Porter defines strategy is the process of making choices about activities that results in sustainable competitive advantage. Both books make reference to activity diagrams – so there’s unity and acknowledgement that choice matters. At the core: Porter explains the ‘why’ of strategic decision making, and Roger Martin describes the ‘how’ of strategic decision making. The ultimate way of showing trade-offs, in my view, is though the Production Possibility Frontier.

What is very elegant about the production possibility frontier (PPF) is that it’s two dimensional and tells a very clear story. There are trade offs between quality and quantity. Luxury and Economy. Bread and Guns. Reporting and Analysis. Accuracy and Precision. Very easy to understand.

Simplicity wins when you’re communicating. And yet, we’re rarely handed a manual when it comes to seeking elegant solutions.

Instead of just plotting two trade-offs and calling it a day, and instead of heading into n-space and Riemann spheres and unicorns and stickers – there is an elegant way of increasing salience while retaining understability.

If you use GGOBI or PASW, you can select ’scatterplot matrix’ and you’ll get a n x n chart of all the activity-relationships. So, if you have 10 activities with tradeoffs and functions interrelating them, you’ll get a 100 charts, of which, 45 will be of use. (45 are just the mirror image, and 10 are straight line functions).

Through the magic of the post-it note, you can expand the number of activities under consideration quite a bit, and then start sorting them out on a very large wall.

What you might end up finding are a number of strategies that are very close any number of PPF curves (at least, the ones that matter). From there you can derive activity maps that correspond to any number of selected PPF curves. The elegant solution would be plot it against a single PPF curve, but I’m pessimistic (at this point) that such a solution would be common enough.

Instead of a single PPF, what if  it’s more of a Library of PPF’s?

I’ve participated in some of that research over the years, and it’s always pretty enlightening.

It’s good news.

The Seven Axioms are:

1. The purpose of analytics is to derive competitive advantage for the organization / firm / entity.
2. Data alone does not yield competitive advantage.
3. A sequence of progressive hypothesis testing is the most efficient and effective method to derive competitive advantage from data.
4. Predicting the future requires an understanding of cause and effect.
5. Correlation is not always Causality.
6. Accuracy over Precision.
7. It is possible for there to be two optimal, equally true, answers to a problem. (And Sometimes More!) (X^2 = 4, x=-2, 2).

They might appear to be fairly straight-forward. And they are. In my opinion.

A statement like Accuracy over Precision was certain to cause problems. And it has.

If you look at the language around cause and effect, causality, and there being many correct right answers to the same problem: you get the point. It follows from the Axioms that, to derive competitive advantage, you need to be able to make predictions about the future, and the only way to really get there is through progressive hypothesis testing with accurate data, and understanding both complexity and causation.

It’s strong work, and quite possibly one of the best reading experiences I’ve had since “Reading Virtual Minds”.

On Page 149, there’s a quote in explaining the common occurrence for ‘fires’ to occur as a result of low-ball estimation:

Curt: Why do they have the fires?

Sam: Yes

Curt: There could be a million different reasons if you think about it, I mean, clients coming in with aggressive timelines period or everybody will come in with big dreams, right?…Like you never lose the champagne dream even if you’ve got a beer bottle budget, right? You always dream big but you might not be, like, okay…”

And I’m in awe.

What a gem.

And I ask myself: how can we optimize and predict dreams? How we do rationalize the denominator here?

What a fascinating business problem.

Of course, I shouldn’t really say simultaneously. I can only read one at a time. More accurate language would be ‘jumping between four books’.

The first is Sam Ladner’s excellent thesis on the commodification of time in the new economy. It’s a pretty awesome read.

The second is Gladwell’s latest book. And it’s a manageable read because the chapters are well contained. It’s called “What the dog saw”, and that line is pulled from one of the Chapters on Caesar Milan. Fun!

The third is a seminal 500 page book about competition. And it’s a sobering read.

And the fourth is about mental structures in the new economy. And I haven’t decided if I’m going to admit that I even read it.

So many at the same time. Sometimes I get to a point in a book where I literally can’t stomach it. It’s either so dense or so depressing or so wrong that I need to put it down and change the channel. Instead of popping open the web browser and heading over to 4Chan, I suppose it’s easier to flip over to another book. Naturally I’m putting off the gratification of completing something. But, so be it.

But at least there’s apple sauce. Apple sauce to wash down all that awful, awful medicine.

And Sam’s thesis is not medicine. I’m actually really enjoying it.