What's needed to manipulate the world

There are too many coasters in the media world; coasters as in those people that coast by in life. There are far too many media people that are satisfied with adequate results, or at least adequate when compared to industry standards, which themselves are nothing to be proud about, and there are far too few people that work in the planning of proliferation of media that have the almost rabid determination to ensure outstanding delivery.

The following story is a subjective (and thus a possibly polarizing) story about a media planner, let’s call him Aleksi, and his unsuccessful media career before he quit (or rather was fired) and changed to a writing career for a small town newspaper instead. From what we can learn of Aleksi’s experience, it appears to be that there are many traits that are required to be an outstanding media planner (or strategist) without which sub-standard results and even blasé attitudes begin to become the norm. I know for a fact that Aleksi’s attitude became exact that over time, because I knew him very well. The required traits cannot be comprehensively enumerated, as they will always differ contextually (by field, communication channel, or even positional hierarchy) but a few overlapping traits can certainly be identified that should apply to all media people.

The first thing you’d be able to tell about media planning, having been close it it, is that there is an absolute need for statistical and analytical ability, as information based on anything else, in the planning stage, will almost always be conjecture and ultimately the mismanagement of a client’s finances. There were quite a few instances where my flat mate, Aleksi, had done so. The only reason I know this is because I went to university with him, and since I’ve known him, he had never had much mathematical competency. So when he came home depressed after presenting post-campaign reports to clients, I knew his failures had to have been linked to the poor numerical planning he often complained about many weeks or months earlier. The complaints were almost invariable when it got to that stage.

Perhaps advanced statistical understanding is not quite needed, but a media planner simply cannot get by without the ability to calculate probabilities given raw market data, or worse, with the perceived ability to do so only to draw erroneous conclusions from the data. Analytical ability is as important if not more important; pattern matching is an analytical skill and dissecting historical reports in order to decide on future plans, or even future iterations of a current campaign, is a skill that cannot be absent.  Aleksi seemingly had no pattern matching ability from what I could gather about the campaigns he drew together, otherwise he would definitely have been aware of several clear-cut patterns.

A media person requires at least a sophomoric understanding of logical fallacies and memory fallacies in order to know how to address them in such a way that the media campaign benefits from it. People are by nature exposed to a wide range of biases that make them act in certain ways, and having intimate knowledge of these will provide people like Aleksi with ways to refine their plans in order to reach their objectives. Of the hundreds available to us thanks to the study of psychology and human behaviour, examples like the source monitoring error become an invaluable tool if used well enough in a messaging campaign. Behavioural economics is a natural extension of this genre of information that is beneficial to have. In fact, further to the above, curiosity in general is an extremely important trait to have. Aleksi used to be a curious person when I first met him; I’m not sure what happened and I can’t prove there’s a correlation between curiosity an success rates, but I’m sure there is at least a weak, positive correlation.

There must be an ability to manipulate the masses, as well as the desire to do so. Ultimately that is what media is, whether the bottom-line message is a call to action or a rhetorical message, media’s goal is to either make people think something or do something, and that means manipulating their mind states in order to do so.  Successful media is still manipulative, technically, even if it is used to “manipulate” for a good cause. I’ve had many an inebriated argument with Aleksi about this matter – he’s of the opinion that it is not manipulation – and it is a matter we’ve never been able to see eye to eye with. Ultimately, I think, this is something that the media person needs to accept and embrace in order to excel at the task, in order to truly become relentless in the execution of the media campaign they want to create. Again, I can’t prove this about Aleksi, or any other media person, but my theory stands.

There is a need for the ability to recognize and dwell on one’s own mistakes without having the ego getting in the way. Inevitably and invariable, mistakes in either a media strategy or the execution thereof will be made, and it will cost money. However, these mistakes offer the chance to be learned from, and they require the kind of person that is ready to both acknowledge even silly errors, and then scrutinize them (in varying degrees depending on the mistake of course) to try to limit future reoccurrences. This sadly, is something that an outward-focused Aleksi was not successful at doing; his ego was too fragile for him to inwardly prod himself with such circumspection, even though the circumspection was about his work and not his own character. Media, like most other things in life, is also a process of trial and error, but without trial and error this is no progress.

Different executional approaches need to be taken not only for different channels, but also for different campaigns as well as a variety of varying factors. It is important to test different approaches to ensure the best ones are being taken, but it is even more important to be able to jump the line in order to get there. This is done by having domain knowledge of various audience types by platform (and many, many other factors) or otherwise having access to historical data that is both filterable and mutable. One way to come about such data is to introduce processes establish a multivariate testing standard or rule. Things like channels, copy and creative work all need to be tested against to uncover underlying patterns, should there be any, to tap into for future media campaigns. Aleksi does like to try different approaches, which is great, but he becomes fatigued far too quickly. This approach is a monotonous one and cannot be discarded once someone cannot bear it anymore. That was his fault.

Depending on who is asked, media planning will be described as an art or as a science, and the latter is the more apt category, or rather is some semblance of science (something that Aleksi and I can both agree on). Unlike art, and as with science, trial and error is the order of the day, even for a media campaign that has already begun. More so than that, it is important to discard of elements that are not working mid-campaign, including entire media channels that aren’t ultimately beneficial to the campaign objectives.

There is certainly a need to be able to work with creative people in the planning of the message copy and the more traditional creative elements (the visuals, the recordings, the interactive elements etc.) to ensure congruency with the selected channels. Actually, it is not only to ensure congruency but also to ensure that the creative work is fitting to the strategic theme of the media campaign. This ability is not as readily measurable as some of the others, but is an inter-personal communication skill that irrefutable has to be part of the overall skillset of the media person. Aleksi is quite the social person, and has great inter-personal communication abilities, so I’m sure he did this well enough.

The media person needs a very strong sense of integrity, as well as direction. This is a crucial trait as it will ensure that even if one does develop fruitful relationships with media owners that generally offer discounted rates for purchases, the integrity toward the client (or rather the objectives of said client) will ensure that one does not feel obliged to spend media budget with said media partner in order to upkeep that relationship, and certainly not to do so in order to receive luxurious supplier gifts at the end of the year, or whenever they may come around. Sadly, my friend was and still is too venal to have said no most of the time.

The same integrity would also be required to stand up against a client (respectfully of course) should they be misguided in the finer requirements they ask for, should the requirements not be congruent with the overall objectives – these things do happen (but alas, knowing Aleksi, he almost certainly would not have stood up to them). It cannot be seen as a bad thing to tell a client that they are wrong in certain circumstances, and in fact it is a bad thing to not do that very thing when the need arises for it, should the media person have the sufficient evidence to back their claim.

Decisiveness is crucial as anything short of it will almost certainly result in a waste of resources – either time, or squandered money on decisions that have been half-made. Budgetary constraints are unfortunately (for media agencies) very real and very often, in fact more often than not, the ideal media campaign in the mind of the planner is only that – idealistic. The limitation of money in any single campaigns adds pressure to the planning process, and begins to beg the question of big or small? The question asks whether the media go through a few large channels, or a larger number of small ones, or rather a hybrid of the two? Channel size should not be of much importance this early in the process, but sadly in far too many cases, it is. However, regardless of when this question is asked, if it is, then the person asking the question will have to have insight into things such as the data about these channels as well as tributary data about things like what type of audience segmentation would deliver the best results. These decisions can only be made with the right data, and the right inferences from market research (about demographic propensities tangential to the insights that said platforms touch on) and these decisions need to be made with a certain degree of definiteness, lest time is wasted on the choices that could be better spent contemplating other aspects of the planning process.

Finally, one needs an amount of luck, or serendipitous luck. Generally luck is not something in one’s control, but serendipity is in that the more a person exposes himself or herself to, the better the chance that they will be exposed something timely or something they otherwise would not have known. This so-called luck surface area can apply to a range of things in the media world, including learning about new media channels, platforms, media techniques, market trends, newly adapted nomenclature or communal attitudes in sub-communities within platforms already being addressed etc. This information is not something that is generally aggregated en masse, and thus is not easily accessible outside of a team. It is important then, for media people to be actively exposing themselves to as many stimuli as possible (but not so much that burnout becomes a risk) in order to achieve this serendipitous luck that may turn out to be beneficial in upcoming campaigns. This ties up with the required level of curiousity that I mentioned earlier, and the amount that Aleksi has is not, and was not sufficient.

A lot of requirements were introduced here, and many of them were introduced through Aleksi, a media planner that doesn’t really exist. Despite this fact, I feel very strongly about many of the opinions stated here and their need in the personalities of media planners in order to ensure the best execution of client requirements in the world of media marketing and communication – after all, is the level of expertise required to succeed not what clients pay for?

The story behind the interactive tap

Very recently we created a real life open tap that was hooked up to an Arduino kit that is connected to the Twitter API and live streams onto www.closethetap.com - The tap is running (in a closed system so no real water gets wasted) and when people tweet with #closethetap the Arduino kit is programmed to slowly close a valve in the tap’s plumbing using a step motor and when it hits 10,000 tweets then the tap will close. We did it for International Water Day as water as we wanted to get people to change their behavior towards water.

Close the Tap - The development journey

The first thing we had decided after hearing about the ‘Close the Tap’ concept was that we needed to make the website part of the project ‘the right way’. The various challenges we faced were:

1. Match it to the conceptualized design while keeping HTTP requests and page load time to a minimum
2. Making it scalable for various viewports
3. Having it supported on IE7+
4. Live streaming twitter feed that falls into a ‘tweet basket’

Match the Design

The design wasn’t web-friendly per se, as it had a couple of strangely positioned gradients and a few textures, including quite a large repeating background texture. When it comes to textures and gradients, we try our best to replicate in CSS since we are unable to include repeating images in the global sprite.

We ended up using CSS3 gradients instead of a giant background-image. We also created a png image gradient to use as a fallback in case the browser doesn’t support CSS3 gradients.

background-color: transparent;
background-image: url(../i/container-gradient.png);
background-image:  linear-gradient(to right, rgba(255,255,255,0), rgba(255,255,255,0.2)  5%, rgba(255,255,255,0.5) 15%, rgba(255,255,255,0.5) 85%,  rgba(255,255,255,0.2) 93%, rgba(255,255,255,0));


/* We have left out the vendor prefix versions for readability purposes of this article. */

We tried to do this as much as possible. Browsers that support the CSS3 gradient don’t load up the image saving an HTTP request and give the developers full gradient control.

Scalable for various viewports

Everyone is pretty familiar with media queries nowadays. We used the HTML5 Boilerplate as a kick-off point for this project, and generally most projects we work on. If you don’t already know, it is a nice HTML framework which comes packed with best practices and useful elements. There are HTML/CSS comments for pretty much everything they do which link to conversations or in-depth articles regarding the subject. However, the #CloseTheTap site had to be scalable. The biggest problem here was the Ustream iframe which isn’t able to change size after it has been loaded without being completely reloaded. The site settled on only loading up the iframe if the viewport is larger than 960px, otherwise an animated gif is loaded up instead to simulate the live stream for mobile devices - a compromise, but a choice we ended up sticking with.

Supported on IE7+

Oh IE7, you are definitely the new IE6. The two biggest problems we had with IE7 on this project is the lack of pseudo-element support and content overlays on iframes. We often use pseudo-elements for little images on button-type elements or for lists as customized bullets. We ended up using spans for this instead - something we prefer not to do.

One of the more annoying problems was that IE7 had big problems with the content overlay on the iframe, as mentioned above. We were forced to load up a flash version of the video for IE7 - definitely sub-optimal, but a workaround nevertheless.

Live streaming twitter feed that falls into a ‘tweet basket’

This was a very fun application to build. We had not previously done something like this before. The way to achiev the twitter fall effect was quite simple:

1. Pull the latest 4 tweets from the twitter JSON file and add them to a ‘tweetHistory’ array.
2. After about 15 seconds check the JSON feed again and pull the latest 4 tweets.
3. Cross check those 4 with the tweets that already exists. If they don’t exist, add them to the list and remove the oldest.
4. Repeat

As you can see, it’s really quite a simple concept and process was really fun, and challenging to create at times. There is inevitably a Twitter plugin out there that may do something similar to this, however, having done it from scratch taught us a whole lot and overall, the front-end team is glad we did it. Perhaps next time you're thinking about including a JavaScript plugin, rather do it yourself and see what is entailed.  How else are we meant to learn?

UStream

We were initially going to try to stream the tap to the site but went with live streaming it through a provider instead - why worry about traffic loads when there are established companies out there that will manage load for you and all you need to worry about it getting the stream to them? There were several providers available and we were initially leaning towards Justin.tv but embedded videos have no option to disable adverts with live streams, and although UStream are a tad pricey, we ended up going with them, and are sufficiently happy with the streaming.

Conclusion

All in all we're pretty glad with how the site turned out, if we were to do it again with more time, we would definitely clean up the JavaScript and make it more efficient, and perhaps create our own Twitter widget JQuery plugin. Aside from that we all had a lot of fun and can't wait to work on something similar in future. Perhaps the next write up will be about how the tap was made.

Also, go check out www.closethetap.com and help close the tap!

The ultimate skeptic

Cogito ergo sum brings us a very interesting problem. Rene Descartes posited this "I think therefore I am" idea which popularized metaphysical solipsism - the idea that no reality can exist outside of your mind or your mental state. As per his notion, you cannot prove to me that you exist, I can only know that I exist, because I experience myself existing. This statement is problematic and thus seen as sophomoric because it is untestable.

However, before discarding it, I'd like to extend it a little bit by delving into the realm of the hypothetical. In a hypothetical world where metaphysical solipsism could be true, it could be either one of two values - true or false. If it is false, and therefore impossible, there is nothing further to discuss. If on the other hand, it is indeed possible, then let's imagine that it is the case. I, the self, am the only thing that exists and everything else is an extension of my mind, or rather a representation of it. What happens when I die? Can I die? How could I possibly know that I've died and would the ceasing of experience signify my death (and thus the death of all existence) or would my death cause my experience of the world to cease?

I don't think those are questions that can ever be answered concretely, but there is another more pertinent question back in the converse. In the hypothetical world where solipsism is false, but in the mind of an actor, he absolutely believes it to be true, is there any way to prove to him beyond a doubt that his theory of solipsism is crazy? If you, as a devout solipsist, die and go to heaven (assuming heaven and God exist), and you see God, and he greets you but instead of reciprocating the greeting, you charge him with your idea of disbelief, how could he prove to you that he's real? The God problem here is that to the ultimate skeptic that is face to face with this God, there is no way to absolutely convince him that the current situation is reality.

In fact, even beyond the skeptical solipsist, even if you just take an extreme skeptic, how can he be convinced that he's in heaven and not in some long-winded dream? Sticking with Descartes' original summary, the only thing you can be certain of is your existence, and everything else including experience, logic and knowledge can be false. This means that solipsism could be self-defeating in that your logic of your solipsist-driven world view could be false or illogical. Again this doesn't prove that solipsism is impossible, only that it could be impossible under certain circumstances, but it does leave open the possibility that you could one day face a God only to doubt his very existence, should you be religiously inclined. However even if you don't believe in solipsism, you could still be the ultimate skeptic.

The API in layman's terms

These days, most web services have an API. The word is thrown around a lot and a lot of non-technical people don't really understand what an API is. Firstly, know that it stands for an Application Programming Interface. The interface bit should tell you that it allows for some sort of interaction, which is right, it allows a piece of software to interact with another piece of software. An example would be an application you use that lets you see your tweets from Twitter, so I'll use Tweetdeck for this example.

The Twitter platform has this API that defines a set of rules or methods about how other software can interact with it to get specific information (such as your tweets). Your piece of software communicates with this interface by "talking" in the different methods. The information is accessed and sent to your application where you access it, instead of using the main Twitter website.

Consider the following analogy:

You're in a certain car, and you'd like to get from your house to the tax offices to submit your tax returns because your internet is down and you can't do an electronic submission. The engine, wheels and axles (for the sake of simplicity) get you to the tax office, but it can't do so without certain rules - You need to drive it on a road. Let's remember that the tax building is a government building (we consider this because the government builds and owns roads), and they provide you with an interface to get to them - the road architecture.

You need this interface to drive to the government building, and you need to follow the rules they've set up when they built it (stop signs, speed limits etc.) to use it. Having a client (your car) and an API (the road), you have an alternative way to send your returns that doesn't consist of using the internet. Remember that there are many cars on the road that can be used to get there.

Another thing worth remembering is that APIs always expose data, as in everything is always available and you use methods (certain commands) to ask the API what that data is, and replicate it in your client. Some APIs require access tokens (sticking with the analogy, only buses are allowed in bus lanes) to access certain data methods.

Also, if there are a lot of cars on the road, the traffic is slow. If lots of clients are accessing a web service's API, there can be quite a bit of lag.

I hope that helps you understand how APIs work.

Existence precedes essence

Philosophically speaking, the proposition that existence precedes essence is a central claim of the existentialism branch of philosophy. This view takes an opposing view to the general philosophical outlook that the esence or nature of a thing is more fundamental than its existence. Jean-Paul Sartre lived by the phrase that existence precedes essence - we first and foremost must exist and only then do we determine values of existence.

This phrase is just as well with projects and start-ups and is congruent with iterative strategy of many start-ups; push your MVP out there and then you change and update it as time goes on rather than waiting for it to become optimal before launching.

This subject has been covered numerous times by other people, but consider this a refresher that fundamental life and consciousness follows the same pattern of bare existence before perfection as most start-up strategies.

Call it a community

There are tons of niche social networks available, and a lot more than are still popping up. Here's just one list of a few of them, and there's hundreds more. If you are an entrepreneur that is planning on building one or growing one, you should know something that might help your endeavor. People have become immune to the phrase "social network" and don't really give a shit about joining your new one just because it has x feature.

What is a niche social network? It's a community of people that access the platform around a specific topic, maybe to learn about it, maybe to share - whatever. That's irrelevant. The point is that they gather, they communicate, and they share - this makes it an online community. What's better about the word community? It's inert. It doesn't have that irritating web2.0 buzzword egoism attached to it. The word also portrays utility, so people actively explore them. It's also been around much longer.

So here's the thing, stop calling your niche social network a social network and call it a community instead. If your recipe-centric website like www.recipefy.com is about cooking and sharing recipes, call it a community because it is exactly that, and people won't feel defensive about joining it. They won't consider it as much of a burden signing up just to be on yet another stupid, fad social network.

Below are two graphs comparing search trends of "social network" vs. "community". The red line is for "social network" in the image directly below.

Sometimes missing out means not missing out

The title is a little less paradoxical when put into perspective. Sometimes missing out on one thing means not missing out on something far more important - something that is beneficial to everyone. I'll keep this brief by being specific. Sometimes being a geeky, loser nerd that can't get laid with a (or any) girl (or boy) means that you end up going on to make something really awesome that everyone benefits from, and end up getting laid with somebody else down the line.

The obvious example is Zuck missing out with his girlfriend and building something cool to get laid with other, new girls. From a utilitarian perspective, this was the far more beneficial outcome as many more people have gained from his 'get-laid' product. Whether or not this specific story is true is irrelevant because this scenario is prevalant in the engineering field and can probably be applied to many, many similar stories. Of course sex can also be a metaphor for any other pursuit of desire.

TL;DR - If you're an engineer and can't get laid, focus on your speciality and keep on at it. It will get you laid and you'll make the world a better place.

Top decision making behaviours that are consistently irrational

In his book Against the Gods: The remarkable story of risk, P.L. Bernstein pointed out the top 10 human decision making behaviours that people continuously making, in error, without realising it. It's a good thing to familiarize oneself with these, as well as other cognitive fallacies, both in order to avoid them, and as an entrepreneur, to know how to best capitalize off of them (in an ethical way, of course). Below are 6 that I thought to be quite pertinent:

1. People overweight new information at the expense of existing information. (The pendulum effect)
2. People are much more sensitive to negative than positive stimuli. (Think time constraints, the risk of foregoing beneficial experiences etc.)
3. People are naturally risk-averse and will generally choose a certain gain over a high probability of a superior return.
4. People are much more sensitive to a loss than a cost. (People would rather keep a cost in order to avoid a loss even though they only differ in terms of perception)
5. People recoil from perceived risks associated with radical innovation on the grounds that it is safer to fail conventionally than to succeed unconventionally.
6. People make decisions with assiduous reference to the information that is available to them, even though they may be aware that it is woefully incomplete.

These points will very well apply to the consumer behaviour model, as well as the communication strategy that you choose as an entrepreneur or start-up founder. Here's a link to his book on Amazon if you're interested in more.

A recap of threewords.me

This morning (SAST), Mark Bao disclosed that his site (threewords.me) recieved 227,000 page views today, of which 50,000 were uniques. At that time the site had 8,285 users (How good is that conversion ratio, hey?). This was all from organic growth (predominantly from Facebook, Twitter and Tumblr) and one hell of a viral coefficient. A couple of hours later the story was picked up by news sites like TechCrunch, TheNextWeb, as well as Digg and plenty of various forums and you could imagine that traffic grew even more.

The traffic got so intense that many requests were timing out at a stage, but the issue was dealt with and the site seems to be keeping stable for a quite a while now. The traffic should wane signigicantly after several days as the novelty wears off, but it might not - no point in speculating. The interesting part for me is one of the insights behind the growth, which shouldn't come as a surprise to anyone.

People are vain. They love being talked about, mentioned, or described. I've got a good feeling that if Mark looks into his Google Analytics he'd see a good share of the total visitors being return visitors with a longer average page view time than from the first time visitors (If he'd set up the tracking goals beforehand). These would be the people who've set up their URL and have later returned to see what their friends have written about them, most likely several times.

As he already mentioned on Quora, Mark's site has a double viral loop, where after they add three words, they themselves are requested to sign-up. Sticking to the premise that people are vain, a lot of them did (see conversion numbers above). Admittedly vanity couldn't have been the only reason and isn't provable to have been (curiousity, for instance could have been a leading factor), but vanity must have played some part.

The conjecture aside, and back to the analytics, I'm of the opinion that this launch could be used to come up with a theory or two about mass visitor behaviour to these types of free applications. The site has chartbeat analytics installed - What I'd really like to see is a longitudinal overview of first time visitors vs. return visitors from launch until a few days after the traffic frenzy subsides. This would shed some light (albeit non-emperical) on how users act when it comes to free, simple applications. A very loose hypothesis could then be develped and used as the basis for an experiment during the next fad (Sorry Mark) whose numbers would also be tracked for comparison.

Who knows, maybe after a pattern emerges, somebody could propose an optimal monetization model for these sorts of things, or even a timing model on where along the timeline to start adding advertising without negatively affecting the viral growth.

Guerrilla media

Guerrilla marketing relies on time, imagination and energy executed in a creative way as opposed to big-budget, strategy-based campaigns. The objective when using guerrilla marketing is to get the best results from minimal resources, and generally to creative a disruptive buzz about a product or service.

This type of marketing is more fitting for smaller business and start-ups because it could run the risk of diluting the core marketing message of larger brands if done too often or inaccurately. This method of marketing can also be extended to media, to create a ‘guerrilla media’ aspect of a campaign.

Most marketing planning is done with tools that allow strategists to determine psychographic details, demographic details, and even purchase-intent estimates for their target audiences. This is achieved by data gathering as well as statistically backed projection algorithms for the latter. A tool that most marketing strategists don’t use is the business intelligence methods that determine short-term trends in cyber networks. An example of this is the measurement of breakout topics that are discussed in clustered communities around the Internet. Even more so is the measurement of user generated media content that augments these discussions – as is regularly seen propagated in the comments sections of larger technology news websites. 

How business intelligence can be pragmatically used with a marketing execution in mind is to use it for trend scraping. Imagine a process whereby a system does period-defined text and data mining - for images, unless it is instructed to read alt-tags for images to determine their contexts – and combines it with predictive analytics to determine how long the trend might last, as well as with what frequency and in which communities.

This data could then be compiled into a structured manner so that marketing or media planners with licenses to access this data could do so based on their required parameters (specific types of channels, channel exclusions, types of audiences based on complementary Nielsens statistics etc.) and would be able to import concise lists of viable guerrilla media options. The lists would provide them with trending topics or even memes, should their brands be non-financial or with enough leeway in their brand personalities to be tongue-in-cheek, which they would then be able to tap into with their own media executions.

The benefit of this sort of approach is that, if executed in congruency with the trend in a manner that is not too commercially intrusive, it could blend into the current disposition of the specific community. Much like large brand communication is defined by a corporate identity (CI) which keeps all marketing and communication activity within a tightly defined field, trend-specific media that uses the same style as the user generated content will allow for smoother integration and a higher propensity for funnel conversion due to familiarity. At the very least, the brand could receive praise for joining in the community’s creative efforts. The exception to this would be larger communities that have a tendency for hostility, which would be filtered out in the aforementioned BI process.