“There’s a big data revolution” says Harvard Professor Gary King. And I agree. However, as Professor King explains, it’s not only the amount of information that is revolutionizing the digital world but the “what” we can do with it that could really make a change for the better.

For the last few months, we have been mulling over the same question at Paper.li: what can we do with all the data we have? After putting things into perspective, one thing became clear to us: that whatever we do, the data needs to be given back to the users.

Here’s a peek at some of the things we are working on to break through the noise and make good use of all the data we collect.


Smart contexts


We have a rich data landscape which has grown exponentially since 2010, when we founded Paper.li to help people cut through the noise of their Twitter timeline.

Our position as forerunners in the content curation world has seen our service grow to a point where we now process over 300 million social posts a day. We surface articles, photos and videos of interest from 144 million websites and blogs, bringing meaningful content to millions of readers everyday. The amount of social data around the content we generate puts us in a great position to do exciting new things.




“We are good at collecting data. Now it’s all about making sense out of it.”


But where does one start to figure out what to do? Hacking right in was our way! We had a lot of fun over the past months doing just that – hacking into our data landscape, looking at it from all kinds of angles and asking ourselves new questions.

We’ve been investigating the connections between people, what they read, what they share and who they share it with. We sifted through millions of articles and the corresponding social signals to uncover how to best track the interests and behaviors of our users, their communities, and how they change over time. Getting these key insights about what our users are doing or interested in at any given point gives us valuable knowledge to customize our future services.

“Context makes the service unique for each user; it makes it personal, efficient and successful.”


It’s all about contextual awareness.  The main outcome of our work has been to translate these insights into actionable queries and algorithms on our data platform.


Give data back to our users


Social data is a valuable currency to marketers and online advertisers who can use it to sell products and it is tempting to simply turn this data over to them. However, we know there’s so much more we can do with it.

Our whole team agrees that our data “intelligence” should be reserved for our users and be used to provide additional value to them. We have a mission to continue investing in contextual data analysis to find and develop new real life applications to give all this “smart” content back. Although, we don’t have all the answers just yet, we do have a clear vision and are full of energy for the long road ahead.

We believe it’s now crucial to invite our users to tell us exactly how they would like their data to work for them. We want curious, passionate and energetic people to join us Backstage at Paper.li to meet the team, join the data debate and have their say on how future products could take shape. You can join the Backstage community here.


Taste of the future.


As a very first step, we will soon be unveiling what we think is a neat little app.

The idea behind the app is to make life simpler for our users, allowing them to make use of our data in a way that will bring instant benefit. A Beta version will be available on iPhone and only English at first.  We’re releasing it in Beta to give our users an opportunity to have a say in how the next versions will take shape.



To wrap up, I can say, these last few months have brought us great insights into the value of our data at Paper.li and we are really excited about the possibilities ahead.  We’re looking forward to hearing what our community thinks about our path towards making data useful and how their needs will dictate our next steps to cut through the noise. With smart data and smart people, the future for content looks really bright.

Tell us, what use would you make of data?




Edouard Lambelet
Edouard is the Co-Founder and CEO of Paper.li. Tweet him @EdouardLambelet.

Published by


32 Responses to “Taming Big Data with Smart Contexts”

  1. Boyink

    I think we need some examples of the sorts of things you are thinking of – just to help frame the discussion.

    I’d love a searchable archive of the content that has run across my paper.li site – so that if I were thinking of traveling to a certain location I could find all the stories that other travelers had written about that mentioned that location.

    Or better yet – geotagged content that would appear as I move into different locations…

    But I don’t know if that’s simply a new paper.li feature or a use of the aggregated data you are talking about here.

    • Ed

      Boyink, Interesting thought on geotagged sources talking about a similar topic from different locations. Interesting angle.

      • Anthony

        Hi there! I have suggested that geotagged /redirected content to Kelly 1 year ago. This is a burning issue! 2nd: I use overblog.com to make an archive (with own domain) of my paper.li’s (see http://www.tangowriter.com and http://www.voxelflow.com) Overblog reads the RSS of my paper.li’s and offers a searchable archive.

        So far for now. Will jump into the discussion later on. Best, Anthony

  2. genaro espinoza

    well i dont have a comment only i want to no if i keep promote ur pepir.li ican make money o is only like a hobby im from mexico my inglish is not to well i wait for responds

    • Miguel Angel Perez Alvarez

      En mi humilde opinión si especializan su cuenta de Paper.li y seleccionan contenidos de la red que sean relevantes pueden conectar su sitio web o cuentas de Facebook y Twitter de su negocio generando así flujo y potenciales clientes.

      A mi me ha funcionado bien para atraer público y dialogar con ellos. Me han contratado y mi periódico de Paper.li me sirve de escaparate.


  3. lock piatt

    You have a great platform that we bloggers with projects can gain exposure and help others learn. So simple. Thank You.

      • lock piatt

        I include it my facebook, tiwtter and ning sites as visted. How can I embed and save time expanding the paper?

          • Mauricio Rodríguez Mojarro

            For Genaro..

            Amigo Genaro.

            El servicio de paper.li lo puedes comercializar si creas una gran audiencia. Pero para ello tienes que pagar el servicio .PRO, para que puedas poner tus anuncios. Pero paper.li no te da comisión alguna porque lo promuevas.

  4. Barry Barney

    I do not possess an iPhone – Android is better (AND CHEAPER!)
    How long before an android version appears? If ever?

  5. What just changed

    Thanks for sharing what your thinking about data. 🙂 the sorts of things I want to know are things like
    – who is coming, from where, when and why and how
    – what is the most interesting type of content they like to see – what is the point of difference that appeals to them
    – is there anything that indicates if single or multi topic coverage works best for them – a llittle often or less and more
    – what of all the sources you index for us to use is the most current – ie content sources that are always refreshing their collections not leaving them to languish and are quality assured in some way

    Just some first thoughts


    • Ed

      “– what is the most interesting type of content they like to see – what is the point of difference that appeals to them ” … This is exactly what we’re currently working on 🙂 Stay tuned 🙂

  6. Peter Young

    Loved the article, but was hoping for you to address the actual “value” of the data. That is, what does all of this “collection” equal in money and “time saved” and a variety of other tangible metric points. I could see a “Part II” to this article easily. I myself have been working on qualifying and quantifying these data points – so I am interested on a number of levels.

    • Ed

      Peter, Ok! Let’s talk about a “Part II”. Connect with us via email please : backstage at paper.li

  7. Paulo Caldeira

    Hi guys,
    I have Android, iOS and Windows devices but sincerely I don’t care about “humanoids” or “apple pies”. The major problem with the Big Data is: how to clean most of the trash at App Stores at the IoT Era?
    In my opinion, we don’t have a “Revolution” but an “Entropy”! Well, to give the data back to people we don’t need more “underware’s”, all we need is apps and projects that really get all connected giving context and localization.
    I don’t believe in a future where marketing is already death. The social web is about how we get in touch with people, we need to get back to the McLuhan concepts and read them correctly, leaving the absurd idea of a “smart” world without “smart” people.
    I use paper.li since the very begining but I don’t have paper.li at my mobile devices! Why? Well one of the problems is: localization! Paper.li doesn’t have Portuguese! When everyone use the English from England, the French from France, the Spanish from Spain, the portuguese is Brazilian!
    So guys: if the Big Data is a problem we all are making mistakes!
    And forget this “supergharge your marketing” slogan!

    • Ed

      Paulo, Thanks. The app you are talking about is on its way 🙂 And I fully agree : data is not the point, but how you turn it into actionable (!) actions. That’s what we’ll try to facilitate in our upcoming new features/apps. And we also second the idea of “smart people”; that’s exactly why we empower content curators.

  8. Jeremy Silver

    Hey Paper.li team first thanks for providing such a great service over all this time. People think I’m very diligent producing this newsletter every day – but you do all the work!

    1. As folks have already said, Analytics would help us – show contributors and subscribers, geo, demo data would all be very helpful.

    2. Ethical data sales – it’s ok to use my data to sell to advertisers if I agree and if I get a cut – figure out how to do that, allow me to choose the kinds of advertisers I would be prepared to sell to and

    3. Be a conduit between brands and users – connect people and brands with their permission in intelligent ways – match us with brands that suit our interests, allow us to retain control, switch off and on as and when we want

    4. consider extending beyond the wonderful world of Twitter to allow the incorporation of data feeds from other social networks and into other social networks – spread your net more broadly

    5 Semantic analysis – you should be able to do some really interesting AI, machine learning led zeigeist analysis and map that on a real time basis.That could be a fascinating consumer site to watch, but it would also be worth a fortune as a service that advertisers or agencies could subscribe to…. delineate the evolving topics of global discussion in real time as they shift and change by geo – not just a what’s trending, but a global view …

    Loads more where that came from. I’m in London – get in touch!

    • Ed

      Jeremy, Thank you. Let’s get in touch. Send us you details via email : backstage at paper.li

  9. Vanessa

    I like to cross promote my customers this platform makes it easier for the data to come to me so that the sharing and brand awareness is easier. The more chefs and food bloggers showcase images the more I can get our message out to the masses.

  10. Holly Massie

    The one improvement to your service we would appreciate the most is better filtering of adult content., We are aware that the paperli papers we have created represent us to the wider world and as a family friendly charitable choir it is imperative that pornography not end up as part of that image. We have already given up our popular paper, Audition Hub, because of the hardcore adult images that slip in from twitter. i looked at all the filtering options available and cannot see how to improve the selection. Note that we are a nonrofit group and cannot afford to pay for the service ( we do appreciate that we can use it at no cost) As far as I can tell there is no way to block certain words that may be contained in the source. I can think of many words I would list that may block the disgusting images that the word audition calls up.
    PR for The Stairwell Carollers

    • Julia Yaziji

      Hi Holly, First, I’m sorry unwanted adult content was pulled into your paper from your #tag search on twitter. We’ve written a “best practices” article which will help ensure that unwanted content will not appear on your paper:
      Unwanted Content, Porn Or Adult Content

      If you have more questions or need more personalized help, please contact me by opening a ticket. A link to contact us will be found in the footer of the article.

  11. Olga Cohen

    You use to print almost every article that I wrote on my blog wwwblog@olgassecret.com but stopped posting them last summer. I use to get quite a few responses from your readers and I also promoted menopause news.
    Just would like to know why you stopped printing my blogs and how do I get you to review this situation and get them up and running again.
    Love your publication and find it very useful as it reaches women all over the world.
    Thanks Olga Cohen

    • Julia Yaziji

      Hi Olga, I visited your blog and we should still be able to pull in your content. Why don’t you visit our help center and open a request. We’ll be happy to help you.

  12. Mark Palmer

    Your post on big data made my day! Thanks to paper.li the archives of The TBI Daily represent one of the largest, if not the largest collection of articles on TBI. The ability to search the archives can be a game changer on how survivors and family members travel the journey of life with brain injury. If I can help in any way, please ask.

    • Ed

      Thank you Mark.
      I’d be happy to learn more. Could please send your details at backstage at paper.li and we’ll be back to you.

  13. Jess Anders

    I’m looking into using Paper.li to replace/update/enhance our traditional weekly newsletter. Curating the content is time consuming, and not very interactive for our people. But for me, the big drawback is the up front time and the lack of statistics. Knowing which articles are ranking over the course of a month is important, but also knowing open rates, click through rates, how much time people spent reading it would also be helpful.

    I know this may not be the right platform for our audience or what we’re trying to do, but it’s unique and different enough that I’m trying to test it out. Data is great, but it needs to connect back to our business purpose and help clarify whether we’re “moving the needle” in the right direction.

    Happy to discuss further!

    • Ed

      Hi Jess. Happy to discuss further too. Especially regarding statistics and “up front time”. Please send your details to backstage at paper.li and we’ll be back to you.

  14. Mauricio Rodríguez Mojarro

    Hi Ed, i would like one improvement to our service – i said our, cause i love/paid – is: That we can write articles directly from admin desktop, combining curate content and personal content – not from personal blog -, directly from paper.li desktop.

    And when we share content in facebook, the link to the article goes to external webpage, but i will like that the link bring the user to my paper.li.


Leave a Reply

Your email address will not be published. Required fields are marked *