Nine months into our Shuttleworth Foundation funding, I’m proud and pleased with where we are in large part because, in getting there, we’ve had to learn fast from mistakes and successes. Some of the lessons we’ve learned:

• The human story is more powerful than the financial one. I thought initially that publishers and copy shops would sign up because we offered them a new revenue stream. However, at first glance no one really believes it when someone promises them a ‘new revenue stream’. They really make their buying decision – which is always an emotional decision – because they connect with our social-impact vision. Then, they go on to justify that decision to themselves by calculating potential revenue, or by citing a need to look for new opportunities in tough times. Similarly, I’ve seen that when individuals don’t connect with our social-impact vision, they use the financial numbers to justify not participating. So most of our Paperight pitches now emphasise the human story – books are the key to upliftment, they save lives, we all have a responsibility to spread education – and then when necessary we move onto the numbers. This was a crucial lesson that took months of trial and error to learn.
• Perhaps the hardest lesson was realising that three months approaching big publishing companies early on was not a good use of my time. Paperight is a classic disruptive innovation: a simple, relatively low-margin product for a new market. No matter how well run they are, established companies cannot justify putting resources into a disruptive innovation very early on. They can only follow smaller, more nimble players for whom new, early-stage markets are attractive. (I wish I’d read The Innovator’s Dilemma sooner; it makes this so clear.) Now that we have a growing stable of publishers and outlet footprint, it’s easier for larger publishers to justify joining us.
• When pitching Paperight to outlets, it’s good to focus on the word ‘legal’. I initially emphasised concepts like ‘easy’, ‘more customers’ and ‘broader product offering’, thinking that the fact that Paperight is the first ever legal way to print books out was obvious and beside the point. While those features are important, we’ve also realised that ‘legal’ is a key feature: copy shops know they are often asked to copy books illegally, and this creates anxiety for managers. Our pitch then speaks to that emotion.
• Building good software requires patience and impatience simultaneously: planning and designing Paperight 1.0 (the current site) took much longer than expected. We had to very patiently thrash a great deal early on, and this paid off in a very smooth build process that resulted in a great site. But none of this would have been possible without the impatiently built Paperight 0.5, a duct-tape solution on which we impatiently registered our first users and delivered our first documents. Even though we had to use 0.5 for two more months than expected, we learned from it right to the end. The lessons included refining terminology, online agreements, book metadata and taxonomies, customer expectations around document quality, marketing strategies (customers love free credit more than books priced at free, even though they’re effectively the same thing), and search and browsing behaviour.
• My initial strategy was to create a large catalogue early on so that users could ‘walk into an outlet and ask for anything’. This was flawed – and not just because it’s very hard to build a large catalogue fast. The flaw is that with a new service, too much choice is paralysing. To gain new outlet sign-ups, we had to focus on one product: past exam papers for grade-12 learners. We have since got much better traction among outlets, who can visualise marketing that to their customers. We learned this lesson while distributing our first Paperight catalogue poster, and watching how outlets engaged with it. (That said, it’s important to note that many users want to browse a range of books not to buy but to evaluate the service before signing up.)
• Unique, tailor-made content is hard work but incredibly valuable. Creating packs of past grade-12 papers involved a serious investment of time and energy. (Nick Mulgrew tells the story on our blog.) Essentially, we’re creating this content from disparate sources (no one organ or government can provide all matric past papers; we’ve had to visit various offices, numerous websites, and beg favours of officials). It is possible that the creation of Paperight-specific content may form a key part of our content strategy over time – potentially more important for growing our customer base than simply gathering others’ content. This is something I’m keeping an eye on.
• A key future revenue model is selling integration with institutions’ user systems to deliver documents to specific people in remote places – for example, distance-learning students picking up their personalised printed course materials from a copy shop, using a code or student number plugged into Paperight, rather than relying on the post. However, to get in the door of large institutions – universities in particular – the outlet footprint has to be in place first. The first question I get is always ‘Where are your outlets?’. It’s a market where vaporware doesn’t cut it. In our first six months, this was a setback that wasted time. Now that our footprint is growing, we can begin making these pitches again.
• Copy shops don’t want to be selling advertising. We had reserved advertising space on the pages of our documents for copy shops to sell to local businesses. It seemed like a good idea. But, for a copy shop, the cost of acquiring advertising is much greater than the likely advertising revenue. We’ve discovered, however, that publishers are interested in using this ad space to cross-sell books. So I’m looking into this ad space as a potential revenue stream for Paperight instead, potentially using it to offset rights fees.