The how and why of making ebooks out of conferences

 by Martin Belam, 14 November 2012

If you are a regular reader of this blog, or follow me on Twitter, you might have noticed that I’ve recently produced free ebooks of my notes from conferences like EuroIA, UX People, and the WAN-IFRA Tablet and App summit. I thought I’d explain some of the how and why of making them.

Why do it?

I think EuroIA in Paris in 2010 was the first time I did an “All your slides are belong to us” blog post. Whilst I was at an event I would be constantly updating a page listing all the talks, and the places that you could find the slides or blog posts or essays about them. Kind of like Lanyrd does now, but by hand. Every couple of hours or so I would tweet out the URL. It provided a service to people at the conference or following it from afar, and helped me drive a lot of traffic to my blog and pick up new Twitter followers.

Ebooks take this a step further. I enjoy writing and blogging, and so it is almost inevitable that having attended a conference I’ll write up quite a bit of it. The ebooks make that content work a little bit harder for me.

Why do it for free?

The business model is pretty simple. I’m making the content anyway, so packaging it in the ebook is an incremental effort that, hopefully, increases the reach of the content. I’ve written it, so I want it to be read.

Writing a few thousand words in the space of a couple of days is obviously a big effort, but there are a couple of reasons I don’t charge for the books. Firstly, although everything is written by me, the book exists because of the generosity of the people who have given their time to talk, and the people who took the time and risk of setting up the event. I’d hate people to think I was trying to profiteer off their efforts to make a quick buck by selling an ebook. Even if it only ever upset a couple of people, the tiny amount of money it would generate isn’t worth it.

But the ebooks do something important for me. Everybody who downloads one has my contact details, and the contact details for my business, on their ereading device of choice. I see it as a valuable marketing tool for my consultancy services.

Here is a little bit about the process I go through to make one of these types of ebooks:

Do some preparation

Before the conference I already usually know the titles of talks and who will be speaking. I can therefore put together a template for both the ebook and the blog posts in advance. Often I’ll do this whilst on my travels to an event. Even just making a list of the names of people talking at an event, and wrapping them in hyperlinks to their Twitter profiles or websites will save time at the authoring or editing stages. Anything you can do in pre-production reduces the amount of time you have to spend working on the book after the conference, and so means it can be released more quickly.

Make good notes. And type quickly.

Making good notes is really important if you are going to try and turn around this volume of content in a short space of time. I always try to make sure I get down some key facts, and any URLs that are mentioned. I’m lucky that I can type really quickly. I use TextMate on my laptop, and IA Writer on the iPad to make my notes. On TextMate I have some shortcut key-combinations and macros that I’ve got set up to help me along.

One key thing I do is always listen out for great turns of phrase or statements that will make good verbatim quotes from the speaker. It makes sure my write-up conveys their personality as well as my own writing style. I make sure that if I’ve taken down an exact quote, I mark it up with quotation marks or as a <blockquote> at the time in my notes — I can do this with a two key combo in TextMate.

What I try not to do is simply regurgitate a blow-by-blow account of what someone has talked about. I think that is a bit like “stealing” their talk. Instead I always try and make sure to relate the points they are making back to some experience in my own career, or to link it with some other ideas I’ve heard on the day, or previously blogged about. I find 600-1,000 words is about the right length to convey an overview of a thirty or forty minute talk.

Making the blog posts

Once I’ve got my notes, I’ll try and write up each session. If I can, I’ll usually try and write at least some of it up in the coffee and lunch breaks during the conference — essentially trying to balance the requirement to network face-to-face at events with my introspective desire to just have my nose in my laptop.

I then usually do a mammoth writing session in the evening as soon as the conference has finished. This is particularly useful when I’m away from home on business, as it stops me going out, getting hammered on expensive cocktails, and then having to explain the state I’m in when I get home the next day. Instead I studiously find a quiet spot where I can focus on writing — admittedly also usually where they serve beer.

I write my blog posts using HTML mark-up in TextMate. Once I’ve got them all into a reasonable state, I’ll upload them into my blog CMS as draft posts.

Making the ebook

The next step for me in making the ebook is to use Pages on the Mac. I’ll copy and paste the text from the draft blog posts into a Pages template I’ll have already made for this particular ebook. Copying and pasting from the browser to the word processor is more than a little tedious, but for me the advantage is that it turns all of the HTML entities I’ll have included for things like accented letters and curly quotes into the actual required characters, and links from the blog post become embedded hyperlinks in the ebook document. You do have to keep an eye out for losing formatting for bold and italics sometimes though.

I’m careful to use styles properly in Pages, so that the title of the book is Heading 1, each chapter title is a Heading 2, and sub-headings within articles are Heading 3 and so forth.

The template I made before the conference started will already have included space for a book cover and some publishing blurb at the front, and a table of contents.

Making a book cover

I’ve kept my cover designs painfully simple — one block of cover and the title. In part this is because you need covers that work well in the tiny thumbnail sizes displayed in iBooks on the iPhone, or, if you are selling them on Amazon, in tiny thumbnails on search results pages and elsewhere. I found the colour block approach just simplifies the whole process, so I don’t have to worry about it. I have a Photoshop template for the covers, which keeps them all at the same dimensions, and I usually pick a colour from the logo of the conference or the organisation organising it. I make a joke of listing the exact hex code for the colour inside the book as the “cover star”.

EuroIA Rome 2012  UX People 2012  Tablet and App summit 2012

The rather generic designs for my ebook covers

Editing and version control

Proofing and editing is really important. I can’t afford to get it done externally — I really miss the trained second pair of eyes I used to get scrutinising my writing when I worked at the Guardian — so I need to try and do it the best I can. I usually go through two editing processes.

Firstly, if possible, I’ll print out the whole of the book. Despite being an utter technophile, I find it much easier to proof and edit on paper, where you can cross things out and make notes in the margin. I also think it is important to try and read the book through in one sitting. This can help alert you to repetition, overuse of jargon, or stylistic tics which you might not notice by looking at each blog post in isolation.

Secondly, I’ll make an interim Kindle edition. I like proofing on the Kindle, as it is easy to use the note-taking facility to highlight problems. Another useful thing I find about checking on the actual Kindle device is that I can see how the contrast of any images is rendering, and adjust them accordingly.

Once I’ve done the editing, proofing and note-taking, I then go in to fix the original documents. I fix problems in both the blog post drafts and the Pages document at the same time, having them open side-by-side. This (mostly) prevents version control issues. I’m not suggesting I’ve hit on the ideal workflow here, but it works for me.

Exporting the different file types

I use Pages on the Mac OS X, and it is a secret weapon in the war on making ebooks easier to publish. You can export to PDF, and you can export as an .epub file. This format will happily work in iBooks, and there is a dialogue box that pops up that allows you to set the title, author and category you want to define the book to be. You need to have inserted the image you made for the book cover at the start of your document, because Pages asks you if you want page one of your book to also be the cover image.

Then the Amazon Kindle Previewer tool allows you to generate a .mobi file from an .epub file. Job done.

Well, sort of. The transition to Kindle isn’t perfect — notably setting up a table of contents in Pages does not generate a valid TOC in the resulting Kindle file. I’ve looked and looked and not yet found a workaround for that. But the .mobi file will work across Kindle devices.

(Incidentally if anyone does know how to get Pages to spit out a proper Kindle TOC, I’d welcome the advice.)

Promoting the ebook

A few years ago, if I wrote a long blog post or a series of blog posts on a theme that I was going to make available as a PDF, I would wait until I’d published all of the individual posts before making the compilation available. I used to have advertising on the site, and I was more interested in the page views than the downloads. The reverse is true now, so I want to make sure that the ebook is downloaded as much as possible. I make the compilation available as soon as the first blog post is published, so people are encouraged to download it to read the as-yet-unblogged material.

In the opening paragraph of each of the blog posts I set the scene about the conference, and then point out that all of the blog posts are available to download as the ebook. I hope that having clicked onto that particular page on my website because the headline has caught someone’s eye, the prompt to download the material immediately will make them download the ebook on impulse too.

I also add a listing for the ebook on the “Books” page of this site. I use a HTML anchor tag so that appending something like #euroia2012 or #wanifra2012 will act as an easily tweeted direct link to the blurb about that ebook.

I make sure that I tweet links to the ebook a few times using the relevant conference hashtag. One of the reasons I try and turn the production of the books around so quickly is so that people will hopefully still be paying attention to the event’s hashtag.

Interlinking the blog posts

I like to interlink all of the blog posts that appear in a conference ebook on the site, and I do this in a few different ways.

Firstly, every blog posts ends with a one paragraph trail for what I’ll be publishing next. When I first publish a post, this is just an informative teaser, but I make sure to go back and add the link in as soon as the next post in the series is made live. This means you can start on post #1 from the conference, and at the end of each blog post simply click on a link to be taken to the next one, without hunting around for the usual blog-style “previous” and “next” navigation.

I also sometimes recap which blog posts I’ve previously published in the opening of each post. You’ll see sentences like “I’ve already published my notes on the talks by <LINK>person x</LINK> and <LINK>person y</LINK>. Now here’s what I made of person Z’s talk about spectrox toxemia”

Finally, I put a “text include” at the foot of each blogpost. I do it with the tiniest bit of PHP known to man or woman. You’ll notice the URL of this page ends with .php, and that means the server that runs this blog looks to execute any PHP code embedded on my pages. I simply finish each blog post with a line that looks something like this:

<?php include('/some/path/to/textfile/conference2012.txt'); ?>

That tells the computer to look for a textfile called conference2012.txt at the specified location, and insert it into the page at this point. The textfile itself will simply be a fragment of HTML, that says “This is one of a series of blog posts about conference x”, and then lists all the blog posts in order. It means that every blog post is linked to all of the other blog post via this “include” on the page. Changing that bit of text across all of the related blog posts only involves changing that one textfile. The “include” also features a final call to action to download the ebook version of the conference nights.

Measuring success

I’m a great advocate of not doing anything on the web unless you’ve worked out how you are going to measure the success. Publishing an ebook is no different. So I use the Google Analytics Event Tracking feature to measure how many people download the books. Links to the downloads include some extra mark-up that looks a bit like this:

onClick="_gaq.push(['_trackEvent', 'Ebooks', 'Download', 'Conference X iBook']);"

That means that in Google Analytics I get a report telling me how many ebooks have been downloaded, and what conferences and formats have been the most popular. If you are interested, here are some instructions on how to use the Google Analytics Event Tracking features.

From the analytics I know that PDF is actually still the most popular format for people to grab the material in.

Why not try one out?

Interested to see what the finished product is like? Why not try one out — in the last few months I’ve published ebooks about Euro IA (iBooks | Kindle | PDF), UX People (iBooks | Kindle | PDF), and the WAN-IFRA Tablet and app summit (iBooks | Kindle | PDF).

1 Comment

I've had plenty of wrestling with TOCs with different eBook formats—apart from attempting to generate page breaks in Pages it's the most complicated thing I've found to do. My work around has been using Calbre to do the conversion (and the 'generate TOC' with expressions function) , but as it can't convert from epub I've been basing both that the mobi on a html version.

And the html version has to come from Word, which is a pain.

Keep up to date on my new blog