Category Archives: Building business

Adopt a prison campaign idea

One of our sponsorship opportunities that we wanted to promote further was a campaign to ‘adopt a prison’. We recognised a need for supplying books to prisons, especially for matriculants.

We identified three potential leads to possibly connect with in the future:

To date we haven’t been able to follow up on this.

Reading Ries and Christensen

I’ve done some fascinating reading recently, too. The most influential books I’ve read recently are The Lean Startup by Eric Ries and The Innovator’s Dilemma by Clayton Christensen. Both have left me reeling from ah-ha moments. Chief among these:

  • Ries’s explanation of how a startup’s product — what it aims to produce in its early stages — are not the things it sells, but the lessons it learns. The faster, cheaper, and more specific you can make what he calls ‘cycles of validated learning’ (build, measure, learn), the faster you can produce something that people want.
  • Ries’s description of vanity metrics vs cohort matrics. Vanity metrics are simple growth curves: ‘we signed up 10 free users last month, and 100 free users this month; 1 user upgraded to a paid subscription last month, and 10 this month’. These metrics always look good, because they show growth in every area. Here, tenfold growth in users and paying users. But they aren’t valuable metrics, because what matters — what tells you whether you’re *getting better as a business* is not growth in users or paying users. It’s how fast you’re increasing the *proportion* of paying users to free users. Cohort metrics are more valuable: ‘what percentage of users are paid users, and how fast are we growing that cohort?’ This totally changes the focus of a sales team.
  • I wish I’d read Christensen ten years ago. Big aha: established companies don’t invest in new approaches (like Paperight) not because they’re stupid or evil, but because they’re fundamentally unable to do so, even if they’re the best-run company in the world. Christensen describes how. This totally changed the way I approached large publishers.

Very few people get through a business conversation with me without hearing about these at least once.

Collating matric exam packs and starting to measure metrics

Michal’s internship finished in March 2012, and Nick began an internship as his replacement.

Our first priority was preparing packs of past matric exam papers. We’d started to source these as part of our initial content list creation, and these were already listed on the site, but the packs themselves had not been prepared. We needed to have them ready in case any orders came in. The primary challenge was creating complete sets of exam papers. The DBE website and WCED didn’t have all of the papers, and their online resources were often buggy or incorrect. We started by creating a list of outstanding exam PDFs, which we then used to individually source as many missing papers as we could (we called and emailed, and bought CD compilations of exams to try to fill the gaps). At the same time we started prepping the packs for those subjects which we had complete sets for.

Nick and I attended some ‘Open Education’ workshops at UCT, in the hopes that this would generate some leads for more content. We found, however, that we already knew much of what the workshops covered (but it was edifying to know we were on the right track).

The aim was that interns or new staff members could jump right in on tasks with a little training, and begin to develop skills themselves. Over the years this has worked incredibly well for the content team. It means that when we do in-person training in those first weeks, it can be much more in-depth (and is thus more valuable) than if we were to do general introductory training sessions.

I began creating and improving upon a series of wiki posts to govern things like document creation and document uploading. The aim was that interns or new staff members could jump right in on tasks with a little training, and begin to develop skills themselves. Over the years this has worked incredibly well for the content team. It means that when we do in-person training in those first weeks, it can be much more in-depth (and is thus more valuable) than if we were to do general introductory training sessions. Ops style posts that give detailed explanations of how to do tasks means that new and old team members alike have something to come back to for reference, and ensures uniformity (which is important when it comes to file naming conventions for version control).

We also began to track metrics for the first time. Our initial focus was on measuring publisher registrations, outlet registrations, and top-ups (i.e. the purchasing of credits in advance). This process of tracking metrics was one that we improved upon over time. It’s interesting how much insight our focus on these three metrics gives to our business goals at the time. We were focused on creating an outlet base, and increasing our content bank, rather than on growing our customer base. And we were more focused on the potential for sales than on sales themselves. The failure here was in assuming that these three metrics were a proxy for other things. We assumed that a wide outlet base represented more potential customers, that increased publisher registrations meant more content (and that more content increased the likelihood of valuable content), that top-ups were a signifier of outlet buy-in, and would ‘naturally’ lead to sales. The reality was that we ended up measuring the potential for success, rather than measuring success itself. It was a lesson we would learn later on.

Publishers approached

  • WITS
  • Hamilton Wende

Publisher registrations

  • Cingela (13/3/2012)

My fellowship newly underway

So, I’m three months in to my Shuttleworth Foundation Fellowship, which is three months into building Paperight full-time. If you don’t know, Paperight is a website that turns any business with any printer into a print-on-demand bookstore. So, what have I been doing with that time?

The first thing has been to get a working demo, or prototype site, up and running, so that we can show the service to others, test some ideas, and develop our vocabulary and sign-up documents in a live environment. So I knocked that together in WordPress during September, along with a bunch of back-room workflow tests and documentation. It’s been hugely valuable.

With that done, it was time to start talking seriously to rightsholders, licensing agencies and content aggregators. So in October I headed off to the Frankfurt Book Fair and London to speak to a wide range of people. And the response was, almost entirely, overwhelmingly positive.

Two years ago, when I first asked publishers about the Paperight concept, they were cautiously optimistic, but many were worried about how their books would look, and how much they would cost to consumers. Luckily for us, since then Amazon Kindle has shown that most readers just want stories and info, and that easy, affordable distribution is often more important than high-end production values when you’re growing a market. Suddenly a book printed out on A4 paper seems just fine. Especially if it’s on every street corner in countries you’ve never sold in before.

So there were far fewer concerns from publishers about Paperight in 2011 than in 2009. Where there were concerns, they have been really helpful in tailoring our message. I certainly have a much better idea of what makes publishers interested in using Paperight. One key issue – which I discussed recently on the Paperight blog – is that Paperight can compete with piracy on accessibility, convenience, and often in total cost (energy, time, money).

Rightsholder agreement

Our messaging is captured largely in our rightsholder agreement, which is really short, and in plain language. It took a lot of time and effort to get it that way. This is really important to us, because Paperight is built on the idea that the once arcane world of rights and licensing can actually be managed simply, and anyone can participate in it. I went through the distribution contracts of a bunch of other businesses, took the most important concepts, and boiled them down to simple sentences and paragraphs. The input of Foundation alumnus Andrew Rens was really valuable here, too. It’s something we’ll constantly evolve, but I’m pleased with the way we’ve started.


Another important area of our messaging is pricing. Most people find it hard to believe it can be cheaper to print a book out than to buy a copy that the publisher printed in its thousands. But now we can show in most cases that that isn’t true. In the video that goes with this post, I give a concrete example of how a publisher can earn as much from a Paperight sale as from a conventional book sale, and yet save the consumer more than 25% on the retail price of the conventional edition.


My conversations with rightsholders and others have also led to discussions about putting a range of non-book content on Paperight, including newspapers, exams, sheet music, classifieds and administrative documents.

The process of prioritising and prepping this content will fall to our content manager.Tarryn-Anne Anderson joined us in November to work on this. Over the next couple of months, she’ll also be putting together a print catalogue of books and documents we think people will like, and we’ll put that catalogue in copy shops around the country. It’ll include textbooks, novels, past matric exam papers, how-to guides and more. And from that we hope to learn more about what print-shop customers are likely to find most valuable.

The website

last-screenshot-live_20120509_10-43pm_cropMeanwhile, all along I’ve been working on a redesigned site that will replace the working prototype in the first half of next year. It’s simpler and will be much faster. And it’ll give us the ability to distribute certain documents in certain regions, which is crucial to publishers who want to reach new markets without competing, for now, with their conventional editions in their home markets.

This means long hours studying and developing user interface and user experience best practice, and chatting to print-shop managers about how their stores work, and how the Paperight site can best work at their point of sale.

Here’s an early mockup of a product page, prepared long before I built the prototype.


The Shuttleworth Foundation

Working with the Foundation has been fantastic. I get to share ideas with and learn from a group of seriously amazing people, who’re working in mobile technology, user-created publishing, biocultural communities, open knowledge and educational resources, peer education, open data, citizen cyberscience, new approaches to IP, and more. And the Foundation staff work tirelessly to support our work and help us focus on making an impact. They all make the Paperight team much bigger than it seems on paper.