About Frédéric Filloux

http://

Posts by Frédéric Filloux:

The New York Times KPI’s

 

Here are numbers lifted form the NYT’s Innovation report (see last week) and other sources. 

Most of The New York Times’ reach comes from its digital audience. Regardless of the metric, viewers on desktops and mobile are crushing print readers.

321-1 - 450

Sources: ComScore for the monthly uniques (US only); internal count for the home page views per 24 hours period and Gfk MRI based on net weekday & Sunday readership, Fall 2013 survey.

321-2 - 450

321-3 - 450

In theory, the Times can get rid of print. Digital revenue far exceeds the cost of running the newsroom, which amounts to $200m a year for 1300 writers and editors. Even if you add $20m for the 200 technical staff needed to run digital operations, and even 30% more for overhead, sales, marketing, and support staff, the result would still be a substantial profit  – but would advertisers come in the same way for a digital-only product?

321-4 - 450

The ad market seems to reward quality journalism over aggregation and listicles: The NYTimes.com monetizes itself three times better than Business Insider and nineteen times better than BuzzFeed. For this graph I simply divided annual advertising revenue for each media by the number of monthly users: 30m UVs for the NYT, 12m UVs for Business Insider according to ComScore figures quoted in this 247wallst story, and a revenue estimated at $20m by Reuters. (Had I used a 25m UVs assumption, BI’s ARPU would have been only $0.80 per visitor and per year).

321-5 - 450

The Times is known to have invested a lot in its digital subscription system (760,000 subs to date). It turns out to have been worth every penny. For those who doubt the paid model’s efficiency, The New York Times provides a great blueprint for quality media.

–frederic.filloux@mondaynote.com 

 

Time to Rethink the Newspaper. Seriously.

 

The newspaper’s lingering preeminence keeps pulling legacy media downward. Their inability to challenge the old sovereign’s status precludes every step of a critically needed modernization. (Part of a series).  

This column was scheduled to appear in the next two or three weeks. Then, on Thursday, the thick Innovation report by an ad hoc New Times task force came to the fore. Like many media watchers, I downloaded its 97 pages PDF , printed it (yes) and carefully annotated it. A lot has been written about it and I’m not going to add my own exegesis on top of numerous others. You can look at the always competent viewpoint from Nieman Lab’s Joshua Benton who sees The leaked New York Times innovation report as one of the key documents of this media age. (Other good coverage includes Politico and Capital New York — I’m linking to the NYT tag, then you’ll have all the stories pertaining to Jill Abramson’s brutal firing as well).

320-Innovation_full

This report is important one for two main reasons:

– The New York Times is viewed as one of the few traditional media to have successfully morphed into a spectacular digital machine. This backdrop gives a strong resonance to the report because many news organizations haven’t achieved half of what the NYT did, whether the metric is the performance of its digital subscription model, or its achievements in high-yield advertising – all while keeping its impregnable ability to collect Pulitzer prizes.

– We rarely, if ever, see an internal analysis expressed in such bold terms. Usually, to avoid ruffling feathers, such reports are heavily edited – which ends up being the best way to preserve the status quo. Even more, mastheads tend to distance themselves from endorsing conclusions coming from the “management crowd” – a coldly demeaning phrase. But, it the Times case, the report was expressly endorsed by the top editors (Abramson and her then second-in-command Dean Baquet who now leads the shop.)

Let’s then focus back to this column’s original intent: Why reinvent the newspaper, quickly and thoroughly.

Until last week, the reference on the matter was an email sent in January 2013 by Lionel Barber, the Financial Times editor (full-text in the Guardian), in which he sets a clear roadmap to shifting resources from print to digital:

I now want to set out in detail how we propose to reshape the FT for the digital age. (…)

[We] are proposing a shift of some resources from night work to day and from print to digital. This requires an FT-wide initiative to train our journalists to operate to the best of their abilities. And it requires decisive leadership. (…)

On unified news desks, we need to become content editors rather than page editors. We must rethink how we publish our content, when and in what form, whether conventional news, blogs, video or social media.

 A year later, key numbers for the FT are impressive:

– A 2013 profit of £55m ($92m, €67m) for the FT Group (which includes the 50% stake Pearson owns in the Economist Group); that’s an increase of 17%, while sales are slightly down by 1% to £449m ( $755m, €551m)

– 415,000 digital subscribers (+31% in one year) who now account for two-thirds of the FT’s total audience (652,000 altogether: +8%, including a staggering 60% growth in corporate users at 260,000)

– A rise in digital subscribers that offsets the decline in advertising now accounting for 32% of FT Group revenue vs. 52% in 2008.

– For the first time, in 2013, FT digital content revenue exceeded print content.

The FT might be on sale – but its management did quite well.

Echoing Lionel Barber’s view of resources reassignments are the equally strong terms from The New York Times’Innovation Report:

In the coming years, The New York Times needs to accelerate its transition from a newspaper that also produces a rich and impressive digital report to a digital publication that also produces a rich and impressive newspaper. This is not a matter of semantics. It is a critical, difficult and, at times, painful transformation that will require us to rethink much of what we do every day. [page 81] 

Stories are typically filed late in the day. Our mobile apps are organized by print sections. Desks meticulously lay out their sections but spend little time thinking about social strategies. Traditional reporting skills are the top priority in hiring and promotion. The habits and traditions built over a century and a half of putting out the paper are a powerful, conservative force as we transition to digital — none more so than the gravitational pull of Page One. [It] has become increasingly clear that we are not moving with enough urgency. [page 59]

The newsroom should begin an intensive review of its print traditions and digital needs — and create a road map for the difficult transition ahead. We need to know where we are, where we’re headed and where we want to go. [page 82]

These quotes from a news organization that never gave up on great journalism will be helpful to those who desperately struggle to transform newsrooms. It is also a plea for the necessity of dumping the obdurate print-first obsession:

– It precludes modernizing the recruiting process as journalists are still too often picked for their writing capabilities while many other talents are needed.

– It limits audience development initiatives. In today’s print-oriented newsrooms, most writers and editors consider their jobs done once the story is filed in the CMS (Content Management System). Unfortunately, in every fast-growing digital media outlets such as Buzzfeed, The HuffPo, Politico, Quartz, Vox Media, now part of the competitive landscape, throwing the story online is actually just the beginning. The ability to cause a news item to reverberate around the social sphere is now as important as being a good writer.

– As stated in the Times report, convincing the masthead on the mandatory resource-shifting in only part of the journey; most of the transformation’s weight lies on the shoulders of the rank and file in the newsroom.

– At the NYT as everywhere else, the old guard (regardless of age, actually), is the main obstacle to the necessary rapprochement between the editorial and the business side. For instance, by rejecting the idea that Branded Content would greatly benefit from the newsroom expertise (although everyone agrees that a news writer should never be asked to write advertorial), or that a conference is indeed an editorial initiative directed to a valuable audience segment, such conservative postures are actually shrinking the company down to its most fragile component.

– The same goes for the analytics arsenal. I heard scores of examples in which newsrooms call for more dashboards and indicators, but seldom use them. Editors should be supported by tactical analytics teams (including at the editorial meeting level) that will provide immediate and mi-terms trends, as well as editorial decision-making tools.

One of the most difficult part of the transformation of legacy media is not addressed in the Times Innovation report nor in the FT’s exposé. It pertains to the future of the physical newspapers itself (the layout of the Times remains terribly out-of-date): How should it evolve? What should be its primary goals in order to address and seduce a readership now overwhelmed by commodity news? What should be the main KPIs (Key Performance Indicators) of a modern newspapers? What about content: types of stories, length, timelessness, value-added? Should it actually remain a daily?

(To be continued…)

frederic.filloux@mondaynote.com

 

The Ripple Effects of Disruptive Models

 

Last week, we discussed the impact of services such as Uber or Airbnb. More broadly, no sectors is immune to major overhauls. Today, we’ll have a look at the impact of Disruptors. 

Nested in Paris’ Le Marais neighborhood, a clever incubator/think-thank called TheFamily, made its mission to chronicle the digital transformation of our society. Largely inspired by the iconic Ycombinator incubator, TheFamily funds and provides all sorts of services to a hundred plus startups. But it also wants to rattle the establishment with an activist posture. Paraphrasing the “Barbarians at the Gate” book title, the incubator hosts a conference series titled Les Barbares Attaquent (Barbarians On The Attack) that examines all the sectors to be impacted by the digital tidal wave.

The latest event (#18) featured the book industry. Prior to that, human resources, retail, luxury, housing & construction, health, transportation, education, garment industry, consulting, insurance, finance and other sectors were dissected by TheFamily partners and guest speakers. Each time with a larger attendance.

One of the founders, Nicolas Colin, recently made headlines when his blog post (fr) denounced the notoriously archaic parisian taxi lobby (see previous Monday Note), triggering a lawsuit from Nicolas Rousselet, the owner of the main French taxi company G7. (Nicolas is the son of André Rousselet, himself one of former president François Mitterrand’s favorite oligarchs, anointed TV mogul in the late Eighties). By suing the blogger, Rousselet Jr. wanted to shut down any criticism of his company’s unrelenting conservatism. In fact, he completely underestimated the reaction of the French digital multitude that rallied en masse to support the blogger (and the media La Tribune, that republished the infamous post.)

This little Gallic tale illustrates the split between the old and the new economy. It could have happened in Brussels, Berlin or San Francisco where lobbies furiously oppose the rise of Disruptors that threaten transportation or short-term rental housing — among other things.

Before we go further, let’s look at the engine of the Disruptors’ phenomenal growth. It can be summed up to one phrase: unprecedented access to capital.

When it comes to technology, Uber or Airbnb are not rocket science. The platform and the algorithm needed to efficiently match supply & demand have been indeed brilliantly implemented, but there is no need beyond off-the-shelf technologies to set up the whole thing. By contrast, when Google started in 1998, it did stretch the limits of the technology of the day (networking and computing power); as for Facebook, despite the relative crudeness of the original concept, it had to deal very early with scalability issues. Actually on its very first day, Mark Zuckerberg’s hottest girl matching system (how nice) crashed Harvard’s network. No such headache for Uber or Airbnb who rely on proven technologies: cellular network, mapping, databases, LAMP-based softwares. As shown in the following three graphs, funding has been equally abundant for these areas:

319-google 319 facebook 319 uber 319 Airbnb

Not only have investors poured big money in Uber and Airbnb but they did so extremely fast, boosting the valuation of these two companies to staggering levels. Since there is very little technology involved, where did the money go? Mostly to market share acquisitions, the only way to leave the competition in the dust for good. Take Airbnb: in just one year, its number of listed spaces grew more than doubled to 500,000 listings in 33,000 cities and 192 countries. Its $10bn valuation puts it head-to-head with the giant group Accor that operates 3500 brick-and-mortar hotels and 450,000 rooms.

In these new models, the American venture capital ecosystem is acting as a weapon of mass domination. When Uber collects more than $300 million in VC money to expand in 100 cities worldwide, its London-based competitor HailO got “only” $77m and when it comes to the French LeCab, it only raised €11m ($15m). It shows how anemic the French system is when it comes to funding its startups; instead of patting the registered cabs sector in the back with demagogic promises, the successive digital economy ministers would have been better advised to act decisively to stimulate access to capital.

Still, the European way of resisting these new models won’t last for long. To be sure, in Brussels, the ill-named “ministry of mobility” decided to simply forbid Uber-like system; in France, the resistance is more messy when hundreds of yelling taxi divers blocked main streets and airport accesses. But grass-root movements are likely to morph into a more anglo-saxon-like lobbying, with highly paid professional hired to defend special interests.

Consider this: between 1998 and 2013, the amount spent in Washington DC alone by various lobbies has grown x16 in constant dollars to a staggering $3.23bn. Today, tech firms are the fourth contributor after pharmaceuticals, insurance and oil & gas: when a big pharma spend $1.00 to influence lawmakers, tech companies now spend $0.63 and the gap is closing.

Why am I mentioning this? It’s because the capital raised by Disruptors will inevitably find its way to effective lobbyism in Brussels (at the European Commission), and eventually in Paris or Berlin.

Disruptors’ lobbyists will argue that new urban transportations system and peer-to-peer housing rental do more good than harm in the community. And for the most part, they might be right. Sharing cars in congested cities via system such as RelayRides definitely makes sense from a environment standpoint when any individual car stays idle 95% of the time. A survey conducted by UC Berkeley (pdf here) on a 6,000 San Francisco residents participating on car-sharing system revealed a drop of 50% in the personal car ownership (the auto industry might not like it, but our lungs will.)

On the economic side, there is no shortage or arguments either. Terminating the paid-for license system (the so-called Medallion) would free €3 billion in Paris, and $10 billion in New York, sums now immobilized and promised to an inexorable deflation. In times of raising inequality, maybe it is not such a bad idea to let people make extra money by renting their apartment or their car — with limitations, of course. To put some figures on the idea: an Airbnb host in San Francisco is making $9,300 per year on average by renting his/er property 58 nights. As for those who makes their personal car available for sharing though RelayRides, they make on average $250 a month.

As for the hotel industry, evidence shows Airbnb’s growth to have very little impact. According to the Boston University School of Management, in the state of Texas, a growth of 1% in Airbnb supply translated into only a 0,05% decrease in the revenue of 4,000 hotels surveyed, while a single percentage point of increase in the supply of regular hotels rooms translated into a 0.29% decrease — 20 times more — in Texas hotel revenues. Of course, cheap hotels are more impacted than the local Hyatt.

Between consumers who are voting with their smartphones, enjoying Uber or Airbnb, and the fact that Disruptors are undoubtedly beneficial to the community, regulators and lawmakers will have hard time defending the status quo.

In fact, they are left with two levers: making sure that the consumer is properly protected form any abuse (that’s already the case, basically) and dealing smartly with the tax issue. The digital economy has a long track-record of linking success to hubris — in practical terms, it means a strong disregard for local tax systems. Here in Europe, the first thing Uber and Airbnb did was setting most of their operations in tax-friendly places such as Luxembourg or Ireland — like Apple or Google before them. On the long run, that’s obviously a mistake as politicians will seize on the opportunity to further single out these new models. In fact, Disruptors would be well-advised to play by the rules in order to insure the sustainability of their services.

frederic.filloux@mondaynote.com

Uber, Airbnb vs Cartels and Regulators

 

Disruptive models for transportation or accommodations are perfect illustrations for the gap between friction-free, agile new models and the cohort of status quo defenders. For their part, regulators, lost as they are in digital translation, are worse than powerless. (Part of an series on Disruptors)  

Last week, a member of the French Parliament released a long-awaited report addressing the fight between taxi-cabs and digital-era car services such as Uber. In the meantime, the new socialist mayor of Paris, Anne Hidalgo, is training her guns on Airbnb, the acclaimed lodging system that is making a killing in the capital.

Not all the disruptors are Ugly Americans, though. Local startups take advantage of an unfortunate side of French culture – bad service at high prices – to put a dent in established markets. To name but a few: Drivy, a car rental system between individual, offers a inexpensive service available 24/7 (most French car rental agencies in Paris downtown are closed on Sundays) and, thanks to the backing of the German insurer Allianz, the company has the resources to grow; similarly, TripnDrive offers free airport parking if you make your car available for rental (of course you get paid if the car is actually rented), again with the backing of a major insurer.

What’s going on in Europe is interesting. Let’s focus on Uber (a serious knee injury made me an assiduous customer of Uber and drivers gave me lots of details about their economics). The office of the Prime Minister commissioned a report after last January’s violent protests by registered cabs who blocked airports accesses and attacked some Uber cars. In France, every government from right to left, has a solid track record of yielding to street protests, which explains why the country is so immune to structural reforms such as the ones implemented in Canada or Sweden. To the PM’s credit, hearings where thorough and the report (PDF here in French) provides a detailed view of the situation. To make it short, city cabs are artificially limited to 17,636 cars for the Greater Paris, thanks to a license system that cost around €200,000 in Paris (€300,000 in Nice or Cannes; in New York a medallion can fetch $1 million). Such malthusian, lobby-driven policies lead to poor supply: Only 3 cabs for 1000 inhabitants in Paris, vs. 13,5 in NYC, 11 in London, 8 in San Francisco, 7 in Seoul, etc.

Practically, a Paris cab driver willing to work as an independent (as many do) must cough up close to €300,000 ($415,000) — that will include the mandatory license, the car, equipment, affiliation to a dispatching system, insurance –  before earning a single euro. Should h/she should choose to lease a taxi, the cost will be €4500 per month for the whole setup. No wonder why taxi drivers are jealously defending their expensive turf.

On the service side, it’s not a pretty sight: filthy cars, credit cards not accepted, beware if you don’t tip, rough manners (like in a bistro, clients are always a pain), endless tales of foreigners overcharged, no cars in sight when it rains, if you book it, expect an “approach fee” of €10-15 (plus €7 minimum fee, plus $35 per hour of waiting time or traffic jam), you’re expected have change on you, etc.  I feel bad conveying such a picture, but that is exactly the situation.

Needless to say, this left a wide open field to Uber-like systems that offer all the agility of modern digital services. Based on multiple interviews I made during my daily trips:
- Drivers are much younger than regular taxi drivers; the oldest ones are former cab drivers who hastily sold their beloved license before it depreciates.
- Above all, they enjoy the freedom of working whenever they want (especially when the demand is high) and the simplicity of the whole process.
- Most want to develop a business around it (some share a sedan with an associate and develop their own clientèle).
- The entry price is much lower. No license needed to operate, and if the driver cannot or will not invest, rental is much cheaper: €2,000/month max, versus €4,500 for a registered cab.
- A swift and friction-free system for booking and paying (no-money exchange with the driver).
- Greater security for the driver who doesn’t carry cash and whose customers are duly identified in the system.

The French government’s response? Restriction and demagoguery. After the cabs’ violent outburst, the administration decided to freeze car service registrations. Today, the report’s authors are willing to constrain Uber’s development.

For instance, it recommends to forbid what it calls “digital hailing”, i.e, the ability to visualize on the app the nearest cab and hiring it, like here:

uber-geol

This possibility of visual geolocation would be left exclusively to registered cabs (although none of them use any app). Uber and its likes would be restricted to advanced reservation — which turns to be a fine line: Uber’s geolocation system is powered by FourSquare’s database of various places. Practically it means booking a Uber car from the Café de Flore  by name is fine, but I can’t use the street address of the place. The parliamentary report is filled with such nonsense and betrays a deliberate disregard for the user perspective (a notion not even mentioned in the opus.)

Legislators and lobbies are missing a key point here. What users enjoy the most with services like Uber or Airbnb are two things:

(a) The lightness and the reliability of the intermediation : an app acts as a trusted third party that, in addition, will smartly address the supply and demand issue. As an example, next month, Uber is going to dispatch more than 100 of its best limo drivers for the Cannes Film Festival to address the surge in demand (apparently, there is so much work on the Croisette that local cabs are willing to tolerate the newcomers.)

(b) The frictionless and robust transaction system: The cell phone is the sole payment vector in the case of Uber, and Airbnb is acting as an escrow agent that guarantees the transaction for both parties.

By and large, the European resistance to the digital modernization of commercial services is driven by two concurrent ideological postures: defense of well-established, well-identified leagues — and strong anti-americanism.

In Brussels, on April 15th, a court order issued a straightforward ban of application-powered car services such as Uber, triggering the anger of the European commissioner for digital policies Neelie Kroes (see also her adamant blog post):

As for the socialist French MP, while his report leans only softly in favor of the old-fashion cab system that no French government wants to upset, its own website clearly expresses his personal bias again what he calls “Uber’s Cow-boy behavior” (always the long-standing cliché) and the fact that Google is behind the service.

Disruptive models are growing like weeds. Everywhere.

Next week, we’ll explore four critical issues:

1. The long term macroeconomic impact of disruptors, once they’ll have percolated in almost every sectors; how to deal with heavily funded players (Uber and Airbnb are valued at $3.5bn and $10bn respectively, and the fact they are not willing to pay taxes in their local markets).

2. The social impact for the new breed of workers who enthusiastically — sometimes naively — embrace such newcomers.

3.  The cascading effects of disruptors that will call for a modernization of many connected sectors (e.g. all sorts insurances, funding systems) and the underlying factors that pave the way for earthquake-like disruptions.

4. The upcoming transformation of industry lobbies and its political impact.

This is just the beginning of the story.

frederic.filloux@mondaynote.com

Religion is a safer bet than Facebook

 

Facebook’s incredible global reach and success appear to forestall challenges. In the long run, though, the social network’s growth and its frantic quest for new revenue sources raise questions. (First of two articles)    

Casting doubt on Facebook’s future is like going to Rome and questioning the existence of God. It’s not the right venue to do so. First, you can’t argue with figures, they’re overwhelming. Each institution features about the same number of devotees: 1.2 billion across the world. As for financials, Facebook’s annual report shows strong growth and wealth: $7.8bn in revenue for 2013 (+ 55% vs 2012), net income at $1.5bn and a $11bn cash pile. As for the Catholic Church, since it doesn’t not issue financial statements, we are left to guesstimates. Two years ago, a story in the Economist provided a back-of-the-envelope calculation putting the operating budget of the American Catholic Church alone to $170bn, the bulk being health and educational institutions, with $11bn for parishes where hardcore users are – which, for that part, is much better than Facebook.

Why, then, question Facebook’s future? Mainly for two reasons: ARPU evolution and diversification.

Let’s look at a few metrics. The most spectacular is the Monthly Active Users (MAUs) base: 1.23 billion people for the entire world. An interesting way to look at that number is to break down the global MAUs into geographic zones and combine those with ARPU numbers (calculated from the quarterly figures stated in the annual report). The results look like this:

316_facebook_arpu

Facebook’s long term challenge comes from these two factors: North American growth will be flat this year, and the rest of the world doesn’t bring much. The company is heavily and increasingly dependent on advertising: from 85% of its revenue in 2012 to 89% last year. Logically, its only option is to squeeze more money per user — which it steadily managed to do thus far. But, in the Facebook ecosystem, making more money from ads means milking more cash from users’ data. This, in turn, will lead to a greater invasion of privacy. It certainly doesn’t seem to bother Mark Zuckerberg, who is a transparency apologist.

Actually: Is he or was he?

As author David Kirkpatrick pointed out in his excellent opus, The Facebook Effect, Zuckerberg once said that “Having two identities for yourself is an example of a lack of integrity” (and judging by FB’s content policy, anyone can wonder if putting a breast-feeding pictures a sign of depravation?)

That was then.

Now, to address privacy concerns, Facebook is said to consider anonymous logins. It’s probably a good idea to back off a bit on the totalitarian pitch quoted above, but since the extensive data-mining performed by the network is made much more valuable by its use of real user names, anonymous logins are sure to impact the ARPU in the more mature markets. Along that line of thought, in Europe, Facebook’s ARPU is less than half of what it is in the US & Canada: $8.04 vs. $18.70. This significantly lower number stems from privacy concerns that are much more developed in European countries. There, the 20-25 segment seems especially worried about the consequences of spending too much time on Facebook.

A remaining lever is what I’ll call the Big Tobacco strategy: Do elsewhere what you can no longer do on your home playing field. Facebook might not be as cynical as Philip Morris (reborn as Altria as an attempt to erase the stain), but it is undoubtedly bound to try and replicate its successful collect-and-milk consumer data mechanism.

This might take a while to achieve.

First because of the ultra-slim ARPU generated by emerging markets users. You might object that the Indian market, as an example, currently enjoys growth along two dimensions: more users, with growing incomes. Granted. But the more sophisticated the India market becomes, the more inclined it will be to create a social network much more attuned to its own culture than a Menlo Park-based system manned by geeks in hoodies. Never underestimate the power, nor the determination of locals. And, let’s not dream too much about a huge Chinese version of Facebook.

Also, for Facebook, the cost of operating its service will make the ARPU question one of growing urgency. Again, based on the 2013 annual report, FB’s Cost of Revenue — mostly infrastructure –  amounts to $1.9bn. Divided by the 757 million DAUs, it costs $2.5 per year to serve a single daily user, that is connecting to his/er pals, hosting photos, videos, etc. If we aggregate all the cost structure components (networking, giant data centers and also R&D, sales & marketing, administrative), the cost of taking care of a single daily user rise to $6.69 per year and $4.12 for a monthly user. It’s still fine for an American and a European, much less so for an Asian who brings a yearly ARPU of $3.15, or an African who brings a mere $2.64 (in theory, the strain on the infrastructure is roughly the same, regardless of user location).

But some will argue Facebook is doing quite well on mobile. Out of its 1.23 billion monthly users, FB says 945 million reach its service via a mobile each month and 556 million do so on a daily basis. And, as stated in its 10-K, mobile is at the core of Facebook’s future:

There are more than 1.5 billion internet users on personal computers, and more than three billion mobile users worldwide according to GSMA Wireless Intelligence, and we aspire to someday connect all of these people. 

Fine, but once again, the ARPU weakens the ambition. While a mobile subscriber in the US and Europe brings respectively $69 and $38 each year (source: GSMA), according to the Cellular Operator Association of India, a Indian mobile subscriber yields only $1.72 per year. This makes advertising projections a tricky exercise.

As it expands, Facebook’s current model will inevitably yield less and less money per user. Hence, its frenetic quest for diversification and service extensions — a topic we’ll address in a future Monday Note.

As for the Church, it certainly is a safer bet than Facebook: The user base is less volatile, the interface blends much better into local cultures, barriers to competitive entries are stronger (and much older), and believers have long sacrificed their privacy to articles of faith.

frederic.filloux@mondaynote.com

 

Lessons from a good vertical: Skift.com

 

For digital media companies, creating good verticals that breed small but valuable audiences has become essential. On that subject, here are my takeaways following a conversation with Rafat Ali, founder and CEO of Skift.com. In 20 months, Rafat’s company has become a reference in the travel intelligence business. 

There is no excuse for not trying to build a vertical digital service (web site & mobile app) for a strong media company shifting to digital. As long as you have a powerful (not to be confused with profuse) newsroom coupled with a well-structured contents system, trying a foray in a specific domain is worth considering. As an example, see Atlantic Media, one of the most innovative media brands, as it deploys a series of verticals nested in its Government Executive Media Group. These units all generate small but extremely valuable and loyal audiences — and enviable revenue per user (more on the Atlantic in a future Monday Note).

Building a vertical is a mere matter of implementation, you might say. But a look below the surface shows how such process demands much more than merely putting a small group of good writers in a digital stable, and asking them to gather news on a specific subject.

That’s why Skift.com drew my attention. In less than twenty months, manned by only 9 people crammed in an mid-town Manhattan office, Skift.com has become a strong voice and a reference in the travel industry: airlines, booking systems, hotels, tour operators – and all the the sector’s disruptors.

OLYMPUS DIGITAL CAMERA

I met Rafat Ali five years ago in Hyderabad, India; we were both of speaking at the same conference. Rafat was about to exit his first and remarkable startup, PaidContent.org (a terrible name he now laughs off), one of the first blogs decoding the media industry’s transformation. After building it from scratch and spending eight exhausting years producing and editing stories, Rafat sold it to the Guardian for a reported $30m – right before the 2008 crisis. (Last year, PaidContent was acquired by GigaOm).

After a short transition, Rafat was free to go. So did he. In 2010, at the age of 36, he left for a two-year series of trips to Oman, Iceland, Burma, India (where he has family), radiating from his bases in New York and London. At last out of PaidContent’s trenches, he took the time to read a hundred books during his journeys. Following @rafat on Twitter, you could feel his excitement, and also his growing interest in the travel sector.

‘You have to remember, it was 2010, the iPad had just been launched, everyone was thinking about what to build on it’, said Rafat. His first idea was to re-invent the travel guide book for the iPad. But he soon realized how crappy the whole travel industry’s information ecosystem was: ‘I was blown away.’ While the transactional part of the travel business had been completely broken apart by a massive, unprecedented disintermediation — benefiting the customer, trade information remained frozen in the past, with its sets of professional printed publications perpetuating a jargon-filled verbiage offering little or no actionable intelligence, nor useful data

Nature (and digital business) abhors vacuum, so does Rafat Ali, who decided to fill the void. When asked to define Skift in a nutshell, he said this: ‘In late 2011, we wanted to build the Bloomberg News of travel’. (When it comes to business information, this is quite a goal. Never aim low, I can’t agree more.) Rafat’s wanted to build something based on a few concepts: rely heavily on data, capitalize on the open-web, use APIs aggressively (to connect with third party data sets), aim at professionals, consultants, experts, and — last but not the least — prosumers who often know more than merchants. (Read Rafat’s post on the “Mediata” Startups).

The other key to Skift’s concept — which means shift in Danish — was tearing apart the silo culture that plagued the travel industry for decades: ‘You have airlines, airports, cruises, hotels, technology… All of these silos have collapsed in global interconnected megatrends, and we knew we could make our voice heard across all…’, explained Rafat while pointing at this graph:

SkiftCircularGraphic-b
Graph ©
Skift.com

As far as editorial is concerned, Rafat believes journalistic content is needed to create addiction, daily use, while-data related products generate usefulness, stickiness, loyalty and, ultimately, monetization. Content-wise, at the beginning, the site was built on four “legs”: aggregation (collecting headlines); curation (with a tweet-length phrase to describe a story); licensed content (full articles brought from news providers); and originally produced articles. Today, Skift is down to two items: 40% of articles are licensed (mostly Newscred) and 60% are original content — about 15-20 short business stories (produced by a staff of three…)

Business-wise, Skift positioned itself primarily as a B2B company, then secondarily as B2B-2C. Its traffic is still modest (1m UVs/mo), but growing fast; so does its newsletter business, expected to reach 75,000 subscribers by year end. No mobile apps in sight as the mobile web works well for Skift: mobile users account for 35% of web traffic and 50% of newsletters readings.

Skift sells few but high yield ads, to the point that Rafat is about to create a tiny studio to create bespoke brand contents. (Maintaining the mandatory Chinese wall could be tricky in such a small structure.)

But Skift’s true gem is its industry dashboards and data collection system, a well-structured tree that leads to scores of statistics and rankings. Inside, you’ll learn that AirBnB — whose valuation is now higher than Hyatt — has a Skift Score (a combination of indicators) roughly twice the “bookings & tools” industry average. Or that Dutch airline KLM scores way better than the hippest Virgin Atlantic. Or that Hertz masters the social ecosystem way better than the trendy Über.

Using data analytics, Skift produces reports — short and updated twice a month (as opposed to quarterly “bibles” prone to quick obsolescence.) ‘We will focus mainly on marketing, strategy and technology to produce competitive intelligence’, said Skift’s CEO. Rafat’s intense focus on doing few things but doing them well extends to the obligatory conference business: Skift intends to do just a single event about the Future of Travel, in a similar fashion to Quartz’sThe Next Billion conference (see the #qznextbillion hashtag for a list of tweets linking to videos). In both cases, these events are built on strong editorial concepts, ‘We want to make a conference about leadership instead of a vendors-to-vendors type…’ said Rafat.

What’s next for Skift? First, an off-site staff meeting in Iceland. Actually, Rafat Ali is considering a global franchise set in Reykjavik. Less anecdotal, Skift founder wants to apply his news and contents formula beyond the travel industry to what he feels are interconnected sectors — at least in discretionary spending — namely food & beverage and retail sectors.

One final note. Looking at the state of travel information, I can’t help but discern a complete failure of traditional, legacy journalism. Too cozy with the main players and their corrupting PR machines, too filled-up with press junkets and freebies, the mainstream media coverage of this $6.5 trillion/260 million jobs sector has become mostly illegible. This leaves a large open field to new players.

frederic.filloux@mondaynote.com

 

Dealing With Data Frenzy

 

Last week, I attended the Newspaper Association of America’s MXC conference in Denver. We were to focus on the publishers’ use of data. A hot topic that sometimes becomes overly broad and leads to unrealistic expectations. Here are some key points I made.   

For any digital publisher, relying on data is no longer an option nor a luxury. It has become a necessity. Each passing quarter confirms the demise of digital advertising: yields continues to fall, programmatic buying (most often operated by large third party players) takes over and continues to fuel deflation. Highly visible media brands — two years ago the Huffington Post, now BuzzFeed — deal with the issue by generating huge quantity of pages saturated with kittens and listicles, each yielding very low CPMs. At the other end of the spectrum, strong media houses develop customized, sophisticated campaigns for high-end brands (see examples on NYT’s IdeaLab page); they also fill their pages and apps with so-called Branded Content items — which very few publishers manage to implement correctly.

For news publishers, the use of data should focus on four goals:
– Increase advertising yields through smarter targeting
– Improve their editorial recommendation engines (hence raising the number of page views per visit)
– Up-sell ancillary products
– Raise the performance of their subscription system (if any.)

Over the recent years, the advertising community managed to find a new gun to shoot itself in the foot. It’s called targeted ads. Everyone has ugly anecdotes about those. Typically, the stories go like this: You do a web search for an item and quickly find it. In the following months you’re deluged by ads for the product you bought. The annoyance prompts many to opt for AdBlocking systems — I did (except for sites I’m in charge of), with no regret nor guilt.

To put it mildly, there is room for improvement, here.

Coming back to profiting from site users’ data, one good example I heard recently is a recent request made by a large airline to the Financial Times: “Find us people who travel on long haul flights and who log on FT.com more than four times per month and from locations scattered along our routes”, i.e. super-frequent flyers used to business or first class, etc. Thanks to IP location analysis yielding geospatial coordinates for each connection, the retrieval of such high-value clientele wasn’t overly complicated. The geolocation principle applies to other requests, such as finding residents of a specific city or suburb, in order to serve effective advertising.

When you think of profiling, use a passport analogy. Anyone who visit a site from a browser (it’s more complicated with mobile apps), is issued a passport in the form of an anonymous cookie, such as this one, injected in my computer by the New York Times:

nyt_cookie

As a digital subscriber, I have inherited no less than 113 cookies from the NYTimes, each one stored in my computer for a specific purpose. They come from every segment of my navigation (pages, sections, articles, blogs), each generates a “stamp” on the passport. The more stamps I get on my passport, the more NYTimes people knows about me. Over time, the process draws a finely defined profile.

The example I often use looks like this (my perspective is from a French business media company):

Based on her past navigation, the user ID:6547dgfc_9088 turned out to be:
– A woman in her 30′s
– Leaving in Toulouse [Thanks to the geolocation of her internet box] She works :
…in the aerospace industry
… most likely in a financial department
… with a special interest in European regulations
… at a fairly high position.
Then we should be able to serve her with:
– Local ads / adjusted for her income and likely tastes [she also visits our lifestyle sections & other online properties in our network] – Adjusted editorial recommendations [related stories] based on her sector and position
– A special deal for our next conference on corporate finance
– A notification when someone in our team or among our partners publishes a book (paper or digital) relevant to her interest
– An abstract of our annual in-depth survey on aerospace
– A sneak-peak at our partner’s COOC (Corporate Open Online Course) featuring four hours of talk by a prominent tax lawyer from Brussels [don't forget the Red Bull] – [And if she's not a subscriber] A promotional, customized, one-time newsletter featuring the economics of commercial airplanes, with past stories from our newsroom, curated links, etc., all of the above driving to the inevitable conclusion: this discerning individual should definitely take advantage of our one-time offer.    

This “internal” profiling can be spectacularly enhanced by working with a major profiling third party. As an example, the large European player Weborama has accumulated a staggering 70 million profiles for an internet population of 52 million French users (each person can be linked to multiple profiles.) All over Europe, Weborama has collected 210 million profiles, roughly 40% of Europeans web users. In our example, by tapping into such large databases, the profile of this upwardly moving female exec from Toulouse will be enhanced up to the minutest detail of her tastes and preferences.

For the media company, reaching such productive interplay between a profiled individual and its ability to serve her with relevant content, services and products requires a well-integrated system — and a critical mass of products.

Understanding someone’s social and semantical genome through internal an external profiling is only a part of the equation. Matching the customer’s profile to the company output (journalism, conferences, publications, surveys…) also demands that the genome of those products be precisely established. If we want to “talk” to the customer’s profile, a story must have its set of tags, keywords and metadata; so does the theme of an upcoming conference that must go beyond a basic presentation, or the description a book. Ideally, every single piece of what the news organization produces must have its semantic genome encoded in a standardized way.

In defining user profiles, media organizations must have a rich and diverse line-up of contents, services and ancillary products. The broader the spectrum of a media brand, the better. All things being equal in terms of editorial quality, an isolated media will be less well armed than a larger company that operates multiple properties ranging from editorial to e-commerce and uses those to construct a wide range of user profiles.

Much more than in print media, isolation is not an attractive option in the digital world.

frederic.filloux@mondaynote.com

On Marc Andreessen’s optimistic view of news

 

A strongly-worded column by venture capitalist Marc Andreessen triggered an intense debate on the future of news. Andreessen might be right places, but his views can also be dangerously simplistic. 

For starters, it is always great to have an outsider’s view. Marc Andreessen’s witty, and fast-paced dithyramb on the future of news is undoubtedly welcome. But, as always, regardless of the depth and breath of the big picture he paints, the devil lies in the details. In no particular order, here are my thoughts on his manifesto.

As a European, I found his piece extraordinary US-centric or, slightly more broadly, Anglophone-centric.

Andreessen wrote :

[T]he market size is dramatically expanding—many more people consume news now vs. 10 or 20 years ago. Many more still will consume news in the next 10 to 20 years. Volume is being driven up, and that is a big, big deal.
Right now everyone is obsessed with slumping prices, but ultimately, the most important dynamic is No. 3 – increasing volume. Here’s why: Market size equals destiny. The big opportunity for the news industry in the next five to 10 years is to increase its market size 100x AND drop prices 10X. Become larger and much more important in the process.

By saying this, Andreessen makes two good faith mistakes.

First, he mixes up global reach and monetizable audience. Evidently, a growing number of people will enjoy access to news (maybe not all the 5 billion cellphone users he mentions), but the proportion of those able to generate a measurable ARPU is likely to be very small.

The Scalability that works for Google Maps or WhatsApp doesn’t work as well for the notion of relevant information, one that is more tightly connected to language, proximity and culture.

Second, he overestimates the addressable news market’s fragmentation. I live in France, a 66 million people country with a high standard of living and good fixed and mobile internet access. In spite of these factors, it remains a small market for the super-low-yield digital news business that brings few euros per year and per user (except for a minuscule subscriber base.) I remained stunned by the inability of good journalistic products, created by smart people, to find a sustainable business models after years of trying.

And the huge, globalized English speaking market does not warrant financial success. The Guardian is one such example. It operates one of the finest digital news system in the world but keeps bleeding money. The Guardian brings a mere $60m in digital ad revenue per year — to be compared to a kitten-rigged, listicles-saturated aggregator generating a multiple of this amount. Journalism has become almost impossible to monetize by itself (I’ll come back to that topic).

Andreessen also vastly underestimates the cost of good journalism when he writes:

[T]he total global expense budget of all investigative journalism is tiny —  in the neighborhood of tens of millions of dollars annually.”

Fact is, journalism is inherently expensive because it is by laborious and unpredictable: An investigation can take months, and yield nothing; or the journalistic outcome can be great, lifting the reputation of the media, but with zero impact on the revenue side (no identifiable growth in subscriptions or advertising). The same goes for ambitious coverage of people or events. No one has ever translated a Pulitzer Prize in hard dollars.

This is also the case for what Andreessen calls the “Baghdad Bureau problem”. It was said to cost $3m/year for the New York Times. In fact, on an annual basis, the Times spends about $200m for its news operations, including $70m for foreign coverage alone. The NYT is likely to stay afloat when it goes entirely digital (which might happen before the end of the decade), but one of the nastiest features of digital news is the unforgiving Winner Takes All mechanism.

As far as philanthropy is considered, I won’t spend too much time on the issue except to say this: Relying on philanthropy to cure malaria or to support ill-understood artists bears witness to an absence of sustainable economic system. (Until, perhaps, the artist dies; as for malaria, there is indeed a very long term benefit for society, but not for those who supply the treatment, hence the mandatory call to generosity.) Saying investigative or public-interest journalism could/should rely on philanthropy is the same as admitting it’s economically unsustainable. Luckily, American society has produced scores of philanthropists free from any agenda (political, ideological, religious) — such as the Sandler Foundation with ProPublica. That’s not the case in France — not to mention Russia and many other countries.

There are plenty of areas in which I completely support Marc Andreessen’s view. For example: A media company “should be run like a business“, i.e. seek the profitability that will warrant its independence (from every economic agent: shareholders, advertisers, political pressure, etc.) This brings us to the size and shape of a modern news factory (I use the term on purpose). We have to deal with an unpleasant reality: Good journalism is no longer sustainable as a standalone activity. But — and that’s the good news — it remains the best and indispensable core around which to develop multiple activities (see my recent column about The News Media Revenue Matrix).You can’t develop services, conferences, publishing, etc. around a depreciated journalistic asset. On the other hand, this asset has to be drastically streamlined: In many cases, less people, better-paid (simply for the ability to retain talent) and with sufficient means to do their job (don’t go for the press junkets because the travel budget has been slashed, you’ll lose on three counts: credibility of your brand, self-esteem of your team, quality of the reporting.)

Unfortunately, as Andreessen noted, there are plenty of hurdles to overcome. In fact, most existing news companies do not fathom the depth of the transformation required to survive and thrive. Nor do they understand the urgency to set this massive overhaul in motion. Such moves require strength, strong leadership, creativity, a fresh approach, unabated confidence, and a systemic vision — all of the above in short supply at legacy media. Note that when Marc Andreessen prides himself to be an investor in media ventures (for instance Business Insider– no conflict of interest), all are digital natives and bear none of the burdens of traditional media. His bullishness on news is selective, personal.

frederic.filloux@mondaynote.com

News Media Revenue Matrix: The Bird’s Eye View

 

Publishers struggle with newer and more complex business models. Some appear stronger than others but, above all, a broad palette is a must. It is a means to capture emerging opportunities and to compensate for the drying up of older revenue sources.

Today, I submit the following revenue matrix for a modern, content-rich news outlet. As I see it, in the news business “modernity” mean this:

A proven ability to produce original content in abundance and under multiple forms: news reporting, investigation, analysis, data journalism, long form (for ebook publishing), enterprise-like journalism, live feeds; all of the above in the form of text, images, graphics and videos.

A cultural mindset to produce contents for the platform with the best fit: a news story for a newspaper, an interactive piece on the web, live coverage for mobile. The collective publishing mindset should no longer allow first- and second-class news products. Every piece of newsroom output must be designed as a contribution to a cascading revenue system in which each element empowers every other one.

– A newsroom equipped with the best tools money can buy or — even better — build. These include a powerful Content Management System (CMS) aimed at dispatching production to every platform. The CMS must be connected to a semantic analysis system that makes all pieces of information — from a feature story to the transcript of a video — compatible with the semantic web’s standardized grammar. In order to extract more value from a piece of content, the CMS must also connect to multiple databases. For example, the name of an obscure city must be able to generate a map – through the Geonames base; a Board Director must be tied to a high value database of business leaders such as The Official Board; the name of a company must lead to open-source corporations listings.

Mastering the semantic web is indissociable from acquiring information gathering capabilities such as aggregation and filtering (see a previous Monday Note: Building a business news aggrefilter ). Such feature is a prerequisite to building high-margin products as well as exploiting the social media echo chamber. After collecting contents through RSS feeds, the combination of semantic news analysis matched against the taxonomy of, say, Twitter, will yield a trove of information on what audiences like or dislike — not only for a news media but also for its competitors. It is a complex and expensive endeavor but, in the long run, it will be worth every penny.

– And more importantly, a global editorial thinking. Too often, newsroom management suffers form what l’ll call “mono-product bias”, focusing on what is seen as noble — namely print. At a very minimum, modern editorship must embrace a widespread digital strategy. But it also must envision a sustainable game plan for a complete lineup of ancillary products that also deserve editorial coherence and strength.

Having said that, let’s have a look at the following matrix. No rocket science here, I simply made a list of 14 products that many news outlets already operate. I then tried to assess the outlook for each revenue stream. (My original idea was to assign a estimated ARPU for each cell, but there are too many parameters to be taken into account).
Click to enlarge the table:

310 table revenue

Now, let’s focus on specific products and revenue streams.

Daily Print Edition. I’m very bearish on print. Granted, it still brings the most substantial chunk of revenue – but also most of the losses. And prospects are bleak: copy sales, subscriptions, even ad sales deteriorate fast. Some light can come from ads – when they are components of customized campaigns. Daily newspapers need to be vastly simplified in order to free up resources for the wide array of other revenue streams — especially digital. I’m a big supporter of Financial Times’ Lionel Barber “Memo on reshaping the newspaper for digital age“.

Weekend editions will do better than dailies for several reasons. First, their function — long formats, portfolios, reading habits — makes them better armed against the digital tsunami that devoured news. Second, they remain a great vector for pricey advertising: on some anglo-saxon markets, weekend editions accounts for half of the print ad revenue. The New York Times understood that well as its full digital access + weekend edition bundle is a hit among customers.

Advertising revenue stream. Let’s face it, traditional ads ormats, print or digital, are dying. The conjunction of programmatic buying and ad saturation/tracking/targeting will seal their fate for good. The best outlook seems to be for customized operations and brand contents (or combinations of the two). They can spread on every platforms, including on mobile where, so far, users massively reject ads. In addition, these customized operations carry high value (huge CPMs or hefty flat fees.)

Event & Conferences. The segment is crowded and success depends on a subtile combination of attendance fees vs sponsorship, but also of editorial content. A conference is indeed a full editorial vector that needs to be treated with the same care as any other publication, i.e, with a precise angle, great casting and first class moderation that favors intellectual density over speakers flogging cheap sales pitches. News media are well positioned to deploy an efficient promotion for a content-rich, sustainable, conference system.

Intelligence & Surveys. Attractive as they might sound, these products require a great deal of expertise to make a difference. Very few media can fulfill the promise and justify the high price that goes along with such offerings.

Training and MOOCs represent an interesting potential diversification for some business publications. They carry several advantages: by addressing a young readership, MOOCs can create an early attachment to the brand; the level of risk is low as long as the media company limits itself to being a distributor (quality MOOCs production is very expensive). For a business publication, such activities represent a great way to increase its penetration in the corporate world where the need for training is limitless.

Premium Subscriptions. Some large, diversified media companies are already considering complex subscription packages for a small number of high-yield clients. In addition to print and full digital access, such packages could include access to conferences & events, MOOCs, market intelligence, and other publications. Testing the concept is a low-risk proposition.

The Business to Business segment remains the province of specialized publications. But the potential is there for general-audience media: corporations are hungry for information. The era of the bulky corporate intranet that no one watches is gone; today, for their staff, companies want apps for mobile and tablets that will save time while being precisely targeted and well-designed. Not an easy market – but  a very solvent one.

Sketchy and questionable as it is, the above matrix also illustrates the complexity of designing and selling such a wide range of products to individuals or corporations. Only a small number of news organizations will have the staff, skills and resolve to address such a broad range of opportunities.

frederic.filloux@mondaynote.com

@filloux

Building a business news aggrefilter

 

This February 10, Les Echos launches its business news aggrefilter. For the French business media group, this is a way to gain critical working knowledge of the semantic web. Here is how we did it. An why. 

The site is called Les Echos 360 and is separate from our flagship site LesEchos.fr, the digital version of the French business daily Les Echos. As the newly coined word aggrefilter indicates, it is an aggregation and filtering system. It is to be the kernel from which many digital products and extensions we have in mind will spring.

My idea to build an aggrefilter goes back to… 2007. That year, in San Francisco, I met Dan Farber, at the time editor-in-chief of CNet (now at CBS Interactive, his blog here) – and actual father of the aggrefilter term. Dan told me: ‘You should have a look at Techmeme. It’s an “aggrefilter” that collects technology news and ranks them based on their importance to the news cycle’. I briefly explored the idea of building such an aggrefilter, but found it too hard to do it from scratch, off-the-shelf aggrefilter software didn’t exist yet. The task required someone like Techmeme founder Gabe Rivera – who holds a PhD in computer science. I shelved the idea for a while.

360 cap

A year ago, as the head of digital at Les Echos, I reopened the case and pitched the idea to a couple of French computer scientists specialized in text-mining — a field that had vastly improved since I first looked at it. We decided to give a shot to the idea. Why?

I believe a great media brand bearing a large sets of positive attributes (reliability, scope, depth of coverage) needs to generate an editorial footprint that goes far beyond its own production. It’s a matter of critical mass. In the case of Les Echos, we need to be the very core of business information, both for the general public and for corporations. Readers trust the content we produce, therefore they should trust the reading recommendation we make through our aggregation of relevant web sites. This isn’t an obvious move for journalists who, understandably, aren’t necessarily keen to send traffic to third party web sites. (Interestingly enough, someone at the New York Times told me that a heated debate flared up  within the newsroom a few years ago: To which extent should NYT.com direct readers to its competitors? Apparently, market studies settled the issue by showing that readers of the NYT online actually tended to also like it for being a reliable prescriber.)

In the business field, unlike Google News that crawls an unlimited trove of sources, my original idea was to extract good business stories from both algorithmically and manually selected sources. More importantly, the idea was to bring to the surface, to effectively curate specialized sources — niche web sites and blogs — usually lost in the noise. Near-real-time information also seemed essential, hence the need for an automated gathering process, Techmeme-like. (Techmeme is now supplemented by Mediagazer, one of my favorite readings.)

Where do we go from here?

Initially, we turned to the newsroom, asking beat reporters for a list of reliable sources they regularly monitored. The idea was to build a qualified corpus based on suggestions from our in-house specialists. Techmeme and Mediagazer call it their “leaderboard” (see theirs for tech and media). Perhaps we didn’t have the right pitch, or we were misunderstood, but all we got was a lukewarm reception. Our partner, the French startup Syllabs, came up with a different solution, based on Twitter analysis.

We used our reporters’ 72 most active Twitter accounts to extract URLs embedded in their tweets. This first pass yielded about 5000 URLs, but most turned out to be useless because, most of the time, reporters linked their tweets to their own or their colleagues’ newsroom stories. Then, Syllabs engineers had another idea, they data-mined tweets from people followed by our staff. This yielded 872,000 URLs. After that, another filtering pass found out the true curators, the people who found original sources around the web. Retweets also were counted as they indicate a vote of relevance/confidence. After further statistical analysis of tweet components, the 872,000 URLs were boiled down to less than 400 original sources that were to become the basis of Les Echos 360′s Leaderboard (we are now down to 160 sources).

Building a corpus of sources is one thing, but ranking articles with respect to their weight in the news cycle is yet another story. Every hour, 1,500 to 2,000 news pieces go through a filtering process that defines their semantic footprint (with its associated taxonomy). Then, they are aggregated in “clusters”. Eventually, clusters are ranked based according to a statistical analysis of their “signal” in the general news-flow. Each “Clustering” (collection + ranking) contains 400-500 clusters, a process that more than occasionally overloads our computers.

Despite continuous revisions to its 19,000 lines of code, the system is far from perfect. As expected. In fact it needs two sets of tunings: One to maintaining a wide enough spectrum of sources to properly reflect the diversity of topics we want to cover. With a caveat: profusion doesn’t necessarily create quality. Crawling the long tail of potentially good sources continues to prove difficult. The second needed adjustment is finding the right balance between all parameters: update frequency, the “quality index” of sources – and many other criteria I won’t disclose here. This I compare to the mixing console inside a recording studio. Finding the right sound is tricky.

It took years for Techmeme to refine its algorithm. It might take a while for Les Echos 360 — that’s why we are launching the site in beta (a notion not widely shared in the media sector.) No surprise, a continuous news-flow is an extremely difficult moving target. As for Techmeme and Mediagazer, despite refinements in Gabe Rivera’s work, their algorithm is “rectified” by more than a dozen editors (who even rewrite headlines to make them more explicit and punchier). A much lighter crew will monitor Les Echos 360 through a back-office that will allow us to change cluster rankings and to eliminate parasitic items.

For Les Echos’ digital division, this aggrefilter is a proof of concept, a way to learn a set of technologies we consider essential for the company’s future. The digital news business will be increasingly driven by semantic processes; these will allow publishers to extract much more value from news items, whether they are produced in-house or aggregated/filtered. That is especially true for a business news provider: the more specialized the corpus, the higher the need for advanced processing. Fortunately, it is much easier to fine-tune an aggrefilter for a specific field (logistics, clean-tech, M&A, legal affairs…) than for wider and muddier streams of general news. This new site is just the tip of the iceberg. We built this engine to address a wide array of vertical, business-to-business, needs. It aims to be a source of tangible revenue.

frederic.filloux@mondaynote.com

@filloux