CaptainDash

Well, you thought you knew the most key marketing trends of 2010 and forward?  There is still one you may have missed, although it is likely to become central in the next quarters to come ; and that is synchronization.

For most adverstisers so far, media spending was (and still is) a mater of advanced planification. There is a goal: increase products output and revenue imput thanks to a media budget that must be carefully sliced and allocated according to the plan. “This year, we will engage with a very creative agency, increase our investment of x% and we will be more focused on direct marketing” is the way the business works.

In ten years though, the internet may have turned this way of working as old fashion if not completely has-been. In fact, online marketing is so real-time and ROI driven that planning and executing according to the initial plan is now perceived as foolish. The notion of dynamic is inherent to the execution mode. Would I detect a better way to increase my ROI, my investment would immediately be reallocated accordingly.

It is surprising that these methods haven’t been adapted to the non-internet media world, that still represents…. 90% of the media spending globally ! Some would say that a billboard campaign cannot feed-back in real-time -and that seems true : apparently a billboard cannot count the number of people who view its ad, and even less the time they spent viewing it.  Same for the tv advertising: watchers cannot click and therefore it is difficult to measure the real impact that the ads have compared to each others.

But there are many reasons to think that if marketing is still so much blind, it is more a matter of incapacity to change its dynamic than a matter of missing information. Actually information is everywhere, it is even overwhelming. A marketing director receives an average of 50 reports directly related to his media spending every… week. Reports from his creative agency -strategic planning would demonstrate that the brand’s consumer are more eager to watch sport tv that go on the internet-, from the Media agency -last campaign has been a success given that the target population reminds of it- from the retailers -sales have decreased over the past weeks- etc. etc. It is not possible to know how many people view an ad on a certain tube station, but it is now possible to know quite accurately how many people were in that station a certain day.

The thing is that it has become quite difficult to deal with all this imputs, it comes in heterogeneous format, at different times, covering different time-slots. It works in a completely opposite system than the media on the internet where there exist some tool that let any marketer access to info such as traffic, ROI, social traffic (of any kind), location, demographic and so on.

SYNCHRONIZATION may then shortly become the new frontier in a business that sizes a mere 3 trillions dollars in media spending. Making this happening is just a question of mind shifting. So far, the marketing is this only enterprise division that cannot really measure its ROI. This may end soon ; and that day the John Wanamaker’s quote “Half the money I spend on advertising is wasted; the trouble is I don’t know which half” won’t be right anymore.

Keyword of the day to remember: Synchronization

CaptainDash is now on a recruiting phase…
We are creating several hiring opportunities, as follow. Oviously we would rather have people who have some interest for the data field, meaning dashboards and datavisualization. Motivation and start-up state-of-mind are obviously key to us.

Graphic Designer:
In the short term, we are offering a trainee position for a young graphic designer, with a good knowledge of the web (he must have already made some web applications) and knowing well Photoshop, Illustrator, dreamweaver… and all the essential web creation tools. This person will work on the main product’s interface as well as on some webdesign (corporate website, etc.).
This position can evolve in a firm unlimited salary contract (and under French law).

Project manager:
We are looking for a project manager with a minimum of 3 years of experience either in the web development or in the internet space at large. Background of handling Business clients account would be greatly appreciated.
This offering regards a full time unlimited salary contract position.

If you have some interest for any of these two offers, please email us at brunowalther at captaindash.com

Datavisualizaton is usually to get things simplier and, most of the time, it does. As a example, to compare the GPB of countries, there are few thing that can match -in efficiency – the interface of Google public data or other interactive dashboards.

But sometimes graphical visualization just screws-up. The best example in our view -is that laughable?- is the following pattern, that was supposed to explain the Afghan situation to military senior officer. When he saw it,  General McCrystal had this laconic comment “When we understand this image, it means that we will have won the War”… He was so furious of it that he then issued a note which forbad his staff to present their memos on powerpoint ! Wouldn’t the visual be about a situation that is increasingly dramatic, it would be of great fun.

So here are are a seven rules that helps to have a dash presentation that is efficient :

1° Don’t put to much information. That is most of the time already enough to compare two series of data -maybe 3.

2° Don’t tweak things with scale.  We all know that the most common trick is to zoom so to dramatically exaggerate a change of say 0,003%. How many journalists have used of that artifice at least once? On paper, we cannot unzoom (or unfold what is bellow…), but thankfully, interactive graphs will more and more allow that.

3° Make the scale obvious. If you really need to focus of the 0,003% change (that can happen for legitimate reasons), make sure that the scale(s) is obvious. That is especially true of you have two different scales  : in that case a good solution is to use two different designs for each data series to clearly show that each one is not of the same nature.

4° Don’t use to much colors : 34% of the World population doesn’t perceive well colors. That means that 1/3 of your readers won’t get the point if you have a two graphics with a red-orange line and a red-violet one…

5°Let people manipulate your data : providing an interactive graph brings more fun and fun is key to convey a clear message.

6° Hide the data spreadsheet : the first thing you draw when you were a kid wasn’t a data spreadsheet, if is more likely that is was abstracted images. Bringing the whole data suit usually get the whole thing boring and doesn’t provide additional information.

7° remove the noise : Raw data sometimes with lots of “noise” that transform you visualization into a worrisome EEG  graph…

Obviously there could be lots of other rules, but these one usually make sense on most of the data-visualization rendering. One of our following piece will be about the visualization that doesn’t follow these rules… Sometimes really funny.

It is difficult to imagine how much data we represent… but crunching the internet helps to figure out. According to IBM, Forrester and a few others, here are the Data that humans create, while paying their taxes, giving calls, listening to music, shooping, working, travelling, etc. I suspect that these numbers are just approximate expectations, but that remains interesting.

We all know that the Countries that compose the OECD are usually some tough ones when it regards Data. Apart the fact that they are the richest places on earth, a measurable difference with other is that they proudly consider that they generally deliver data and financial reporting that are clearly serious stuff (ok, I can see your point, are the Greeks part of it are you thinking ? Ok… yes they are and so are the Portuguese and the Spaniards, but they are the tiniest by far).

Nonetheless, three years ago, the OCDE countries altogether made a quite significant declaration, that they eventually called “Istanbul Declaration” ; Istanbul being then the Capital city’ summit location.

The “Istanbul Declaration” may be a cornerstone of the new democratic era. It simply puts that all OECD countries shall commit to give access, freely, to as much as possible Governmental & Pubic Data. There is not a single mention of how this should happen, nor is it referring to any specific data format, type of content, media format. Anyhow, as soon as 2007 the 32 richest countries on earth decided that they had to let data go…

That’s huge isn’t it? So let try to see what is the outcome of that all, almost 3 years after that Grandiloquante outcome. And well, as far as we can see, it takes time to turn that into concrete action, even for the four bigest OECD countries. A mere six of these have got it operational so far:

United Kingdom: Tony Blair made a big deal setting up the  e-gouvernment since 1997 and it is fair to say that the UK Prime almost invented the concept of e-gouvernment and data.gov (.uk). The beta version of data.gov.uk is online since the 30 September 2009 and by January 2010 more than 2,400 developers had registered to test the site, provide feedback and start experimenting with the data. When the project was officially launched in January 2010 it contained 2,500 data sets. Developers, Universities, officials, and a significant number of subcontractors are working on it and as a result it may well be the largest datagov initiative appart the US one.

United States: during his Presidential Campain, Candidate Obama repetedly said that he wanted to use the Internet to improve the way democraty can work for everyone. When arrived in office, he passed several laws to make it work while his cabinet released the Open Government Directive and granted some funding to several agencies to make it happen. The hassle is that the US Federal government source the data from thousand of agencies which, as a result, remain a challenge to agregate. That is still ongoing work but the outcome is already stunning.

Germany: another Federal country, but a more simple task. Since the government of Gerald Schroeder, Germany has been serious about data, and thanks to whom they can probably pretend to a quite honorable position in the top most advanced data governance countries with their Genesis and Statistische portal which improves its content every day.

New Zeland : in Nov 2009, New Zeland prime, John Key was quite proud to annonce that NZ datagov portal had launched. Thanks to a healthy statistic agency, they are able to bring some of the most consistent and large data fied. As a result there is a surge of economic, sociology, agricultur students who make some university research based on their data.

Australia : it seems that Australia and New Zealand copy each other. The only missing data in the Australian portal are the economic ones. But, the government has commited to provide financial and economic data as soon as this Summer. All data set are accessible via a creative commons licence.

By comparison, France is just getting nowhere. Although Nathalie Kosciusko Morizet, the Gov. official that cover the data topic vowed that France would soon release a datagov portal, nothing has come up yet.

A few others countries have launched data portal but it is unclear that these are operational ones, Austria, Netherlands, Korea and Japan are part of it. China doesn’t seem to have anything near to be ready (apart a vague National Bureau of Statistics) yet but, they are not part of OECD…

In itself, the term Datajournalist isn’t very old. It has been largely popularized during the last American Presidential Election but, more accurately, it is fair to say that it more probably got its first moment of fame with the the initial ‘Gapminder’ shows of Hans Rosling, whom became almost a Data-visualization rock-Star, at the very least in the US since he first performed its demonstration at TED 2003.

His magistral show is said to have been played over than 200,000,000 millions times on the web. Since then, lots of journalists have tried to come up with either breaking news based out of Datajournalism approach or to provide so in-dept understanding in any given topic. The New York Times, with its visualization lab, was the first to come up with a mainstream approach, it was quickly followed by the BBC, the Times magazine or USA Today. On this subject, pearltrees.com comes with an increadible interactive map of the most consistent initiatives regarding datajournalism.

interactive map of the top tv programm in the US (New York Times)

But the founding notion of datajournalist is certainly broader than the principle of big media pushing a new category of information. Its definition is probably even a bit larger than the wikipedia one and could be summarized as such:

1° data inquiring: Reveal some hidden information in the data that surround us and that grow exponentially.

2° Provide interactivity, and let anyone make its own research out of the interactive map. and in a minor tune, let people publicize the result they have come with

3° Provide UGD (User Generated Data): maximize people contribution, help them to bring their own data or participate to the data analysis.

What is interesting is that the next pulitzer guys may not be journalists that were found of interviews in some doubious strip-tease club late at night. Think more of a 21 years old, talented internet tech-core geek: the information is there, all around us, it is just a matter of removing the noise and giving some constituency to it.

An amazing result of that type of contribution, but also of an efficient UCD work, is the MP expenditure scandal in the UK, revealed by the Daily Telegraph. From the very beginning the journalists felt that they would not come up with a breaking news unless they were able to search through a huge mass of documents and data. There requested the help of hundreds of  Internet users to go through 96,000 documents -invoice, reports- and built up a small IT system to monitor that case. It hasn’t come with lots of datavisualization results yet, but it has reach a level of excellency regarding datajournalism, that would deserve a Pulitzer if only that thing happened in the US…

The notion of Datajournalism is a serious play to save the news media industry. you feel it is a too easy short-cut? Well, consider this: one of the reason that makes that journalist aren’t able to come up with unique information is that everyone is a journalist. We, the authors of that post are actually doing the job of a journalist without any press-card. And some of us are able to come with some very relevant insight on any topic.

With the datajournalism, this whole notion of uniqueness of information gets back: expertise -although it is more of a programmer expertise than of a journalist one- is back and investing on a case means setting up a team gathering technical experts, journalist and sometimes ‘basic’ UGD participants…

As explained above, some newspapers magazines have already got some soundable success with datajournalism. it creates an interactional dimension that consumers -readers- clearly perceive and is probably ready to pay for. On the editor side, it brings an additional team value (inquiring journalists, programmers, data experts, graphic designer, etc.) that is not easily accessible to any lone blogger. US and UK newspapers have clearly understood it. In Continental Europe though, there is a lag that is preoccupying. It may be also due to the fact that Governments are just starting to open up the data box…

Talking recently to a Deputy Mayor of Paris -Jean-Louis Messica- we had the positive surprise to discover that Paris has a plan regarding Data. In a initial step, it would be just to progressively give access to a set of raw data and let the community of data-geeks play with it. Just doing that is a challenge, in the sense that it changes to way processes work in their -big and internal- IT department. But the plan is to progressively implement some processes that make that any new data becomes accessible to anyone -except, obviously if it comprehends some sensible information regarding security or personal issues-. You wouldn’t guess how large is the amount of data that a Mayor House can gather: If covers as broad as data of the pollution, traffic, energy, school attendance, nursery allocation, rivers level, number of trees in any given street, street cleaning frequency, etc… No need to say that these data are as many way to increase the democracy transparency, and furthermore the efficiency of the City’s operation departments.

But when it comes to budget and finance lines, it seems less obvious that these would become available soon. As in many organization, the concern is that if people have no way to benchmark the invested amount in a given item, budget transparency might feed-back badly with nothing else than criticism that would blunt the effort of transparency. This aspect might become a reality thanks to the commitment of people like Stanislas Boutmy, who founded the Public Evaluation System, which aims at providing a financial rating for each public office either Region, Department, City, etc. He doesn’t necessarily works for the City of Paris, (they have some agreement with Fitch NetRating), but his approach is one among many which force the public agencies to further understand that it is now time to shift the due transparency to a higher level.

There is still a long way to go before that we get into the democracy 2.0, but we are moving toward it.

We noticed recently that there are lots of people who argue with the fact that data are widdely accessible to the masses. They believe that most consistant data are hiden in very well protected stronghold datacenters. In addition, the common belief is that if the data were accessible, it wouldn’t mean that they are of any use for the mass audience.

These arguments are partially true and we found interesting to try to summarize the main blocking points for a mass usage of data.

We believe that there are three of these bootlenecks :

1° Deal with the absence of standard

Data come of many ways and even more numerous formats. There has been some atempt to standardise data into some international and unified formats, but the less we can admit is that these effort haven’t paid yet. As a result any platform that wants to use eterogenious data has to reformat them prior any consistant usage. That is part of the work the Data.org makes among many other players. There are good reasons to be optimistic there ; not that data are becoming standardised, but there are an increasing number of technologies that can reformat the data so that they can be exploited more easily. On another topic, and on the contrary to the common belief, data are widely available. One can access to almost any type of data about traffic, weather, pollution, demographics, export/import goods (deep detailed), energy, etc. On this topic the article  The Data Deluge From the Economist is quite accurate.

2° Ease the access to data.

Although there are almost unlimited amounts of data transiting through the Internet, it remains very difficult to find them. There is no real search engine for data and no real one stop shopping place for them yet. A few initiatives -among which data.gov is probably the most well known- are currently trying to overcome this issue, but we still lack a Datapedia offering.

3° Provide fun

Data is usually boring. Nothing comparable to typing your friend’s name into google or looking at Twitter. There are generaly few people spending their week-end playing with series of data and Excel. In order to let anyone play with the data, it has to be through a very friendly interface that has more to do with the right side of our brain (the emotionnal one) than the left (the rational one). It mean that it must emerge some news tools, that are revolutionnary compared to the ones that have existed over the past quarter of century. But overall, it means that the era of the data spreadsheet -The Excel Data standard- shall end to be replaced by the era of the  datavisualization, or in other words, of the graphical representation of data.

Unlooking these three bottlenecks would be critical to make data becoming the next revolution phase of the Internet. It would also probably require that the data become more culturally accepted ; i.e. that we understand that it is not because an information is made out of data that it is necessarly a challenge to understand compare, and use. This might be the more difficult issue to deal with, but there is no doubt that it will be overcome very soon.