Dirty secret: “Phase 2″ almost never happens.

Since Saaspire started, we explored what turned out to be a very undefined space.  I was shocked to discover that not only were there not decent technology solutions available, but on top of that there was very little product design work being done in this area.  In fact, there wasn’t really even a good language for designers to articulate personalization behavior.  (The typical solution, wire-framing a million states in an application, was kind of like sketching out a movie one frame at a time).  I even discovered a fundamental organizational problem: companies were simply playing catch-up to the latest Internet technology: mobile, social etc.  They had no way of creating a communications strategy that could flex with new innovations, and thus were constantly in crisis.

However, it has come to light that the aspects of this new landscape that I want to explore are different from my partners.  Consequently, it has been mutually decided that I will spin off from Saaspire, creating a new company that will focus on the issues that I find the most compelling: this new world of Conversational Architecture that has emerged.

Axilent

I’d like to introduce my new company: Axilent.

Axilent’s initial offering is the Axilent platform: a foundation upon which one can easily create “relevancy” features: personalization, related content, product recommendations and contextual advertising.  By reducing the cost of implementing these features, I intend to move them into Phase 1 of the project, and to show that the sky’s the limit when it comes to designing a new generation of dynamically personalized, cross-channel, conversational Internet services.

The Axilent platform is now in private beta.  If you’re interested in participating, please send me an email at loren@axilent.com.

To demonstrate the platform, I’ve re-launched Whiskey Engine.  Now entirely focussed on whiskey recommendations, it’s an example of how easy it is to create product recommendations.  I’ve also put up a post on the new Axilent blog, explaining how Whiskey Engine works.

I am really excited about this – I think there is an amazing opportunity to create the next generation of web sites, mobile apps, social network integrations and other concepts based on innovations that we haven’t even made yet.   Axilent will be there to make that creation easier.

Somewhere in the middle of college I started to get the feeling I was being ripped off.

Admittedly, I am the canary in the coal mine over this kind of thing, but having reached the point where one starts to contemplate career choices (for which I had been prepared by my philosophy major, har, har) I found everything that could be construed as a standard choice a kind of raw deal.

Not to be all Karl Marx about it, but I started to feel that the educational system and awaiting employment market was trying to shape me into a cookie-cutter worker-bot, ready to take my place in the cubicles of the world, and do stupid meaningless things, never really capturing any value for myself, regardless of effort.  By value I don’t necessarily mean money; rather the degree to which your situation changes to move you towards the goals that are important to you.  For many people and most institutions this often does mean money.

In time I put a name to this phenomenon: optimization.  It’s so vast and all-pervasive that it’s difficult to even talk about it without sounding like Captain Obvious.  Essentially, it breaks down to two crucial ideas:

1. Institutions optimize to capture value for themselves.
2. Therefore, any “standard deal” offered by such an institution has been optimized towards their benefit, and away from yours.

It’s easiest for me to discuss this in career terms, although I believe it’s considerably more far reaching than that.  Typically institutions (companies, governments, schools and so forth) present individuals options that require much effort from those individuals, but provide limited upside in return.

In my grandfather’s generation he could at least get a tradeoff concerning stability: he could take a job and expect to keep it throughout his lifetime, receiving a pension when he retired.  Then came institutional optimization.  Corporations discovered they could improve their bottom line if they simply normalized a behavior that was previously considered anathema: the regularly occurring layoff.  Now routine and barely newsworthy, it is a regular occurrence to hear of companies laying off huge numbers of workers at the slightest economic contraction.  Even in boom times benefits are cut, and jobs are eliminated through re-organization, outsourcing, automation and so forth.  As individuals it seems naive at best to assume we can count on capturing any real sustainable value from a job.  Jobs are not engineered to deliver value to those who hold them.

Of course, optimization can face resistance occasionally.  In the area of software development, despite some extremely powerful and sustained efforts to commoditize (and thus reduce the cost of) development, software developers have stubbornly retained high value and are considered a critical resource.  This shows in their salaries, benefits and work environments.  Developers have captured value.  However they are the exception; just walk into any big chain pharmacy to see the opposite situation.  There you’ll find miserable, interchangeable, valueless workers, slowly being replaced by automated checkout machines.

This treads on the political.  Depending on your political leanings you may or may not think this trend is a good thing, and you may or may not care too much about it at a societal level.  However, my point isn’t that institutional optimization is right or wrong, it’s that it is predictable, natural and inevitable.  Institutions will always optimize value towards themselves, and the optimizations will always be most pronounced when the power relationship with the other party strongly favors the institution. Usually this means that it manifests where institutions intersect with individuals, because institutions are usually more powerful than individuals.  Think of you versus the cable company.

And that is what leads me to the second part of the idea: the Standard Deal.  The Standard Deal exists in so many aspects of our lives: it is the job we hold, the cable package we subscribe to, the political party we vote for, and the news we consume.  Little consideration is required in our adoption of the Standard Deal, because it has already been structured for us.  We sign on the dotted line for the same package as our neighbor.

The problem with the Standard Deal is that it’s optimized to take value away from you.  That’s why it’s a standard.  Following the natural process of institutional optimization, any situation in your life that has been worked out by some institution is specifically designed to move value away from you, and towards that institution.

The Only Winning Move is Not To Play

The Standard Deal is your enemy.  The only way to capture value about things that are important to you is to get out of the deal you’ve been handed, and forge your own path.

Now this (like most things) is a tradeoff, and getting out of the Standard Deal has a definite cost.  Specifically, you are taking on more risk, you are setting the barre for accomplishment higher for yourself, and in absence of the roadmap that comes with the Standard Deal, you will spend a lot more time figuring out what the hell you are doing.

Additionally, because of the pervasiveness of the Standard Deal, you will have to pick your battles.  For example, I certainly don’t need to “have it my way” when it comes to selecting socks.  The impact of stocking selection to me is so minimal that I’m happy to accept the Standard Deal in this regard.  (However the opposite is true for a neighboring issue: “time spent shopping for socks”.   Because I place a very high value on my own time I’ve switched to manpacks.com.)

For me the issues that really burn have to do with how I spend my time, and with my lifestyle.  The way I like these things is certainly not for everyone, but they are highly significant to me.  I view time as my only real resource, and a finite one at that.  Activities that feel like a waste of that time (bad jobs, too many unpleasant obligations etc) are unjustifiable in my mind.

Now that puts me at odds with the Standard Deal in many cases.  For example, I absolutely hate commuting, so I’ve made arrangements so that I can walk to work.  I spend about an hour each day walking, giving me the benefit of exercise and some home/work separation, but without the mandatory time sink of a commute.  Additionally I’m obsessive about autonomy, so I’ve attempted to push my career towards ever-increasing autonomy whenever I’ve been able.

As a result I probably spend more time than average trying to figure out what to do next, and wondering whether I can keep things going.  This is an acceptable tradeoff to me.

Your deal is probably different.  Perhaps you want to maximize the time you spend with your family.  Maybe you want to earn as much cash as you possible can for the next decade.  Or maybe you want to work on fascinating and challenging problems.  For each goal there is a path out there that you can cut through the jungle; a path that will optimize things so you can achieve that goal.

However, no one will ever hand it to you on a platter.  None of the Standard Deals being offered will, on their own, let you accomplish your goals, because that is not how they have been optimized.  Instead, you have to pick your battles, reset the situation, and come up with your own way of doing things at critical points in your life.  Then you need to have the courage and fortitude to see it to fruition.  With that your life can be optimized for you.

Come the industrial revolution however, there was suddenly an issue of scale introduced into the equation.  Suddenly instead of a handful of customers, an enterprise might have hundreds, thousands or, in the modern world, millions and billions.  One of the first and most fundamental actions of the industrial revolution was to cut personalization out at its heart.  That’s because personalization didn’t scale.  The factory needed to get thousands of units manufactured on their assembly lines, and couldn’t stop to deal with pesky user requests like “colors-other-than-black” on their Model Ts.

Then cometh the Internet.  With its oh-so-seductive lack of physicality and supposed promise of infinite supply (and the associated supposed-zero-incremental-customization cost) there was suddenly a burst of activity around creating customized experiences for individuals on the web.  Personalization was back.  All sales leads would convert and prosper.

Wait a second, though.  Go look for some books on personalization.  You might notice something interesting, as I did when I went looking.  Nothing has really been published about personalization since about 2002 or so.  That’s kind of funny.  Wonder why that is?  Almost like…something went wrong…

Black Boxes

New technology always debuts as a black box that promises to solve all of our problems.  Generally speaking, the people signing the checks don’t care to understand how it works, only that it’s going to bequeath competitive advantage on them (in the case of early adopters) or because all of their peers are signing on to it (in the case of the majority).  Also, sadly, the role of technology has been treated in a very arms-length manner within much of the business world.  That’s too bad, because it has some of the greatest impact on business.

In the case of personalization, engagements typically went like this: you fed all of your data into it (products, articles etc) and let it perform some sort of proprietary analysis on that data.  Then you dropped some recommendation widgets onto your site, clicked your heels together three times, and suddenly you were looking at theoretically relevant content and/or personalized product recommendations.

Over time, however there have been concerns by people who have deployed such systems.  In a nutshell those concerns are:

• How do we know our recommendations are working?
• How do make sure the black box personalization system doesn’t do something weird or inappropriate?
• How can we use personalization to distinguish ourselves from our competition?

In the world of black box personalization, there haven’t really been good answers to these questions, because you’re dealing with a black box.  You don’t know how the system is really working, there aren’t any controls or visible workings that you can have mastery over and it’s difficult to gain insight on how things work, and how your business strategy could be incorporated into the personalization system’s execution.

I believe these combined factors took a lot of the sheen off of the face of personalization.  Business operators, unable to really get any idea about how these systems worked, decided they weren’t that happy about delegating such a fundamental aspect of their marketing to an automated system that provided no means of understanding its workings.  From there the only practical choice was to go back to manual curation.  And that, as we know, doesn’t scale.

This problem is, I believe, at the heart of why I’ve had trouble finding any writing about personalization that has been published within the last 5 years.  I have a hard time believing that the technology behind personalization has come to a halt, but the business application of it seems to have hit a brick wall.

Splitting the Dilemna

So faced between a choice between the scaling problems of manual curation, and the unpredictability and the unknown factor that comes with black box personalization, business hasn’t been left with a great choice.  Generally speaking, it seems like businesses have mostly come down on the side of manual curation, and have put personalization into a ghetto on their website (a recommendations sidebar or such).  Again, that’s too bad, because there’s so much being left on the table that personalization could address, but without more control and insight into the process, business are unable to trust it.

What would be nice is a mechanism for businesses to be able to define the principles, or rules, under which they want to extend personalized service, and then from those rules deploy automation to execute that service.  A kind of  ”white box” personalization, if you will.  This is where Conversational Architecture comes in.

I propose that we treat personalization a little differently than we have in the past.  Using Conversational Architectural terms, business can define how their personalization system should respond to particular cases that are important to the business, in the way the business prefers.  Because the business is essentially defining rules, however, and not individual cases, the system can scale through automation.

Tuning the Content Mix

Conversational Architecture thinks of product recommendations, or related or personalized content, as all part of the same problem.  We call these “content policy” objects, and treat them as components that can be included in various contexts.  A content policy object is a placeholder for a set of business rules that describes how a component should be populated with content (content meaning articles, products, profiles or whatever).  The point, however, is that the mix of content within these policy objects can be tuned by the website operator.

Here’s what kind of content you can get in a content policy object:

• Personalized:  Based on the user’s expressed affinities, personalized content tends to be related (via metadata) to content that the user has endorsed in some way.  For example, if we’re talking about products, then personalized products would be related, (via metadata, tagging or collaborative filtering style aggregation), to products in which the user has expressed an interest in some way, such as by buying them, adding them to a wishlist or so forth.
• Related:  Related content is similar to personalized content, but instead of user affinities, the starting point is something on the page that serves as a base content.  In the case of e-commerce, the starting point could be the product featured on a product page, or perhaps a particular metadata or tag value represented by a product category page.
• Curated:  Sometimes you just want some specific content to show up in a particular place.  This is curation, meaning the individual content objects are specified directly.
• Random: With all this content targeting, it’s occasionally nice to dial in a little bit of noise.  Random content could be anything, and allows the policy editor to back off a little from complete content targeting, allowing for the possibility of serendipitous discovery by users.  You never know.

How Content is Related

Another important idea with regards to personalized and related content is the idea of vectors of relevancy.   This has to do with the specific metadata that is used to relate content to each other, and the relative importance of similar values within that metadata.  Not all characteristics of content are equally important when tuning a content policy.

For example when we built Whiskey Engine (whiskey-engine.com), we consulted with an expert who explained to us the elements of the taste of whiskey.  He broke the taste of whiskey into elements like “herbs/spice”, “flowers” and “smoothness”.  Based on his tastings, we accumulated data about the taste profiles of each whiskey in his collection, with a value between 1 and 5 (1 meaning the least, 5 the most) for each taste characteristic.

However when we spoke with him further, it turned out that people tended to place a premium on the “smoothness” factor above all else.  In other words, once people decide how smooth they like their whiskey, they tend to follow that vector when looking for new whiskies.  People place far less concern on other whiskey taste values such as “herbs/spice”.

For a different example, think about music: if you like the artist who recorded one track, there are reasonably good odds that you’ll like another track by the same artist.  On the other hand, the record company that release the recording is much less significant.  People usually don’t care (and don’t know) which record company released a particular recording.

Buy carefully defining the metadata for content as well as specifying its importance as a vector of relevancy, recommendations and related content can be improved within policy objects.

Gaining Confidence: Measuring Content Policy Performance

So how can you know if your personalization strategy is working?  By treating the content policy object as the head end of a conversion funnel, and measuring how well it performs (we call this “Confidence“).  At its simplest, you should simply be able to see how well each content policy object is performing.  If you’re not satisfied with the results, then go tweak your content policy mix, or try altering the relevancy vectors of the content metadata to increase the quality of the personalized and related content in the policy.  Specific personalization deployments should never be treated as “set and forget”.  You need to work on them, adapting them to your specific audience.

Of course the ultimate version of responding to confidence measurements would be for the content policy object to reconfigure dynamically based on performance.  Conversational architecture has a name for this too – we call it “conductivity“.  Essentially we measure the performance of recommendations and weight the effectiveness of the current vectors of relevancy.  Over time we’ll start to adjust the vectors (“oh look, it turns out that herbs/spice is a bigger deal than we thought”) to increase the performance of the content policy objects.

What about Modality?

In the original conversational architecture article I talked a fair amount about modal websites.  How does modality apply to personalization then?  I think that first one should realize that while you don’t have to use modality to build a personalized website using conversational architecture concepts, modality can provide you with a way of describe some special features that could take your personalized website over the top.

The basic guideline is: when moving away from ideas that can be expressed easily in terms of content policy objects – modality can be very helpful.  For example, imagine an e-commerce website that provided discounts for people who frequently shared products from the site through social networking channels like Facebook and Twitter.  In Conversational Architecture terms, you could identify this kind of person as a Persona (call them “social sharer” or something) and trigger a mode from that persona that caused a discount to appear in their shopping cart.  This is an easily expressed business rule that could then be implemented through automation on the site.

Wrap Up

So we introduced five new concepts that Conversational Architecture provides that allow businesses to more clearly articulate what they want from personalization: content policy objects that provide customized mixes of content, products etc., vectors of relevancy that determine the relative importance of content metadata when calculating recommendations, confidence which expresses the performance of content policy objects in terms of conversion, conductivity which expresses the evolution of vectors of relevancy over time, and finally how modality can be applied to a personalization problem.

This is obviously a lot of new ideas, however it’s entirely possible to just try them out one at a time and look for incremental improvement in your personalization implementation.  By using these concepts, you have acquired a language in which to discuss personalization, which will allow businesses to more closely align personalization with their specific approaches to operating their business.

On one hand, the hackers/programmers were acutely aware they were sitting on top of a bunch of potentially useful technology for the publishing world. Semantic web, collective intelligence, mobile apps, location-aware, social network integration and so forth. There are now many, many tools in the swiss army knife of technology with which one could build a new-generation media empire.

On the other side of the fence, we found the business people we encountered quite open and interested in the new technologies. Everywhere we’ve been, we’ve encountered attitudes ranging from “oh yeah, I’ve heard of that – I should really learn more about that” to “yes it’s awesome. I’m really interested in getting on that train!”.

What we have not found, however, is a bridge between these two worlds.  When we’ve started to speak with user experience, visual design and product development people, they seem at a loss as how to incorporate these technologies into their products and leverage them to their advantage.  There is no common design language or methodology for making sense of the whole collective intelligence world in a comprehensive manner.

Before we go any further though, let’s go camping.

The Idea

Let’s pretend for a moment we have a clothing retailer whom we’ll call Rugged Clothes.  They want a complete coordinated digital marketing strategy.  Here’s what it might look like:

The Smith family goes camping.  During the camping trip, the kids wear some new clothing their parents have bought from Rugged.  The parents take pictures of their kids with the Rugged iPhone app, uploading the pictures.

When they get home, the mother opens up the Rugged app on her iPad.  Because they’ve recently uploaded the photos, the app automatically opens up in photo editing mode.  Mom goes through the photos, picking out the best ones, adding comments to them.  Then she elects to “publish” them.

The photos are published to the Smith family page on Facebook via the Rugged Facebook app.  Friends and family can access them.  The photos are automatically tagged with the articles of clothing that appear in them.  Clicking on them will take users to a special category page on the Rugged website based on the Smith family.  From their the individual pieces of clothing can be purchased.

This story tracks the experience of users across four different media (iPhone, iPad, Facebook and the website) but describes a single, coherent experience that is aware of the current state of it’s users.   This is the kind of story that catches the attention of the more visionary business people these days.  It’s the promise of the collective intelligence technologies, and of the integration of mobile, social and web services.

The Page: A Bootstrapping Metaphor

We’re in a period of time with collective intelligence technologies analogous to the early days of film, a century ago.  When motion picture technology first came on the scene, people simply leveraged the methodology of a previous medium, theater, and filmed it.  It was several years before they started to realize they had a completely separate medium on their hand, and started to experiment with film editing (montage) and moving the camera during a shot (tracking and panning).

Similarly, in the web world, we’ve had the page.  The concept of the web page swiftly became an incredibly convenient metaphor for designers in the early days of the commercial web.  It allowed people who had a background in print design to make the jump into web, because they already knew how to lay out a page.  (Put aside the endless problems created by designers who assume web pages work like print, it was actually a net advantage: it bootstrapped web design).  By framing design decisions in the metaphor of a page, and a website in the terms of a “collection of pages”, we had the foundation to structure the question of how to build in this new medium.

Unfortunately, all that is breaking down now.  The page metaphor is becoming increasingly strained and less relevant in our modern world.  Consider the following:

• Mobile applications on multiple platforms
• Highly dynamic AJAX/DHTML/HTML5-style websites
• Social Networking Platforms
• Location aware services
• Collective Intelligence / Semantic Web technology

None of this stuff has much to do with pages.  But without a design language or metaphor to fall back on, a chasm emerges between business people who can see the potential of these technologies and are willing to fund the right projects, and the technology folks who stand ready to build this stuff if only someone could let them know what, exactly, they should be building.  There is no way to capture this stuff simply by discussing “pages”.   It’s time to put that wireframe down, and step back.

Brave New World

Let’s refrain, for a moment, from discussing the specifics of any individual new medium (web, mobile, social etc) for a minute and try and consider the big picture organizational communications.  There are three basic characteristics to which we should aspire in our communications strategy:

• Multi-Channel: As is largely conventional wisdom now, just having a web page, or just having a Twitter account etc., is usually not sufficient.  Different media have different mechanics and areas in which they are effective, and the best approach would be a comprehensive communications strategy that takes advantage of the strengths of each platform and leverages them in such a way that makes the most sense for the organization.
• Multi-Modal: This is an important concept: people’s interactions with organizations are modal.  Often they are driven by some purpose, specific or not, held by the individual.  One of the biggest design challenges on the Internet has been to try and present what are essentially modeless designs (e.g. “good for everybody, all the time”) that are actually used in very specific, modal ways.  A great communications approach would be multi-modal, rather than a “one size fits all” mode.
• Multi-Directional: Communications is really a two-way street.  While it’s one thing to have a touchy-feely marketing message in which you claim to listen to your customers, actually implementing it in a quality way at scale is extremely challenging.  Nonetheless, an organization that can actually respond to feedback and requests from individuals is at an incredible advantage.

So what metaphor can we use to pull together these qualities?  What exists in the natural world that’s a good fit for this?

Enter The Conversation

Conversations have obviously been around forever, which conveniently means that most people have something of an intuitive grasp of what they’re about.  Looking  at our criteria above, we can see that this metaphor maps nicely onto what we’re trying to accomplish.  Conversations can traverse various media (multi-channel) can shift modes depending on various action of the partipants (multi-modal) and involve two or more parties both listening and speaking (multi-directional).

Conversations should be used as the foundational design metaphor, at the point after when the initial concept for online communications has been proposed, but before specific user interfaces are designed.

This let’s you know what to build.  By modeling the conversation (or conversations) your organization is having with the outside world, you will be able to shape your online communications strategy in a way that is targeted towards specific audiences, over specific channels.  More targeted communication means less noise, and the more fruitful the conversation will be.

Conversation Design: How to Do It

So how exactly does one design a conversation?  Let’s break it down into steps:

Conversation Mapping

The first step is conversation mapping, or essentially to determine what conversations exist between participants.  This is a high level, strategic activity, that creates some shape to the universe into which specific design thought can be injected.   The first step is to identify the participants.  Here’s an example from my company, Saaspire:

Saaspire itself sits in the middle as the reactor.  This isn’t meant to indicate that it’s passive in any sense, but because it’s the participant that we actually control, any automation we build will live there, so the term “reactor” is accurate as far as describing the process.  All around are the main constituencies with which Saaspire communicates: customers, investors, developers and press.  Those are the “actors”.

Now we need to establish the conversations we’re having with each participant.

Basically what we’re saying is that Saaspire is having four different conversations with different participants.  For customers our communications tend to be about product support and education – similar to this is the developer conversation in which our communications are more technical and platform oriented, but again about documentation and support.

To investors we speak about the value and potential of the company itself, and finally to the press we try and make an attractive “next big thing” story.

Identify User Contexts

So once the existence of a conversation is established, how do we gain some insight on how it functions?  First we need to look at the driving forces of conversation modality, which I call context.  Context is an aggregation of the various factors about a participant that, in combination, drives the conversation from one mode to another.

Context consists of four factors:

• Personas are behavioral aggregates of participants.  They represent any long-lived group of user behavior that’s worth addressing en masse.  Traditionally in web design, personas have been expressed in demographic and (high level) motivational terms (“Cindy is 36 and wants to get things done fast.”).  While we’re less concerned with the demographic aspect of personas in this case, it can still useful to think of them in motivational terms: How do they think?  What do they want?
• Affinity represents the stuff that people like.  It might refer to content or advertising on a media site, or it might instead refer to products on an e-commerce site.
• A state or goal represents a temporary condition in which a participant exists.  For example, a user that has started a checkout process on an e-commerce site could be said to be in a specific state that will conclude with finalizing checkout.  The main difference between states and personas is the temporary nature of states.
• Environment is a catch-all meaning the circumstances under which the conversation takes place.  In digital terms, it tends to refer to the browser, the operating system and the form factor of the device used for access, but can also be broadened to include concepts like location.

This leads up to a very important concept in conversation design:  contexts trigger modes.

Mode Mapping

Once you’ve identified the possible personas, affinities, states/goals and environments that you’re going to support, the next step in conversation design is to determine your response.  This is done by having specific combinations, or contexts, trigger modes.  Modes are your response to that context.  For example, within our customer support/education conversation, we might identify the following mode:

In this particular case, we are looking at participants that we’ve tagged in two specific ways: first, they are considered part of the “qualified customer” persona.  This might have been established in any number of ways, such as requiring them to log in or otherwise establishing that they hold at least once license for one of our products, or it may just be some much softer form of self-identification on their part.  Secondly, they have exhibited behavior (perhaps a search on our site, or in inbound link from a specific Google search) that let’s us know that they have the immediate goal of seeking information.

Given these two criteria (and we don’t care about their affinity or environment in this case) we trigger “customer support mode”.  Within customer support mode, we might provide facilities on our website that are slightly (or substantially) different from what other users might see.

So What’s A Mode, Exactly?

A mode is a building block for your web service.  They are the states that your web service passes through for individual users as they are triggered by those users’ contexts.  In different modes, your web service might contain different functional components, variations in user interface, different content and so forth.  The question of what to vary between modes will be one of the foundational skills in this approach to design.

For example, you could use different modes to optimize for the way the user likes to interact with the site.  Perhaps certain users tend to search for information within your website whereas others are more browsers (they use the navigation system).  The site could adjust modes to emphasize the elements of the UI (search vs navigation) that most suit those users.

Modal design could emphasize the state in which a user exists at a given point.  For example, consider the “offline buying decision” in e-commerce.  A visitor goes to an e-commerce website and browses around, looking at various wares.  Then he leaves the site, and while he’s away from the site, decides to purchases something he’s seen.  At that point he goes back to the website and immediately purchases the item.

Most websites don’t know what to do with this behavior.  They see the first session as a conversion failure, and the second session as a success without any explanation.  But a modal site would recognize this as a state shift for the same profile, and the site could optimize for the appropriate state (browsing vs buying).

If the site suspected it had an offline purchase decision maker (a Persona), it could switch between “browse mode” and “buy mode”, based on the inferred State/Goal of the individual user.  In browse mode it would be always showing more options to the user, up-selling, suggesting more items and generally just extending the engagement between the user and the site.

If the person came back to the site, and immediately added the item to their shopping cart (something they had looked at before) the site would switch modes, into “buy” mode.  In this mode, the objective of the site is to let the customer check out as fast as possible.  No more distractions, no upsell, no additional options.  The user is now in buy mode – just let them buy.

More to Come

So that’s the quick introduction to Conversational Architecture.  I’ll be drilling down into more of this in future posts.  Please let me know what you think – I’d love to discuss this.

The fact is that on an average day most people who have careers don’t really think about them.  They think about what they’re doing at work that day.  They think about how many days are between them and the weekend.  They certainly don’t think about the fact that the things they’re doing (or not doing) now may affect them profoundly in a year, two years, five years and so on.

About 20 years ago I was an unemployed rock musician on the verge of dropping out of college – with no degree, no career and no plan beyond “being a rock star”.  I lived in a mouse and cockroach infested apartment in Toronto, that shared a bathroom with the apartment across the hall.  The other tenant was an unstable man on medication that threatened to kill me in the hall one day.  Life has gotten better for me since then.  :)

Over the next two decades, I developed two careers – first as an audio engineer, then moving on to programming and later technical management.  Over that time I learned that there’s an art to managing the career path itself, not just in doing the work.

The friends with whom I’ve been talking seem confused and frustrated with their careers.  They’re looking for answers to how to manage them, but find objections with any ideas that I present.  So I thought I would present the “big picture” as I see it – the fundamental principles of technical career management that drive success and failure.  Maybe within the big picture specific strategies for people’s careers can be found.

Two notes first:  One, I speak not only as someone who’s been an employee, but also someone who has hired probably about 30 people over the last 5 years.  I have a good idea about what makes a successful employee, and other experienced managers will agree with me on at least some of these points.  Second, for some reason I’m seemingly out on the end of the axis as far as risk-tolerance is concerned.  A lot of what seems acceptable risk to me is apparently intolerable to most folks.  I think it’s to their detriment, but it’s worth pointing out I don’t hold the “center” opinion on this matter.

Okay then, without further ado:

Your Career Is A Moving Target

Understand that every opportunity you take, and every opportunity you pass up in your career has consequences not only for the present, but also for the future.  When you decide to learn a new technology (or not), or try your hand at managing others, learn about UX and so forth, you are changing your career prospects three years from now.

The most important idea here, in this entire blog post, is that you need to look at all career decisions (and inaction is a decision too) in both the sense of how it affects you in the present, but also how it will affect your future prospects.  You have to actively manage your career. Hopefully you can make choices that are beneficial to both present and future, but sometimes it’s a little harder.  Sometimes, it means you have to make conscious decisions rather than just drifting, which brings us to the next point…

Inertia Will Kill You

While making active decisions to learn new things, change jobs, expand your options and so forth, will not guarantee success, I can promise you that consistently passing up on these things will absolutely put a bullet through your prospects.  Inertia, the act of just continuing to do what you’re already doing, is deeply insidious.  The longer you let it go on, the harder it is to overcome.  Unfortunately there is no way out but to overcome it.

I was speaking with someone once and told them that I felt like I was stuck in a rut.  They said something that I’ve always thought was deeply insightful:  ”Ruts can be comfortable.”  The terrible, insidious thing about inertia is that it can be quite comfortable.  If your career is in an inertial state it means you aren’t having to deal with new, strange and uncomfortable situations.  You aren’t having to do anything experimental, meaning you don’t have to deal with the fallout of failed experiments.

As a manager I’ve occasionally come across resumes from people who’ve finally been forced out of a job they’ve held for a decade or more.  There’s no indication of advancement, no attempt to upgrade their skills.  These people have accumulated a ton of inertia, and now their finally coming out into the light.  The shock must be terrible.

It is much better if you try to deal with this in small bite-sized pieces.  Always be learning new skills.  Always be trying new things.  Make sure you have something to add to your resume each year.  However, if you’ve allowed a long period of nothing to occur in your career – there’s still only one way to go: start doing stuff now.

As you can probably gather from this, leaving a state of career inertia can be quite painful.  Well, guess what:

No Pain, No Gain

I’m going to reveal something to you.  Last year in March I quit my job as the Director of Technology at HUGE, where I managed a department of about 35 people that operated technology projects with multi-million dollar budgets, to go sit in a glorified phone booth at Green Desk and startup my little tech company.  My rough calculation is that I’ve taken something on the order of a 30% pay cut in order to do so.  I couldn’t be happier.

In order to beat the inertia, you are going to have to sacrifice.  It may not necessarily be money, it may be familiarity, status, comfort or stability.  But it’s likely that something is going to take a hit.

The good news is that often this sacrifice is only temporary.  I can think of one occasion where I agonized for weeks about quitting a dead-end (but comfortable) job in order to take on something that was going to require real work from me.  Two weeks into the new job, I couldn’t believe that I had ever hesitated.

Sometimes it just means you need to work harder.  Turn the damn TV off and code something.  Or read a book.  Or show up at a meetup.  Eventually, your efforts will pay off.

How To Do-Si-Do

Time for some square dancing.  Know what a do-si-do is?  You go into the square and move past your dancing partner, then pass behind them, back to back, and finally retreat to your original position.  Career path movement can work the same way.

Often you are not going to have the opportunity for straightforward advancement.  In order grow, you will need to make lateral moves.  Below is may career path throughout the last decade.

I noticed something interesting.  The path of least resistance is to try and gain seniority at small or obscure companies.  From there you create legitimacy for your seniority in larger companies.  Experience in the larger company will in turn make smaller shops take a chance on you for a more senior position.

So in other words: first you hold a position at a small company.  Then you get the same position in a larger company.  Then you get a more senior position in a small or obscure company.  Repeat.  You are systematically building credibility.

Skate to Where the Puck Is Going To Be

A while back I was looking at developer resumes and I was really impressed by one in particular.  This person had started out in the late 80′s as a C++ guy.  Then around 1994 or so they hopped on the Java train full steam. (Correction: It’s been pointed out to me that this is too early to have been possible. The accurate date was probably a little later than this.)  They held several Java positions of increasing seniority until, mid-2000′s they suddenly switched gears and became a Ruby developer.  By the time they got to me they were a senior Ruby dev.

On the other end of the scale, I knew some FoxPro guys in 2004 (!!!) that were still trying to figure out how to switch to Java.  Eek.

Listen to the Great One.  Technology changes, and you must keep up if you have a technology career.  Pay attention to what’s going on – things change from year to year, and you need to make sure that you are learning what’s next.  The winning horse may not be obvious at the time but there’s no real downside to learning something even if it doesn’t work out.  All that time I spent learning Jini is unlikely to ever do much for me.  (Don’t know what Jini is?  Don’t worry about it.  It’s not coming back.)

Obvious trends right now are mobile computing, massively multiprocessor computing, big data computing.  Get on it.  Stop clinging to the idea that Cold Fusion will be making a comeback.

How to Leave the Party

How long should you stay in a job?  Primarily this is an intersection of the forces I’ve already talked about, but they converge on this question and create a specific model.  Let’s apply them first:

• A job is good as long as it’s adding credibility and experience.  If the job is serving you in the do-si-do, then it’s worthwhile.
• A job is good until it starts being inertial.  When you find yourself in a job simply because it’s too uncomfortable to move to another one, that’s exactly when you need to move to another job.
• A job is good as long as there is somewhere to go within the job.  Notice on the chart above I list two jobs under one company: Alexander Interactive.  The opportunity for me came because the company was expanding at the time, and the larger number of people meant I had the opportunity to grow into a more senior role.  Not all companies are like this, however – I found that in other places the bias is towards hiring from the outside for new positions rather than promoting internally.

Now I’m sure that you’ve heard that if you jump around too much it spooks employers.  This is true to an extent, but there’s more to it than that.  First it is partially a function of how conservative the employer is.  So partially gauge the number by the nature of your targeted employers.  Generally tech tends to be more forgiving about job-hopping than other industries.

My general feeling is that, unless there is an internal career path available to you at a company, the value a job brings to you peaks at about 2 years. Less than that does seem a bit flighty on a resume, but more than that and you really stop learning new things in the same position.  Time to go.

The other thing to keep in mind here is timing.  If you’re near the point where you’re ready to make a transition, try and do it during a hot economy.  It’s much easier to make job transitions when there are jobs to be had.  It’s easier to take risks when there are alternatives.

And speaking of risk:

Managing Risk

You are going to have to take on some risk.  Basically it’s unavoidable – you can only pretend otherwise.  No job is 100% secure anymore, no company is 100% secure.  The only real job security you have is optimizing your ability to make money.  In other words, no one can hand you job security – you have to make it for yourself.

By actively managing your career, by overcoming inertia and getting out of the rut, by strategically moving through jobs and by staying on top of trends, you can create value around yourself that, on average, should work out for you.  However all of these activities mean you will be taking some risks once in awhile, and you may have to deal with consequences.

See the second job I took in the chart?  ”Developer at AppNet”?  I realized on my second day there that I was in the wrong job.  Unfortunately I had just taken their relocation money to move out to California, and it was 2001 and there were no other jobs available.  I had to stick it out there for the next year.  It sucked.  Oh well, you win some and you lose some.  That’s risk.

And finally…

Make Yourself Awesome

I love finding people who have come up with creative ways to work on their career.  For myself, although I also thought it would be fun to do, I was aware that co-founding Django NYC with Kevin would be a good career move.  It paid off a mere two months later when I was interviewing at HUGE.

Start a group.  Write a book.  Make an open source project.  This stuff all counts as career work, even though it may not be immediately connected to your current job.  What it does is show to future employers that you actually care about this stuff, and are not just there to pick up your paycheck.  These kind of activities are one of the fundamental ways to tell good employees apart from mediocre ones.

Okay, that’s it.  Have a happy new year, and go out and kick some ass in your career.

Hover’s biggest feature is simply that it doesn’t try to mess with you.  I’ve found all the domain management stuff to be sufficient, although not amazing – but at least it just lets you work instead of screwing around trying to upsell you on things.  Additionally, Hover’s website is nice and clean and refrains from having NASCAR Driver Supermodels offer you AMAZING DEALS on .biz domains.

(Side note:  Don’t ever buy a .biz domain.  You might as well have your domain at .spam.)

My long term thought is that I want to move the various saaspire domains away from GoDaddy where they’re mushed up with domains that I personally own.  But I’ve actually never tried to move a domain from one registrar to another, so I thought I should give it a dry run with something that was real (i.e. not parked) but not mission critical:

This website right here.  Antipatter.

The procedure turns out to be simple.  First you have to “unlock” the domain at the current registrar.  I believe there are restrictions on the unlock (perhaps preventing you from changing registrars just before you have to renew – just when I want to change the most).  Then you need an “authorization code” from the current registrar, and you must give it to the new registrar.

Then you have to make damn sure your administrative contact into is up to date.  Now – here’s a secret:  I didn’t want to pay GoDaddy’s surcharge for not publishing my info, so I’ve given them really stale info to work with.  If you ever have run a WHOIS on any of my domains, well, I haven’t been in any of those places for awhile now.

Hover offers privacy built into their pricing model – there’s no option to NOT have privacy for your WHOIS info.  Note that their one year price is a little higher than GoDaddy’s, but the extra $5 is totally worth it to me.

Once the auth code is provided the request is sent to the current registrar, which then asks permission from whoever is the official administrative contact email recipient.  You reply in the affirmative.

Then they sit on the whole thing for about a week, during which nothing happens.  It reminded me a lot of a bank clearing a check.

In the end the domain transferred and I got a maudlin “Aw, we’re sorry to see you go” message from GoDaddy.  Pretty clean.  I did manage to screw up on not updating my nameservers fast enough, which is why Antipatter (*cough*) went off the air briefly yesterday.  However it wasn’t too painful of a process, so I’ll probably start moving other domains over.

In the past I’ve been in several organizations (businesses, in my case) where management has a sense that something isn’t quite right, and that they have to do something about it.  Often these kinds of problems can manifest in morale, in productivity and in the ability of staff to make good decisions and execute effectively.

Unfortunately, the frequent reaction of management is to attack a particular aspect of the organization, assuming that all of its problems can be addressed in this single space.  They will decide “we need to improve our processes” and focus exclusively on that.  Or maybe they decide that if they can only hire the right people, all of their problems will magically vanish.

The truth is that organizations are powered by an ecosystem of forces that interact with each other.  Good management examines the entire ecosystem, making tweaks (or sometimes wholesale changes) within to address the problem.  Good management knows that there is more than one type of adjustment to make, however.

The Forces at Work

As I mentioned above, a number of forces affect the organization:

People

Organizations, like Soylent Green, are made of people.  The competency, character and societal culture of those people will clearly affect the organization in all sorts of interesting ways.

Generally I divide “people issues” into two categories: the first is what people bring through the door with them – their mannerisms, how they work with others, how they react under stress and so forth.  The second category has to do with how well they fit with their role within the organization.

On the “through the door” side, these issues are well covered by the studies of psychology and sociology.  The best hope is for management to become sensitive to these kinds of issues, and encourage a productive environment.    Frankly, this is an opportunity for strong leadership – management can encourage people to act in a certain way (I’m a proponent of “nice”, for example) by setting the tone themselves.

It should be understood, however, that people’s personalities are what they are – and that they’re unlikely to change for the organization’s sake.   If you can make this work with your organization, great.  If not, well, don’t expect to be able to correct it.  If you can’t live with it, that person probably needs to go.

The role aspect has a fair amount more play.  The essence of employment is that its an agreement by a person (the employee) to fulfill a role defined by the employer.  How well that role is fulfilled is the success of the employment.

Sometimes what is at first glace a problem with a person, is actually a problem with the person’s fit for their role, or the definition of the role itself.  For example, if a role has an unusually high turn-over rate, it may be that there are other issues at stake.  A position that no one can fill is worthless.

Issues in the “role-fit” category can often be addressed by adjusting the role.  This doesn’t have to be negative: sometimes the action is to expand the role – increasing the person’s responsibility.  Sometimes (often) some degree of lateral movement is required.  Often a role organically changes under a person and becomes a poor fit, even though it originally was a good one.

Process

A lot of consultants make a lot of money in this area, because process adjustments are relatively easy to define and execute, at least in comparison to some of the other dimensions in the ecosystem.  Again – process isn’t everything.  It’s one dimension of the organization, and while it can be incredibly powerful as an agent of change, to focus on it in exclusion of the other dimensions is to hold an incomplete understanding of the organization’s health.

Process, or “how we as an organization repeatably do things” is part of the day-to-day structure of the organization.  It intersects with roles, systems (automated or not) and defines the day for many members of the organization.

There are three big ideas to understand about organizational process:

1. Process has weight: Every procedure, every form, every mandatory meeting and so forth places a little (or in some cases, a lot) of additional load on an organization.  This weight adds up.  Process weight can have the very serious consequence of adding drag to the organization – slowing it down, preventing it from executing on things that are actually important to the business.  This is typically the situation that people describe in large organizations, and is what they seek to escape by fleeing to small organizations.It is important to understand that processes don’t come for free.  Every time a new process is added, the drag on the organization increases.  For this reason, management needs to ask itself if the new process is worth it – are they buying enough of a benefit to justify the drag.  Also, they should question how much drag they are adding, and if there is an alternative that could be implemented in a lighter, more efficient method.
2. Processes need to be refactored.   No matter how good an idea it seems out the gate, any process can stand to be improved.  (And sometimes replaced, or just scrapped.)  The strongest organizations provide mechanisms by which people who live with the processes on a day-to-day basis can report on how they’re working, and submit ideas for how they can be improved.  Then, the organizations actually act on those suggestions.  ”Dinosaur” organizations occur when the organization is not capable of change – either through lack of willingness or ability to execute a change.  Dinosaur organizations become weaker and weaker over time.
3. Process requirements change with organization scale: If you are in an organization that is changing size, there’s an additional factor: what worked when you were ten people fails when you are thirty.  What works at thirty fails at 100.  Scaling is the phenomenon of processes breaking repeatedly. If you are in a scaling organization expect process to break more frequently, and that you will need to refactor the way you do things.

Business Quality

Here’s one that often get’s missed: if you are in a bad business – if you are unable to make money (or otherwise succeed in a non-business organization), things are going to be bad no matter how many amazing people or awesome processes you have.  There needs to be a business engine that keeps everything going, or you have nothing.

This third pillar can often be the limitation for many great operational people who know the ins and outs of scaling a business, but don’t operate well at the actual layer of the business itself.  Why does it work?  The business model, the marketing, the various outward-facing ingredients that determine the success or failure of the organization in the marketplace, these simply cannot be ignored when thinking of the overall health of the organization.

As a personal example, I spent a while trying to build a startup focussed on the music industry at the exact same time the music industry was going through its existential collapse. (From Napster to iTunes, basically).  I had tons of great ideas for how to scale the business from a personnel and process perspective, but the underlying quality of the business wasn’t strong enough to support it.  So it didn’t work out.

Leadership and Culture

The final two dimensions are less of pillars and more of the medium through which quality spreads through an organization.  You can think of them as magnetic poles, because they interact with each other and make things happen.

Leadership comes from the people at the top: the founders or top management.  Leadership obviously defines a great amount of strategic and executional aspects of an organization, but additionally, leadership defines the original culture of a company (often subconsciously).

Over time, through hiring and firing practices, and internal interactions, leadership creates the culture of the company.  If leadership acts unethically, it will become part of the company culture that unethical behavior is acceptable, and in turn the company will attract unethical people.  For all aspects of culture this is true: drama begats drama, creativity begats creativity, hostility begats hostility and so forth.

In a future post I’ll start having a look at treating organizational issues in a holistic manner, by applying this model to understand where problems intersect, instead of approaching issues exclusively from a single dimension.

Not to be overly simplistic, but there are three basic perspectives of web development:

• Usability, or user experience.  In its broadest definition, this is a measure of how good the web app is to the user; how much value it creates for them.  In a more specific sense it reflects a stack of best practices (and very, very occasionally innovations) that add up to the user experience on the site.
• Business value.  In the broadest sense this is “what the web app is doing for the people who pay its bills”.  This can be interpreted as monetization (via ads, conversion etc), or simply controlling the budget and schedule of the development project.
• Feasibility:  Often taken as a technical subject, it its broadest sense, feasibility simply describes how much effort goes into making something happen.

The catch is, to truly make something successful, both from the perspectives of  project management (launch on time, on budget) and product management (build something that is successful in the real world), you need to keep your eye on the ball for all three perspectives.  Otherwise you’ve failed.

For example, I’ve seen user experience people run wild with feature ideas, not paying attention to the development cost that such features incur.  I’ve seen the opposite as well – web apps with terrible user experiences that are vulnerable to losing market share to better designed products.

I’ve seen project sponsors focus on business value at the expense of user value, and then are surprised when no one uses their product.  I’ve seen projects that have been managed perfectly for schedule and budget (perfect scores to the project manager) fail in the open marketplace because they didn’t bring value.

Which Brings Me to the Siloing Part…

So to really succeed, you need to synthesize all three perspectives into a unified vision.  That’s why it drives me crazy to see so many people in web development so oblivious and willfully ignorant of perspectives other than their “native” perspective.  UX people that make features up without understanding development cost.  Developers that launch features without considering a usable interface.  PMs that manage schedule and budget without comprehending the quality of what they’re developing.

Do you really want to be good at this?  Then learn about the perspectives that are different from what you deal with every day.  UX people should understand the constraints (and advantages) that technology puts on development.  Developers should appreciate the business goals and usability issues related to the products they develop, and PMs should understand the vision of the products they’re managing.

Process is Not The Answer

Process alone, especially any process that involves creation of a bunch of intermediate documents or getting people to show up for a bunch of meetings, is not going to fix this.  The people who work on web applications need to internalize the other perspectives, and make it their personal responsibility to understand this.

Once this happens, an organization can support people by offering pathways to education (courses, books, conferences etc), but the people need to personally commit.

Admittedly this makes a barrier to scaling, because not everyone is interested in other perspectives.  They got into their own discipline (development, project management, visual design, UX etc) for a reason, and that’s where they want to stay.

My view is that the willingness to see the other perspectives is what separate the truly talented from the mediocre, and the mediocre will not be working for me any time soon.

I’m going to go out on a limb: it’s the ultimate remote control.

This is entering the age of networked hardware.  The “internet fridge” concept that nose-dived 10 years ago is back with a vengeance: hardware will be on the net, addresses web services.  Cars, appliances, strange boxes that you connect to your TV and home – all of these need user interfaces.

However, UI’s are both expensive to build, and frankly hard to pull off well.  That’s why we’ve seen generation after generation of terrible UI’s from hardware manufacturers – each being forced to reinvent the wheel over and over.

But now – why bother?  Apple creates tools to build pretty good application UI’s without an overwhelming amount of work – and a streamlined mechanism (the iTunes app store) to distribute them.  So instead of sinking a lot of cost into developing your own complete user interface for your hardware – why not just hook up to an app on a device that the user already owns?

There are already examples of this in the wild: the ZipCar iTunes app will actually unlock your car.  If you invest in a Sonos sound system, you can run it from an iPhone app.  And apparently the Ford Sync system is slated to run apps that you’ll be able to buy from the iTunes store.

So what might be possible if you had a 10-inch device that could act as a remote?  Run a gigantic entertainment system?  Dock in any piece of hardware and become a touch screen control for it?  How about I use an “iSlate” to start my car?  A $700 device isn’t that much against a $20,000 car.  Or – how about running your home?  AC, Heating, security etc could all be run from the iWhatever.

Using the standard user interface SDK and guidelines, people build controls for hardware based on behavior that is already well understood.  The learning curve is reduced, and companies don’t have to sink a lot of money into UI development for their devices.  It’s a slam dunk.

Having spend the last 4 years or so working in web agencies, I’ve seen the inner workings of the IT department in many, many companies – from mom and pop operations to Fortune 500 outfits. More often than not, the IT department is really, really broken.

First, let’s have a look at what broken means:

Symptoms

Lack of Prioritization

Faced with conflicting demands from different business units, the IT group is unable to plan where to spend its energy, and so lives constantly in a reactive state, servicing whatever the squeakiest wheel happens to be at any given moment. Essentially it’s locked into a state of putting out fires, and is unable to put energy into actually making things better.

Often this can be a result of lack of “informed power” regarding the IT department: senior management doesn’t really have the understanding to view IT as an allocatable asset, and no one in IT has the power to push back on the immediacy of the next business stakeholder’s fire – regardless of whether it’s important to the business or not.

Lack of Bandwidth

Related to the prioritization issue, but slightly different, is the fact that most IT organizations simply don’t have enough people to do the volume of IT work demanded by the organization. At the same time, however, there is resistance to hiring in IT, because IT is viewed as a pure cost center in most organizations. In other words, it is not viewed as a vehicle through which the organization can make money, but rather just a place where spending occurs. In an environment of cutbacks, it will be trimmed.

Communication Problems

Often business stakeholders and IT just can’t understand what the hell each other is saying. Business users generally have a sense that certain features are vastly more important than others, but don’t always communicate this prioritization if they’re not asked. Additionally they may jump immediately to a specific feature implementation (that may be taken as a requirement by IT) that isn’t really what is best for them. Business stakeholders are not going to have any sense of feasibility, however: they don’t know how hard it’s going to be to do things.

IT comes from the opposite end. They understand feasibility, but not business importance. Unfortunately IT has the habit of taking all “requirements” from business as of equal importance, and just plugging away, looking at things purely from a feasibility perspective. When combined with the business stakeholder’s habit of proscribing features instead of just stating needs, it can lead to a lot of wasted effort.

Fortress Mentality

When things go wrong with technology, who get’s blamed? IT. When things go right, who get’s credit? No one.

In a sane business decision, one weighs the upside against the risk. There are many cases where it’s worth it to take a risk in order to achieve a big payoff. But IT is essentially balancing risk against an upside of zero. Naturally, that means they will seek risk avoidance in every case. This minimizes the chances of things going wrong, but it also minimizes benefits that might be had from technology.

Consequences

These symptoms turn an organization’s interactions with their own IT group into a big, dysfunctional dance. Business stakeholders feel that IT is unresponsive, slow to complete projects, incomprehensible and too restrictive. They start to come to the blanket judgment that IT is not going to be a solution for their problems.

IT for their part, feels under-appreciated, constantly putting out fires, under-staffed and forever dealing with clueless users. They have ideas for how to make things better, but never have time to implement them. On the projects that they do get to work on, they’re shocked to find out that features they’ve worked so hard to implement are being unused by business stakeholders.

This dysfunction has even led to a culture of circumventing IT. Business stakeholders (as much as they might like to) realize that they simply cannot live without having certain technology services performed, but feel that IT will be unable to help them. At this point they may look for outside help – either from product vendors that claim to “do it all”, to agencies that usually make the same claim.

The product vendors or agencies may be just fine, but that doesn’t change the fact that the organization has chosen to outsource a part of its business, not on the basis that it isn’t core to their enterprise, but rather on the basis that they are incompetent to execute. Since IT work is often the automation of a core business process, outsourcing it can lead to a weaker overall organization.

So there’s a picture that’s probably familiar to many. How do we fix it though?

How to Fix IT

Fix the Accounting

If technology has benefit to an organization, then that benefit needs to be reflected in the way the organization accounts for itself. From a financial perspective, a business will look at the profitability of its various components (at various levels of formality, depending on the size of the business) and make decisions about where to invest.

Businesses need to stop looking at IT as pure cost centers. That means that the credit for adding value to business units needs to find it’s way back to IT somehow, and be measured against IT costs. Here’s two possible ways to do that:

Charge the Business Users

If business stakeholders use IT services, they have to pay for them. That way the “cost of IT” lands squarely on the business stakeholders that use IT services. Of course, the side effect of this is that the quality of what the business users are paying for will immediately fall under scrutiny, but that’s okay. The point is to make the organization stronger.

Credit IT with Results

Let’s say IT builds a system for a business group, and as a result there is a reduction of cost for that business group (or even better, a measurable increase in revenue). Some of that savings/earnings increase should again be credited back to the IT department, and shown as income (or something equivalent). Again the political side of this action will be determining how much credit to kick back to IT. Too much would additionally have the negative side effect of making the business groups less interested in such activities, as they aren’t being credited with the results.

Hire Technology Leadership

Particularly aimed at the communication / prioritization issues, the solution is to hire a senior person with a technology background who:

Can Translate: Someone who can speak the feasibility-heavy language of developers, and the priority-heavy language of business stakeholders. Someone who knows when to ask context-breaking questions, such as identifying the real business issues instead of just accepting prematurely prescribed solutions dictated by business stakeholders.

Can Design Solutions: Someone who can take the real business requirements and come up with a technology-based solution, then take that solution and articulate it as technical requirements for the IT Staff. This is an individual who understands both technology and business.

Has Authority: This person needs to be able to push back. It is inevitable that forces within the organization will attempt to mis-use IT, either pre-selecting the wrong solution, or by short-circuiting the priorities of the IT group. The Tech Lead has to be able to say no to those forces, and find a real solution that balances the real needs of the business stakeholders within the larger context of the organization.

Decompose IT

The most radical step is to send IT staff out into the wilds of business stakeholders. Instead of a common pool of IT resources, an organization could choose to send, for example, a technical PM, a couple of developers and a sysadmin out to exclusively work on the problems of a particular business group.

This is an approach that probably makes the most sense when the needs of a business group grow beyond a certain point. At that point using centralized IT resources becomes less effective than maintaining their own technical staff. The obvious concern is that efficiencies of scale are lost when technical resources are distributed amongst business groups, but this is a good choice when the effectiveness of those resources within the individual business groups makes up for any loss from decentralization.

The upside to decentralization is that the accounting problems just go away. The efforts and results of these distributed staff are just measured within the individual business group, who will either be able to justify the expense or not, on the same kind of merits that they use for everything else.

It is important in these cases that technology leadership/management is included in the distribution, or the business units can find themselves saddled with a lot of the communication issues mentioned above. Because of this, this is a significant and potentially expensive step for a business group.

Conclusion

IT in most organizations is broken – it’s not providing the kind of support and advantages to business users that it could, and is the source of frustration and feelings of helplessness for many employees, both business and tech. By re-structuring the way IT fits into organizations, it can deliver on it promise at last.