Paige: Your personal Librarian

Paige organizes your book notes and highlights. It syncs with your Kindle and you can easily add your own comments and tags. Paige's mission is to help people learn more effectively through reading.

This project is currently in beta development, due for release in Q1, 2017. Get your beta invite:

Case study highlights

  • Why reading is the ultimate meta-skill, and how to learn more effectively through reading.

  • The challenge of focus: how to prioritize product design, and the role of research.

  • The thought process - sketches, wireframes, tradeoffs, and how we brought this unique vision to life.

Reading is the ultimate meta-skill. If you learn how to read, you can learn to do almost anything. Whether you’re challenging your imaginative boundaries, or picking up new skills, reading is how you take control of your own education.

Like any skill, reading takes practice, and has many stages to mastery. There’s no right or wrong way to read, but there are certainly more effective and less effective methods, depending on your goals for reading.

For non-fiction books, one common “mistake” is reading them word-for-word like they’re great works of fiction. Fiction invites us to enter the author’s world for a deeper, more engaging experience, whereas nonfiction seeks to inform and elucidate the author’s arguments. When the goal is to learn, you don’t need to read these books. You need to understand them.

The more we read, the more we practice. We start to build a knowledge base on a variety of topics, and can more efficiently process new information relative to existing knowledge. The more we practice, the easier it gets. Well, up to a point.

Have you ever read a book, only to forget even its main points months later?

Learning is a process that requires frequent application and usage or revision. What we don’t use, we forget. While there’re a range of both new and time-tested note taking tools, they focus only on the “input” side of the information equation. Organizing these new notes, and recalling them for application, remains largely left to our brain. Although our brains are still the most versatile problem solving machines (for the time being), let’s face it, declarative memory isn’t one of its strengths. As Einstein famously said,

“Never memorize what you can look up in books.”

In the age of Google and Wikipedia, it has never been easier. Paige’s mission is to help readers organize and access knowledge from books they’ve read just as easily and intuitively.

One book to rule them all

Technology has touched upon the way books are written, published, distributed, and consumed. Though physical books and the “smell of an old book” remain an irreplaceable part of many readers’ lives, there’s no denying the impact ebooks and ereaders have had on changing reading habits, and creating new audiences.

Paige was founded by a team of developers and book lovers. As someone who has set a goal of reading 52 books this year (I’m currently nine books behind I’m ashamed to say), I jumped at the opportunity to take on such an audacious challenge. Organizing and retrieving book notes and highlights was just the beginning. If all goes to plan, the potential was much bigger:

The value of information is limited in a vacuum. Ideas don’t stand alone. The true potential of ideas can only be realized when they exist in a cross-linked network, like Wikipedia. As Kevin Kelly, founding executive editor of Wired magazine, describes in his book, The Inevitable,

“Turning inked letters into electronic dots that can be read on a screen is simply the first essential step in creating this new universal library. The real magic will come in the second act, as each word in each book is cross-linked, clustered, cited, extracted, indexed, analyzed, annotated, and woven deeper into the culture than ever before.”

In this single networked literature, incorporating texts past, present, and in all languages, we’ll have a clear sense of what we as a species do and don’t know. But, we’re getting ahead of ourselves here…

Keeping it simple and focused

Rather than build the electronic Library of Alexandria straight away, the goal for the MVP was simple: create a simple product that makes it easy to sync, organize, and find readers’ notes and highlights. There would be no social element yet, but it would allow us to validate or invalidate some core assumptions about our target market and their reading habits.

In the early stages of a project, there is never a shortage of things to do. To reduce complexities and make sure we focused on the most important, we constrained ourselves to two main areas:

  1. Content organization — what system is most intuitive to organize and display content from a wide range of sources?

  2. Information input — if there is a lot of friction in onboarding, before users realize the product’s value, user activation rates would be dismal. How can we make onboarding, import, or sync as friction-less as possible, while not sacrificing value?

Of course, these were only our initial hypotheses of what we have to get right for the product to be successful. To test those assumptions, we had to get out of the building and talk to our users.

Just enough research

Getting feedback is the best way to measure progress. It’s a key step in any design process, and whether qualitative or quantitative, we set clear goals that provide actionable results.

In the conceptual phase, we relied on user interviews to understand what direction works best. Based on the initial key audiences hypothesized, we conducted interviews with 53 people in the target groups. This allowed us to understand existing reading and note taking processes, and identify the pain points.

We established 3 main personas based on interviewees’ different reading processes, book preferences, reading frequency, and goals. This process narrowed the target audience of the Paige MVP to non-fiction, iPhone carrying, Kindle readers who read 20+ books a year, with an existing note taking habit, for the purpose of professional improvement. Organizing and the retrieval of information surfaced as the biggest pain points.

Quantitatively, we ran rapid design tests with users to take some of the guess work out of a range of design decisions. The bookmark and themes icons for instance, were selected through some quick A/B tests. The process was a scaled down version of our design process: set clear goals, create design elements or prototypes for the test, test, then analyse the results.

Multiple small tests would often run in parallel and on the same day. So while the test designs weren’t perfect, they conveyed ideas clearly, and surfaced the most promising for further exploration.

The pleasure of finding things out

The Paige brand aimed to encapsulate an intuitive organization of ideas, and the adventure of discovering new ideas through reading. With these guiding values and a deep understanding of the problem, I began by sketching out some initial user flows and concepts:

Some early concepts


Some early concepts

Sketches allow everyone to quickly visualize and identify what works and what doesn’t. The best ideas are then selected, amalgamated, and further explored through wireframes.

Wireframes: laying the foundations


Wireframes: laying the foundations

During wireframing, we crafted accurate copy to reflect how “real” user content might look. This ensured variations in content length and formatting were taken into consideration. We could also test the robustness of various layouts when displaying edge cases.

What we made

For the MVP, the focus was on getting the UX right (or as close as possible), and making sure Paige solved the problem.

Simply Sync

The Paige experience starts with content — readers have to first add or sync their content before the product can add value. This step is perhaps the biggest “ask” of our users in time, effort, and trust; therefore, we needed to make it as frictionless as possible.

We considered three methods of input: syncing with users’ existing ereader accounts, image-to-text photos, and manual input.

Sync, image-to-text, and manual input


Sync, image-to-text, and manual input

By constraining our target audience to Kindle readers, we reduced the scope of the MVP to just sync and manual input. Sync also requires the least amount of time and effort, and can populate the app with user data instantaneously.

Once synced, users can see their collection of book notes and highlights in the “Library.” Here, notes and highlights are sorted by book. Each individual note can be bookmarked or tagged.

Find individual notes from the Paige library


Find individual notes from the Paige library

Personalized organization style

Tagging is the main method of notes organization on Paige. Users can tag each note, or a selection of notes by themes or categories of their choosing. Later iterations will incorporate elements of machine learning to scan each note for key words to streamline this process.

From our research, book note-taking systems differed a lot between each reader. We learned about systems including ones built on top of Evernote, to hand-written index cards organized thematically.

Index card book note-taking system. Courtesy of Ryan Holiday


Index card book note-taking system. Courtesy of Ryan Holiday

Therefore, we wanted to create a system that is intuitive, easily customizable, while being scalable enough to be useful long-term as users add more and more content.

Drawing inspiration from email and note-taking apps interviewees use, we settled on a simple customizable tagging system for the initial design. When users tag their notes and highlights, it links similar ideas from various sources together. This is similar to how our brains organizes new ideas. By linking ideas semantically, it also makes it more memorable, and easier to recall.

Linking and remembering key concepts by theme


Linking and remembering key concepts by theme

We also learned that users revisited their book notes for a variety of reasons and in different contexts. This may include writing a book, article, or blog post; revising for an exam, or as a habit to internalize key ideas.

If certain key words can be remembered, users are likely to prefer the search functionality. But for more general browsing and brain storming, the thematic browsing is preferred. We designed the main themes view with this in mind. This was about flow — quick ways for users to swipe through saved notes, and also about serendipity — of discovering old ideas that haven’t been revisited in a while.

Testing our assumptions

Usability testing time!


Usability testing time!

We observed 22 attempts at specific key tasks with users throughout the design process. The success rate has steadily improved to over 60%. The data collected helped identify navigation issues, informed design decisions, informed the product road map, and also highlighted areas for improvement of Paige’s branding and positioning.

Usability test results for select key tasks


Usability test results for select key tasks

What’s next?

Paige is currently in beta development, due for release in Q1 of 2017. Initial prototypes received positive reviews, with early testers offering to pay to help accelerate development — also super useful data for Paige’s pricing strategy.

We’ve also discovered Paige’s branding needs more work. Initial survey results have indicated the messaging does not convey the product with enough clarity.

Analytics will also be implemented prior to launch to evaluate the success of the product’s design. Key metrics to evaluate include user activation rate, user retention, and also navigation flow tendencies.

What I learned

Spending time on research saves even more time in design.

As a designer, especially for projects I’m excited about, my mind would be bursting with ideas that can catch me like a runaway train. Jumping head first into an idea is, more often than not, not such a great idea. Without first understanding the problem and who you’re solving it for, you run the risk of building something no one wants. The first rule of a great product is to build something people want [link to quote], and the best way to do that is to get out of the building and talk to users.

Focus. Everything is not a priority.

Figure out what the right things to do are. What are the most important? What is a need vs. a nice to have? Like the point above, speaking to users will help narrow down the options. You’ll get a much better product if you can really nail a few things, rather than being so-so at a bunch of things.