Tech

Why build a digital newsstand?

A kiosk owned and controlled by publishers will be beneficial for all involved

On 30 June the French consortium ePresse opened its digital kiosk. Six months of hard work for a very small team (the ePresse consortium is a three-person operation: a CTO, a marketing person, and a manager), and still a long way to go. ePresse brought up eight titles: five dailies (Le Figaro, Le Parisien and its national edition, Libération, the sports daily l'Equipe and the business paper Les Echos), and three newsweeklies (L'Express, Le Point, Le Nouvel Observateur). This is only the rocket's first stage: an iPad/iPhone app allowing per-copy purchases within the App Store; more to come this Fall.

Knowing I'm charge of this development, editors and news executives abroad inquired about the experience. Here are a few early observations.

[English version of ePresse demo here

First, the big question: Why build a digital newsstand? After all, there is no shortage of places for buying online editions: Zinio, deployed globally; Relay.com and LeKiosque.fr in France. And, of course, Apple, which will roll-out its own Newsstand before year-end.

The answer is of a strategic nature: we're dealing with concerns over control and technology.

For publishers, retaining full control of all commercial aspects of their digital sales channels is critical. They must safeguard their freedom to decide prices, marketing strategies, discounts, bundles, special deals. They must also protect their ability to collect valuable customer data, without having to beg permission from a third party to do so. Marketing being the tactical engine of the trade, it is also one of the most underdeveloped assets of the press – and not just in France. A kiosk owned and controlled by publishers will be immensely beneficial for all involved.

Now, let's take a walk through a usage scenario. You start by downloading the (free) kiosk application on your mobile phone. Next, you launch the application. A welcome screen greets you: for one euro (or dollar, or pound), you get unlimited access to the entire kiosk for one (or two) weeks, all you can eat.
Publishers might not like this: it amounts to a "leak" of digital copy sales that won't be counted by the Audit Bureau of Circulation. But savvy publishers will also consider the upside: (a) the customer leaves his name and credit card info (that's what the one-euro thing is about); (b) he will leave a trail of data. Then, when the almost-free trial period ends, a tailored offer is pushed to each individual customer, based on his recorded readings. An individual's preferred title would be well inspired to offer him/her a steep subscription discount.
Over time, as reading patterns build up in the customer database, it becomes easier and easier to push offers not only based on title preferences, but also on a predictable news cycle. A political newspaper might cook up special deals six months before an election; a sports paper might do the same with Olympics and similarly attractive events. Here, tactical flexibility provides inordinate payoffs. As for occasional customers sticking to per-digital-copy purchases, they should be offered an incentive to give an email address, the ultimate goal being to convert them into digital subscribers.

Now the reality check: this scenario doesn't work for current kiosks; pricing policies are constrained, promotional offers are not possible (they will eat up the kiosk's margin) and the newsstand keeps customer data for its own marketing purposes. Plus, most kiosks charge around 30%, roughly three times the cost of an efficient digital delivery system.

The same goes for bundles. Currently, platforms handle those rather crudely. For publishers, beside per-copy sales, subscription systems end up as value-killers. In France, the Hachette-operated Relay.com kiosk offers a 9.90€ a month digital bundle for up to 30 magazines. A great bargain indeed, but one that yields a mere 0.30€ for each publication -- before the kiosk's cut. In other words, nothing. One of Relay.com's bestsellers is said to be the 19.90€ a month all-magazines-you-can-read, with a similarly puny outcome for magazines.

In contrast, a publisher-run kiosk can introduce more bundling refinements such as a combined daily + weekly subscriptions system or any other such combination that makes sense marketing-wise. Deploying such arrangements will require a great deal of cooperation among titles – something close to performing unnatural acts, a delicate aspect of the job.

Building the system also involves deploying multi-title CRM (Customer Relationship Management) systems. This, in turn, requires weaving together customer databases belonging to different and sometimes competing titles – again, plenty of diplomatic issues in sight. I might be a little naïve, but I think media groups have done a great deal of progress recently when it comes to understanding the benefits of building integrated systems. With this in mind, for a consortium such as ePresse, the goal is to yield more value that the sum of its parts.

Now let's jump to the technology aspect. ePresse.fr, launched ten days ago, is but the first stage of a much larger setup. Today, we limit ourselves to proposing an iOS app with per-issue sales only, through the Apple app store (lower case ''app store'' with intent as it seems Apple won't be able to own those words). Obvious next steps include other mobile platforms and, more importantly, a subscription system directly available to smartphones, tablets and, of course, the web. In the process, we'll add a couple more titles, but we intend to remain selective.

Mobility is a critical component. Currently, digital kiosks offer mostly PDF-based editions. As discussed in a previous Monday Note, PDF is by no means the future of digital media. PDF once was a fantastic invention, but it wasn't designed for today's task: encapsulating news.
With this in mind, during the first months of ePress development, we spent a great deal of time aligning the output of the different publications to what we knew was the right target for mobility: XML feeds for stories on top of a "zoned" PDF that defines the placement of a story in a page. Such feeds were supposed to come directly from each publication's CMS (Content Management System). Some were able to supply the correctly formatted feeds right from day one, other needed upgrades to their CMS output. At publications, tech teams were very cooperative. We also got serious help from EDD, a French company specialized in digitizing media contents (EDD indexes and distributes 50,000 articles per day). EDD collects publishers' PDF files, send those to India where the files are taken apart in order to produce the required output, all of it done every night within two hours.

Once clean XML feeds (standardized for the eight titles) became finally available, we had to put those on our content-delivery platform. We did this by re-aggregating all the components (PDF, zoning/mapping files, XML files, summaries, graphic elements) within a transaction-tracking mechanism. For this, we picked miLibris, a French startup that provides reading tools and cataloging systems for publishers, and for the French ISP and mobile carrier Orange.

Again: the idea was to use native XML to publish each title we serve, fully formatted for each article.

Three reasons for this:

- Readability. You don't comfortably read by constantly zooming and pinching. The screen of smartphone covers only a 1/60th of a broadsheet newspaper. For a reading a "facsimile" rendering of a 30 pages publication you'd need 1800 pans and zooms. Insanely unrealistic. XML gives us the ability to automatically reformat the text to fit the device, smartphone, tablet or, eventually, PC browser. No more pinching and zooming, just scrolling.

- Functionalities. Relying on XML and text opens the way to a broad set of additional features: font-size adjustment, social sharing of articles, ability to create users' folders, search, recommendation engines, etc.

- Future-proof. At some point, we'll get rid of PDF. As mobility usage rises, readers will demand quicker downloads over 3G or Edge cellular network. Two obstacles remain: one is each title's graphic identity; legitimately, publishers demand the preservation of the visual aspects of the publications. As shown below, we're making progress; the PDF version of a page:

and its XML/HTML5 translation (click to enlarge both):

… But we aren't yet able to translate the minute details of a refined newspaper layout in XML and HTML5.

The second aspect is more economical. In some countries (such as France), the entity in charge circulation audits (equivalent to ABC) refuses to take in account digital copies as long as they are not exactly identical to the print version. This outdated posture explains the remanence of PDF formatting: it is accepted as a ''carbon copy'' of the print original. My take is this will evolve over time. Already, titles such as the Economist offer an encapsulated version of their print edition that carries the same editorial content, but with a different advertising setup in some parts.

The evolution of the "edition" concept of is indeed a key question. On the one hand, the notion is deeply associated with the idea of branded news encapsulated in a "cognitive container" – yesterday the paper, today the digital edition tied to an app. On the other hand, digital news also begs for real-time. This can be implemented through a variety of techniques: overlay real-time news display, or permanently updated editions, which, in turn, push hard in favor of a subscription model vs. per-copy sales, the latter a mere (but necessary) transition.

— frederic.filloux@mondaynote.com