I’ve hypothesised on the future of analytics before, on how companies will adopt and integrate information from web analytics into their Business Strategies for the greater benefit of both the business and their customers. In July the investment by Google and the CIA in the Temporal  and Predictive Analytics firm Recorded Future has crystallised my view on what the future of analytics is – a critical standalone service.
Since the first time I drew information from a website that related to the company’s products rather than the websites simple usage statistics, I’ve realised the strategic importance that analytics could play in the future of a business. I am not belittling the importance of good website monitoring here, what I am saying is that there is more gold to be mined.

This article reflects my thoughts on how these analytics tools will evolve in importance and soon become critical to any serious ongoing businesses strategy.

The Way it will Be
The analogy I want to use is that of a room with a desk and three windows, each window is a different view of the world, and on the desk is a blank page which represents your business strategy. The room represents your business.

The first window looks out over the online presence which is under your control, primarily your own web properties, but this can extend to other online elements as well. This view is fed by website analytics, this is the analysis of Clickstream Data to determine what customers like and dislike about not only your website, but your company. This can answer questions like which products they are most interested in or even what questions they most need answered about your products or business before they engage with you.

The second window looks out over conversations that your customers, the media, and the general public are having and what they are saying about your company and products out in the world right now – what they think is right or wrong about the things you do. The view to this window is fed by Social Media, Polls and Tailored Website functionality which is designed to extract customer opinions and attitudes without them being aware.

The third window is a view of the future. This view is fed by Temporal and Predictive Analytics and it is based on two things, the first element is what people are saying now about your company or products performance  in the future, the second is based on trend analysis of companies and products like yours. Temporal Analytics analysises what has been, scraping data and information from databases all of the web to construct a historical landscape. By analysing the complex relationships between elements and their context on this landscape and looking at their evolution over time, it is possible to build a comprehensive model of what the future will look like, which is the function of Predictive Analytics. (I’m grossly simplifying the nature of Temporal and Predictive Analytics for the sake of brevity).

Obviously you can easily choose to change the views of the second and third windows to give you a view over your competitors. For the first window you need to resort to competitor data from companies like Compete.com.

Now let’s get back to the desk and the page representing your “Business Strategy”. You now have three sources of information with which to view your customers- what they are looking for, what they think about your brand, what they expect from your brand, what’s making them buy or not buy your products. Furthermore, you can see what people and the market in general are predicting for your brand, and the general trend insight into where your brand is going.

This is an incredible starting point for any business strategy, a near complete view over the online landscape in which your brand or business operates. And here’s the kicker: It’s all available in real-time.
This means that your strategy need not be a closed document, in fact it can be a living document measured against real world metrics in real time.

Part 2, The Way it is Now, will follow in a couple of days

About 11 years ago (OMG) I had my first experience of Web Analytics. My boss at the time asked me to produce a Weekly report from our log files on what our users where up to. It was a painful process and one which come every Thursday I would begin to dread. I would produce reports and send them out into the company (read ether) and wonder who if, anyone read the reports.

We ran an e-commerce site and things like goals, funnels and metrics were really not quite on the radar yet. I did manage to identify a couple of hacking attempts, and roughly identify the Geo-location of our customer base, and our most popular items. There really was no critical information which added to our companies’ bottom line. I fondly remember my Monday morning report back which could have been a recorded playback of the previous Monday. The coffee was brilliant.

Flash forward to today, as I sit producing my monthly reports for clients, providing them with rich data which is relevant to their online and offline marketing, and their business bottom line. So what’s changed in that time to create this living breathing thing we call analytics and reporting which can literally change the way we do business for the better.

The Changes
Well most importantly the tools have changed. The primary tool for the analysis and reporting on Clickstream data Google Analytics turned five a couple of weeks back. The purchase of Urchin by Google and the choice to make this tool free was for me one of the defining moments in Analytics, sure there were other tools available, too many to name, and many that have fallen away. Google Analytics though introduced Website Analysis to the masses. Being free, it also enabled a lot of websites to dip their toe in the water and begin gaining an insight into their users.

By today’s standards the measurement metrics and benchmarks by which we measured sites have changed radically, and this is also in no small part pushed by Google and its competitors. Our ability to segment data and drill down into true user trends is giving us a radical insight into the behaviour which affects our business.

Second to this, web analyst’s have changed. We understand the capabilities of the tools more, we are beginning to understand how to bend our websites to supply the data, and most importantly we’re beginning to demand more from our data.

The 90/10 Rule
There’s a 90/10 rule in analytics. It basically suggests that you spend 10% on your analytics software and 90% on your analyst. Yes, I know I just said that Google Analytics is free and you think the jokes on me, but if you really think it’s free then you haven’t installed it properly and you haven’t customised any code to the benefit of your data – so the jokes on you!

Anyway, back to the 90/10 rule. What makes me proud of the 90/10 rule is that the critical change it suggests is that as Website Analysts we’ve earned this 90% by proving that data alone is not enough to answer questions about website behaviour. Regardless of the solution you choose to implement to monitor your website, you still need someone to analyse and understand that data.

As much as Analytics is not something you just put on your site, Analysis and Reporting is not something you simply throw at your webmaster or content editor. It’s a specialised role which requires a wide experience and understanding of all aspects of website creation, digital marketing, and online strategy.

By the way, the 90/10 rule is also the reason I love Google Analytics so much, and why I believe it is perfectly suited for the South African market. In an environment where our budgets are smaller than those of our overseas counterparts, it gives companies the ability to collect data and have it analysed without breaking the bank.

Where Are We Now?
Right now as I sit, with a pretty good view over the analytics market in South Africa, I would say that we’re still relatively far behind. There are a few companies which are actually utilising incredibly advanced reporting and analysis for the benefit of their businesses, but at the same time there are also some large companies which are running embarrassingly small implementations. In fact they’re pretty much checking the “Analytics, Have we got that? Y/N” Y box.

The blame for this though doesn’t really fall solely on their shoulders. It’s really the responsibility of both their website solution provider and the client company to be demanding a better insight into their website and customer behaviour.
There are three obvious reasons why website solution providers don’t push Analytics as much as they should:
1. The Clients don’t demand it
2. The Website provider does not understand it, or have the capacity to analyse the data.
3. The website company actually doesn’t want the client to understand how the website is working. A lot of companies and website solution providers still view websites as an open and shut project, a project with a finite end. So why report on anything about it when the project is finished. In addition why expose yourself to potentially damning data.

The truth is that all of these reasons are part of the old world of website’s, Web 1.0 as it is so charmingly referred to.

A website cannot be looked at like a finite project. Analytics and reporting offers us the opportunity to analyse what may not be working and to fix it.

On deeper levels we can actually extract information which can become critical to the bottom line of the business. For instance wouldn’t you like to know what the most pertinent questions are about your company or product? How about a breakdown of these questions by province with a real time report, that a marketing department can access so that they can ensure that these questions are answered in all marketing collateral.

How about understanding which of your product features clients really want instead of the ones they can only afford to buy, I’ve seen that one answered before with the reward being a reconfigured sales package and a substantial increase in sales. Okay, I’m getting too excited now.

Although, if you think I’m excited, there is very little that compares to the tangible excitement of a business owner who realises for the first time that this data is readily available, and that they are seeing real insights into parts of their business and customer behaviour that they previously did not know existed.

We Have Something for Everyone
Customisable reports in Analytics also offer the opportunity to offer different pieces of segmented data to different levels of the organisation. And this for me is the greatest source of long term Analytics joy.

A good analytics implementation will give you clear insights into the website metrics that affect your business, a great analytics implementation will give every level of your organisation an insight into how their role in the website and related activities affect your business.

A custom report for marketing for instance can deal with the questions users have, what products and content attract them to the website, which pieces of content chase them away, what website content engagements increase their propensity to purchase… the list goes on.

Where Will We Go?
I’m very confident that Analytics and Reporting will continue to grow radically. I’m not going to say that 2011 is the year of analytics, because it doesn’t work like that. The progress companies will make in utilising the services of Analysis and Reporting will be far more subtle, although the improvement to their business will not be as subtle (in a positive way!).

Analysis and Reporting cannot but become more pervasive because of the insight and benefit it provides. It is also never going to become a totally automated process the role of the Analyst will always be critical in this relationship. As good as reporting systems may get they will never really be able to totally investigate, understand and draw conclusions the behaviour of human interaction online. There will always need to be someone with an advanced understanding of how websites, marketing and users all relate to each other and work.

Conclusion
One of the simple and most important things about Analytics and Reporting is that it removes much of the guess work. The kind of guess work you get in Agency and Client Company meetings where everyone has a shot at second guessing the needs of the customer, and the opinion that wins is generally directly related to the rank of the owner of that opinions position in the Organisation. With good Analysis and Reporting you should have a clear view of the customer’s needs and at the end of the day no-one out ranks the customer.

I love analogies, I use them often when I’m discussing online related discussions because I think it’s useful to use real world equivalents of on line processes because the concretise concepts in the audiences mind by allowing them to relate to a process through their own experiences.

So my analogy for building the propensity to for the user to commit an action – for instance an online purchase – is not going to be the process of purchasing a product in an offline store, because I think it’s almost a totally different process to online purchasing, and it think that where each may have their own unique advantages, they are actually worlds apart. My analogy actually doesn’t relate to purchasing at all, but the emotive experiences that predicate a certain type of behaviour. My analogy is “catching a flight”.

The Issue
When we consider the actions we would like a user to undertake during their online experience of our website we often spend a lot of time upfront designing and developing strategies and processes to expedite their path to the action. We spend a lot of time analysing the competition, the User Interface, and the Architecture of the site. Essentially checking all the check boxes which we consider good web design.

Where I find that we often fall flat though is in the actual day to day modification of the website to enhancing the user experience and building the propensity for them to commit the actions which the entire site was built around. I think in some ways we’re either burdened by red tape or budgets, we can’t do endless a/b testing because the budget doesn’t allow for it, or the chaps from IT are just getting miffed because we’ve asking them to change a page 23 times in the last three months.

Most importantly though, we don’t really understand exactly what is going on inside the head of our users at critical points in their website experience. We don’t understand whether we are increasing or decreasing their propensity to meet one of our website Goals.

Why should we even Consider Propensity?
Let’s say you want a user to purchase a product on your website, they’re going to go through several stages, most of these involve education about your company and your product. This learning process involves the answering of some basic questions, Are you a reliable service provider? How fast do you deliver? Are you the cheapest? What are the product specifications?

These are all basic steps we undertake before trusting a company and a brand enough to buy their product, barring a few exclusions. There’s a process user will engage in before buying your product, and it will involve them going through all of these steps (and more) before they commit to handing you the numbers for their plastic.

The Analogy
So I arrive at the airport to catch my 18:30 flight, immediately I’m a bit confused because the airport has been done up and the flight desk has moved along with the notification boards. It raises my stress levels a bit, but I search around and finally see a board with my flight number and booking desk. I cross the concourse and see a queue of people rivalling my memory of the voting queues for the last election. My level of tension rises slightly more, did I choose the right airline? Am I going to make my flight? Where does the queue actually start?

The last question is answered about four feet in front of me and about 300 feet from the actual check-n desk. I gather from conversations around me that I’m at the right desk, going so far as to ask another customer if they’re on the same flight. And so begins the wait, it’s not a normal wait though because I know that for every minute I wait in the queue, it’s another 30 seconds I’m going to have to shave of my sprint time to the boarding gate. Stress level is now at a stage 3. I then find out that the flight has been moved and is now departing sooner than expected. Needless to say that I and 30 other people start to run through freshly carpeted halls to our boarding gate. Stress level is now at Def-con four.

At this point I decide to take a step back, and for the first time notice the level of aggression amongst my fellow passengers. They’re now just generally rude to everyone, from the girl selling muffins at a coffee counter, to the lady collecting boarding passes.

What’s even more interesting is that when we do eventually board the plane, the unsuspecting staff are greeted by a bunch of rude, and belligerent passengers. The flight staffs probably don’t have a clue why they’re behaving in this way.

What strikes me as interesting about this entire process is that it wasn’t a single moment in the process which led to the behaviour of the passengers, but a sequence of events which led a bunch of people to behave in ways which they would probably normally be embarrassed by. They all seemed like a nice bunch of people when they arrived at check-in, I got a few smiles, and people were courteous enough.

So we can only assume that this sequence of events has built a propensity for these passengers to behave in this way.

Admittedly it’s an extreme example, and it is demonstrating the propensity for a negative rather than positive action.

How Does This Relate to a Websites User Flow?
We all have an action we want users to commit to on our website, sometimes a few actions, hopefully never many actions. It’s obvious on sites that I visit that they have a single goal, or action they would like me as the user to commit to. These goals are clouded often not with obstacles which stop me from moving from one part of the process to another, unless no attention has been paid to usability, but with obstacles which lead me to a propensity for suspicion or discomfort.

These emotions may be so minor that I as a user do not actually acknowledge them, but subconsciously prompt me to delay the purchase, or seek a service provider who provides me with more answers or a greater sense of security.

This can have a devastating impact on this websites conversion rate because the user has now been prompted to look at options from other service providers.

How Do We Understand A Websites Effect on Propensity?

Well the first answer is to build a funnel. A funnel is a series of steps which the user must complete on the way to an action. In order to have a Funnel we need to have a Goal, and the Goal is obviously the action we would like the user to complete.

Setting up a Funnel with which we can immediately identify problem area’s along the stages towards reaching the goal. A Funnel will show us not only where users are leaving the funnel, but what other information on the website they are looking at. The idea here is to then incorporate as much of this “missing” information into the process as possible without making the experience uncomfortable.

So, if you notice that a fair amount of people go off and look at the guarantee for the product, which takes them out of your purchasing funnel, then you need to finds a way to either incorporate the guarantee information into the funnel. If you do this and they’re still leaving in droves, then there’s obviously something wrong with your guarantee.

Which leads me to the second point, which is how do we tell what they user is feeling when they leave. Analytics and tracking is good, but it’s not good enough to be able answer this kind of question. It’s a conundrum best expressed by this question: I have an Average Pages per Visit of 7, are people looking at this much content because they’re interested in my product and want to read as much as possible, or because they can’t find what they’re looking for?

You could probably answer that by looking at Time on Page, but the best way would really be to look at utilising surveys. Exit polls are actually a great way to assess user satisfaction with your website, and I’m not going to go any further on this because Avinash Kaushik consider this in his excellent article “Eight Tips For Choosing An Online Survey Provider”

The last option is to actually commit to some usability testing, and this can be costly. The issue really with cheap user testing, asking a friend a to to try buy something off your site is that you’re immediately slanting their actions by defining parameters for their experience. Good usability testing needs to be broad enough to allow consider the experience of discovering your site and products, and that is not a simple task.

I’ve mentioned A/B testing, it’s incredibly important, although in some instances very difficult to justify to clients. A/B testing is a fantastic way to solve a problem once it has been identified, the issue as I’ve described above is actually all about identifying the problem. You would utilise A/B testing to test the potential solutions identified by your users or business analyst.

Last Piece of Advice
Really the last and most important piece of advice is to be totally brutal. Be brutal with your website, and remember that there are no sacred cows. The first port of call should actually be you going through your website and actually trying to commit the action that you would like your users to commit. If it’s not working for you, then you can be damn sure it’s not working for your users.

The announcement on Friday on the Google Analytics blog More choice for users: browser-based opt-out for Google Analytics on the way is already creating a lot of hype. Questions are being asked about how wise it is a move for Google Analytics, with some even suggesting that it will bring about it’s downfall. The implication is that this is a gesture of good faith by Google based on the criticism they have received recently about their approach to privacy highlighted with the launch of Google Buzz.

Right now the analysis on the opt-out tool is a little thin and the sensationalism is rather, well thick. The main fear is that many users will install this tool when it becomes available and simply opt-out of stats and disappear of the tracking radar, and where this is potentially true, the idea that this will actually happen is pretty far-fetched. It also represents a kind of old world thinking about Web Analytics that doesn’t stretch beyond Unique Visits, Visits and Pageviews. As much as I resent seeing Visits drop at any stage, Web Analytics is not about tracking the individual or simply monitoring the stats, it’s about identifying trends and actionable data, and developing strategies to act on that data. In fact a lot of the data you see in Analytics is actual based on a sample, this is most common when you access custom reports.

Huge number crunching is required to generate pivoted data, and in order to do so more rapidly Google will grab samples of data to represent a larger segment. This action doesn’t compromise tracking, in fact it expidites it allowing us to view our data from more angles more rapidly.

It’s also worth pointing out that Google isn’t the first to do this, similar tools exist for Coremetrics and Omniture’s Site Catalyst. We also already have a segment of users who turn of cookies in their browsers, effectively disabling all cookie based tracking. Complex calculations between log files and cookie trackers can give you a reasonable estimate on the size of this loss, and no doubt if you’re worried about the loss you’ll be researching that methodology right now.

So where this opt-out tool may cause us to loose a segment of users, which is never a good thing, we will probably not loose the insight Google gives us into user trends and segments.

The Google Analytics Map overlay is a particularly useful tool for segmenting users in South Africa. The Map Overlay page itself offers the further segmentation of users by other Metrics like Time on Site, Bounce Rate, Goals and some other very useful metrics. The question that I have had for some time though is how accurate is this data, and depending on its level of accuracy how actionable is it?

It’s also important to note that Google Adwords offers segmentation by region in their Ad Planning Tools. So the question of geographic accuracy has implications on the geo-segmentation of Pay Per Click (PPC) campaigns.

My proof is not that complicated but if you want the simple answer you’re welcome to skip to the Conclusion.

The Issue

Google Analytics Map Overlay utilizes Geo IP’s to locate users. Google captures the IP address in order to do the Geo-IP look up. So this seems like a logical point to begin testing the system. Being an army of one I chose to test the system from three locations where I will be assigned IP’s work, mobile phone, and home. I utilized Maxmind and the GeoIP tool.

The first test from work got South Africa> Johannesburg> Rivonia, which is great because regrettably that’s where my company is based, on the border of lesser Tuscany. The second test from my mobile phone got South Africa> Johannesburg, which is still okay for city and regional data. My third test from my home connection got South Africa> Western Cape, eish, that’s quite a commute.

This example demonstrates an already known issue with Geo IP, which is that if your service provider is based in another city and assigns you an IP you will be Geo IP located in that city. Even though I’m in Johannesburg my ISP is based in Cape Town I am being assigned an IP in that City. Now this is not always the case, as you will see further along in this post, but it is an inaccuracy that needs to be considered.

Now, I’m not going to jump up and down and claim to have discovered a flaw in Google Analytics, pretty much any Web Analytics tool paid or not utilizes Geo IP and will therefore have the same issue. The truth is they’re all aware of it, and do not state that this data is totally accurate.

There is a second less obvious issue with Geo IP, Maxmind states that at a country level Geo-IP is 99.8% accurate, and for South Africa it is 68% accurate to within 40km of the location. Which means that Geo-IP probably gets city level Geo IP wrong 32% of the time.

What this means is that when you look at an Analytics Map Overlay for South Africa you should consider that accuracy falls with every drill down.

So How Bad is It?

So where does this leave the Map Overlay? Well actually it’s not really as bad as it seems because the major ISP’s seem to do a pretty good to excellent job of geo-locating users, and in South Africa the major service providers, predominantly Telkom & Internet Solutions, account for the majority of most large websites traffic. I did a test on some of our websites to prove this, and it’s probably worth setting this up on your own website as a litmus test of map overlay accuracy. Basically in Custom Reports I created a report that segmented users by Internet Service Provider, and then drilled down to city.

This report shows that large portions of users are from the major Internet Service Provider’s, and that if I drill down those Internet Service Providers I can see that they’re actually segmenting customers to a city level.

So the major portion of my traffic is actually being located geographical with a reasonable amount of accuracy.

The final confirmation of the general accuracy of the Geo-IP data is pretty simple. We have a high traffic site serving Durban and the surrounds. I created another custom report for this which was simply Country>Region>City.

This report indicates that 72% of our traffic is from the correct region, and by extension the cities and suburbs add up as well. What’s interesting is that if I modify the custom report to be Source/Medium instead of city.

It’s interesting to note that for the Regions/Provinces other than KwaZulu-Natal organic traffic is almost equal to direct, and in KwaZulu-Natal the reverse is true. This indicates that the large portion of users outside of KwaZulu-Natal are finding the website through organic search which makes sense because users outside of the region will stumble across the website through information which is non-regional specific.

The other piece of information that shows where users are from is the regional access to content, the pattern for each region/province is different enough to show that users from the different provinces have different patterns of content access.  That’s another simply modification of the custom report.

Okay, so in summary. It seems like on the surface Geo IP, and the Internet Service Provider issue really nullifies the accuracy of the data in Map Overlay. However, the issues actually don’t seem to have that much of an effect on the regional data considering my findings on the KwaZulu-Natal site. This is because the majority of users are coming from the major ISP’s which actually do a pretty good job of segmenting data and obviously updating Geo IP information. Having said this though at a city/suburb level I think there are significant issues, most of which Google Analytics are actually aware of and are working to fix.

Conclusion

The conclusion really is that the data on a Country and Regional level is totally actionable. So, you could actually use it to feed information to the companies about the different customer behavior in different provinces, and it could actually be used in offline campaign strategy.

So the data is not anywhere as accurate as GPS, but it is accurate enough to be actionable, which is what actually counts. PPC geo-segmentation is also pretty safe because it only segments to the provincial level.

Having said this though, on a City and Suburb level you’re going to have to do a lot of work outside of Analytics to make sense of the data, and I do believe the integrity of this data is probably so skewed by the ISP and Geo IP issues, that is actually isn’t worth it. At this level you’re probably better off using either surveys or information gathered from site registrations or competitions to figure out what users in different areas of a city actually want.

When a customer asks you for a report do you deliver or engage them? There is a shift in Web Analytics in South Africa as more clients are becoming aware of the availability Clickstream Data and free Web Analytics options like Google Analytics and are requesting information on their websites performance.

The question is when they request this data from you do you just send them the numbers or do you engage with them? Sending them the numbers in the form of a Excel or PDF report is often the most common solution to making these kind of requests go away, and I’ve noticed that reports from our clients that I have seen, even ones that have been the result of massive campaign expenditure, more often than not take this format.

When you send a client a report on something they have to interpret, more than likely they’re either just going to dismiss it or simply email you back a list of questions, which is actually their way of pleading with you to help them understand their data.

What you are not doing is actually explaining why an banner placement out performed another of instance, or why users who landed on a certain page had a greater propensity to purchase. All of these learning’s from web analytics and reporting are thrown down the drain with the same mistakes to be repeated over again. This leaves no gap for campaign optimization and in the long run leaves you open to an attack from another company who will raise this as a concern when pitching for your clients business.

Aside from this obvious downfall of not engaging clients, you are also not helping clients understand how their business fairs online, you are not offering them insight into business critical questions which their website may be able to answer, or giving them insight into improving their business online. By simply delivering a report to them you’re trivializing reporting, what you’re saying is that they shouldn’t care about reporting.

The question for them is then if they shouldn’t care that much about web analytics reports, why then should they care when you increase their Search Engine traffic by 150% with a 18% increase in conversion.

Engaging with your client in reporting is part education and it does require effort. The effort is with it though, because when the time comes and you spot an opportunity through your web analytics reports you can be on the same page about the importance of a large scale or rapid roll-out of a solution.

A lack of engagement is a massive opportunity loss in the development of your relationship with the client, and your ability to justify additional development expenditure on the website.

The feeling I get is that most Suppliers who have the capability to deliver reporting are afraid that by educating their clients in reporting they will expose the clients to the bad decisions made during the development of the website, or the underperforming area’s in their conversion strategy. And yes, you probably will, but if you get your client to that level of understanding then they will more than likely understand that the web and in fact business in general is fraught with strategic mistakes, and we all know that it’s better to try and fail than to not try at all.

When was the last time you surfed your site through a Site Overlay?

Site overlay is really one of the most underrated tools in Google Analytics and I think several other Web Analytics systems. I know that there are sexier systems dedicated to site overlays like Crazy Egg, but before you go down the path of a paid solution have a go with the Site Overlay in GA (Content>Site Overlay).  I think it often gets a bad rap by serious Analytics fundi’s because it seems like a bit of a basic tool.

For my own part I often get distracted by the nitty gritty numbers, or the custom reports focussed on some of my customers needs, but once in a while I find that a quick trip through the website through Site Overlay brings home a few realities about the website I’m reporting on.

There’s a wake up call when you actually see what users are clicking on visualized in this way, suddenly those fancy buttons the design team spent days designing actually seem like a bit of a waste when you realize that 0.002% of users clicked on any of them.

One of the really nice things about Site Overlay is that it’s also totally accessible to anyone whether they understand data or not, so your designer, your CEO, and even your receptionist can get a basic feel for what people want from your site.

Avatar is being hailed as the savior of cinema, James Cameron himself pretty much expressed that when he accepted the first of what I anticipate will be a boatload of awards.

It may be a masterpiece of style over substance and blue people discrimination debates aside, what the pundits seem to be ignoring is that it took him four years and $500 million to make, in fact each frame alone took 100 hours to render.

The fact is that his effort will not be repeated soon, and yes where the technology will be re-used it will simply not match the expectation of what viewers now expect, and in fact I predict a slew of really crap what should be straight to video releases getting the 3D makeover and being punted as the next Avatar, or at least “From the producers of Avatar…” (Movie Voice).

What makes Avatar such a stand out movie is essentially the concept of the Sublime , and I’m not talking about “this chocolate is sublime” definition of the word. Sublime as a concept in philosophy and the arts for hundreds of years, it’s roots, or at least the roots of it’s first definition originated in large scale landscape painting. The concept in relation to the arts is simple, sublime is the experience you feel when the image you see is greater than what your imagination can conceive. When you cross the crest of the mountain and you look out in awe over the incredible expanse of natural beauty, that is sublime. Not your chocky milk shake, or your cappuccino magnum.

Landscape painters managed to master replicating this experience. You have to remember that people didn’t see all that much visual imagery, and that in a sense paintings were the movies of their day in that they were windows into other worlds. Movie makers consciously and unconsciously have also mastered this skill, although the competition these days is a heck of a lot tougher.

With the array of digital tools for creating sublime experience you really have to stand out from the crowd, and this is where Avatar comes in. The problem with the sublime is that, although it is a repeatable phenomenon, it does with time begin to lose it’s impact. Arguably some art works remain sublime for other reasons, scale, craftsmanship, and the representation of a context in a particular point in time.

My point though, is that sublime in movies tends to date real quick, and the reality is that if Avatar and 3D in general is going to save the cinema going experience it is going to have to keep on getting bigger. Unfortunately this is actually not true of story, but as you can probably see I’m one of those people who laments the death of song writing and you can ignore my further rants on this.

The additional fact is that if 3D is actually all that, then people will buy 3D screens capable for their homes and return to the kind of piracy that is killing cinema. James Cameron said – Avatar is what cinema is about, it’s about shared experiences. Stuff shared experiences, shared experiences talk through the start of the film, get up to go pee during the climatic moments, and frankly smell weird. Cocooning is a trend that is not going to go away and pretty much 90% of technology for the home facilitates this process.

The Opportunity

What cinema and television has to wake put to is that people are not going to stop pirating or downloading media at their own convenience, and that trying to clamp down by banning and fining people is not going to stop them. These actions only create public relations nightmares, swaying public opinion against film distributors by enforcing the perception that they’re fat cats out to get the little guy. You need to remove the need to pirate by giving people what they want: high quality, downloadable media.

As radical as this seems, the option is viable as visual entertainment has far more subtle branding and advertising opportunities than music. You cannot interrupt a song with an ad (or at least you can I suppose, but then it’s called radio) or imbed product advertising into a song (or at least you can I suppose, but then it’s called hip hop). Seriously though there is the opportunity to offer high quality programming for download with global branding and advertising. There are also mechanisms to embed advertising on the fly for regional downloads. Yes this would require a shift in the way advertising for Film and TV is sold, but it’s not such a radical shift if you look at how online advertising is currently sold, segmented and delivered.

Any way you slice it up the opportunity and argument for branded entertainment is actually pretty strong. Especially if you’re losing the battle the way film production and distributors currently are.

Measure the Bleed

This all brings me back to the analytics point of view. It’s really about what cinema can learn from illegal downloading. When I look at the stats on what the film industry is losing to illegal downloads, I wonder if anyone is really tracking what they’re losing it on. Take the larger torrent sites for instance is anyone really tracking what is being downloaded. Yes I’m sure there’s a huge focus on Affiliate tracking, but what about the content. I mean I can’t think of a more incredible way to judge the peoples choice awards for film (tracking downloads), or what the demand is for a film (tracking search).

These kinds of stats are an amazing indicator to the potential eyeballs an ad placed in a downloaded movie can get. It’s also a good barometer to see what people actually want from film, I mean if people won’t even watch a film when it’s free then you gotta know that it’s stinker.

Possibly one of the most exciting apps I’ve seen to come out of the launch of the Google Analytics API is the free application Trakkboard. Basically it allows you to create dashboards for multiple Google Analytics Profiles or Accounts. One of the limitations I experience with Google Analytics is an overview of detailed traffic for multiple accounts. Yes, the dashboard in Analytics does give an overview which is useful but at times not granular enough and to some extent Google Intelligence & Alerts is very helpful in identify periods of large over or under activity.

The issue really comes in when you’re monitoring several accounts and need specific data to display for each separate account in one dashboard. This is where Trakkboard comes in. Trakkboard is a German Adobe Air application which was posted on the Google Analytics Blog some months ago, with it you can create widgets of all you tracked sites with relevant data and time spans applied to each of those widgets. The relevant data includes visits, visitors, pageviews, medium, source, top/flop keywords, pages per visit, bounce rate, time on site, ecommerce, and goals. In addition to this the new release includes a twitter widget which can track up to two twitter keywords.

Trakkboard is no  substitute for Google Analytics, and if you manage one website you don’t need it. For multiple website it acts as a layer in front of Google Analytics, a dashboard where you can view once a day to check all of they are all on track. Obviously if you notice anything out of the ordinary you’re going to log into Google Analytics to troubleshoot it.

Trakkboard is available for download here: www.trakkboard.com remember you need Adobe Air running in order to install it which is available here: get.adobe.com/air/

For other programs that utilize the Analytics API have a look here

When people comment about the success of Google the one factor of this success that they often leave off the list is Google’s understanding of functional design and the presentation of data. What the interface architects and designers at Google seem to understand is that human’s are creatures who more readily interpret visual information then information that needs to be interpreted, like straight tabular data.

The relative ease at which we interpret visual data when compared to other forms of data is a basic necessity for survival, we need to be able to instinctively understand that a hairy beast with 12″ fangs and matching claws is dangerous and should not be approached without a lengthy question and answer session or data analysis on average survival rates on encounters with such a beast.

Now admittedly great design is not something that is readily associated with Google. Their product releases are usually accompanied by a series of expose blogs by industry pundits analysing the functionality of each new feature. They don’t really garner the same kind of passionate reviews that the release of a new i____ does.

The impact of their design is far subtler, but its effect is rather more intense. I’ve noticed that they utilize their design to smooth an otherwise steep learning curve, and this is evident in the speed at which most of their products are adapted.

The seem to have two approaches, the first is to base the product on a known entity, like their search engine, which most users now access with almost hypnotic ease. You can see the basis of the search usability in most of their Trend, or Aggregator tools. The second approach utilized in their more complex tools such as the Adwords or Google Analytics is to make the initially experience deceptively simple, allowing the user to stumble up the complexity as their experience or need grows.

Google Doc’s is in a sense an exception, but that’s because it is after all an office suite for which an existing preferred user accepted layout exists. Why Microsoft chose to deviate from this in it’s latest office version for the PC is actually another reason Windows Vista users should have been standing outside a “campus” in Redmond with pitchforks and flaming torches… but I digress.
The reason that Google is not readily acknowledged for great design, or even usability is because it is actually so perfect in it’s function that it does not interfere with the task we utilize it for, which in essence is the holy grail of utilitarian design.

Now I’m not saying that the world needs to be Google-ised, I firmly believe that there are two sub sections to design, the first is functional design which is all about designing the product around it’s end use and the second is creative design which is designing the product solely to heighten it’s aesthetic appeal. (This is not a new theory. I encountered in Holland and I can’t find a resource for it, but I know it’s not a new idea.)

Anyway, my point is that if we look at Google it is essentially about the first type of design, functional design, and yes where their products are not “beautiful” in the traditional contemporary aesthetic sense, it is absolutely beautiful in the functional sense. The evidence of which is the millions of users who use their tools daily without having to undergo hours of training.

For the most part I find most digital design is really confused about what it wants to be and what function it is actually serving. It’s almost like the designs are overworked by designers who are afraid to let their work be “uncool”, they can’t seem to hold back on the third gradient level, busy image, or unreadable typeface.

Then again, maybe they’re right because I’ve never seen a purely usable functional site win a design award in South Africa, and anyone who has ever tried to sell a website to a client will know how hard it is to do this without pretty visuals.

Don’t get me wrong, I’m not advocating the late 90′s Jakob Nielsen view of the world of the one website interface internet, and obviously there is a large segment of the web which is purely about aesthetics. The point I am trying to make though is that the function of a website, it’s purpose and it’s goals need to be the pre-eminent consideration of the websites layout. Designers need to ensure that these goals are enforced by usability and complemented with design that does not act as a roadblock to these goals.