Writing Website Test for Nonfiction Authors

Most of what you read about web design and web marketing is based on a decade of third-hand anecdotes and old wives tales. To find out what works for nonfiction authors today, the only fair test is to build a new writing website from scratch and see how it performs over time. I set out to do that a year ago, and wrote a series of posts updating readers on the progress of that website. The first post discussed the goals for the new website, the second was a progress report on search and video traffic and the third was focused on YouTube for writing websites. Today I'm going to wrap-up the first year of that website and start a parallel test with a major new section on the FonerBooks website.

The IFITJAMS.COM website is collection of about 20 static pages and a dozen blog posts on the subject of automotive troubleshooting and repair. I don't have any "authority" in the field, I'm strictly a shade tree mechanic, so I think it's a fair proxy for a new author or publisher starting a website with a project in the early draft phase. The total visitor count for the first year fell just shy of 100,000, and the site is now averaging around 400 visitors a day at the end of the slow season.

It's important to note that there are no community features on the website to inflate the visitor numbers. I didn't add a discussion forum or any scheduled content, request subscriptions, etc. Around 80% of the visitors come from search. The top ten search phrases for the month are shown below:

Currently in the sixth spot is the full domain name, IFITJAMS.COM, and just missing the top 10 was the partial domain name, without the COM. I've noticed that many people today navigate to websites they remember by typing the domain name into Google rather than putting it in the address bar, but I believe that in this case, most of people arriving after searching on variations of the domain name are really coming from YouTube. The graph below shows the YouTube video view count, which is on its way towards 1,000 views a day, and the total number of video views for the channel is already over 200,000.

In addition to the little single-take automotive repair videos being far more popular than my self publishing harangues, it turns out that viewers find them far more compelling. The final graph shows the attention span for my top ten auto repair videos, all of which are well over 60%, meaning that people usually watch them straight through:

Before moving on to the new website experiment, and why it's necessary, a few observations on what worked for IFITJAMS. First of all, content worked, both on the website and on YouTube. As an author or a self publisher, content is the one thing you already have plenty of, so it's just a question of packaging it for the Internet. The most popular content on the website is the the troubleshooting flowcharts, which are just a graphical representation of text content. One reason integrated graphics of all types work well on the Internet is because they demonstrate effort on the part of the writer. Blogs and e-mail have made it abundantly clear that for many of us, talk is cheap, but go to the trouble of drawing a graph and people will say, "Hey, this may be worth something." If all you write about are your opinions and heart rending stories about your family, you may build a blog following, but you're never going anywhere with search.

Yet the videos I did for the website were more popular than the website itself. Is this a problem? No, every video view brands the website, and several people a day click the link on the YouTube channel that brings them to the website. If I had written a book on automotive troubleshooting, I could go back and add a banner to the videos announcing it (instantly promoting the book to over 5,000 viewers a week), and continue to produce more videos based on the book content. If you're curious how much time it would take an author to put together all of the content on the website and YouTube, I'd estimate it was on the order of 200 hours, but keep in mind I partially rebuilt a car in the process. Another way of looking it would be that for each hour of time invested, I drew 1,000 viewers on YouTube and 500 readers for the website over the past year, and those numbers will likely rise over the next year or two without any further effort required. If I had really been writing a book about troubleshooing automotive problems, much of the work would have been used for the book as well.

So now we come to the beginning of the new test, which I suspect will prove that starting a new website for each new writing project is both unnecessary and counter-productive. My new writing project will be on the subject of self employment and self sufficiency, something I'm very passionate about as a self publisher. And not surprisingly, I think that most of the advice I see peddled by self employment gurus focuses either on the easy stuff or the unethical approach. Rather than setting up a new domain, I'm simply adding the subject to this website, with a couple new links on the homepage. Also in accordance with my standard practice, I'm posting draft material as I go along, which I'll edit and improve as the ultimate parameters of the project firm up.

The new project starts with the Sole Proprietor blog, which I published over an older blog about investing. I've experimenting with changing the subject of an existing blog in the past, and on the whole, it seems to hurt as much as it helps since any existing links misscategorize the content. But I don't take blogging seriously as a way to draw visitors in any case, I just wanted a simple way to interact with readers who don't like sending direct e-mail. The core of the new project for now will be a collection of pages about taxes for self employed people in the US, which I'm putting in a sub directory.

My ultimate goal is to create a broader guide to business self sufficiency. If I can draw enough interest for the tax nitty-gritty and basic business philosophy that I'll dribble out on the blog at one or two posts a month, I'll tie it all together with some new and existing material about basic home finance for a complete picture of sustainable self suffiency in a self employed business model. It seems to me that there's too great a focus on short-term income in most modern business models, whether it's self employed individuals or giant corporations, when sustainability of the business model should be the primary goal. This doesn't just apply to self publishing where I'll take four titles earning over $12,000 a year over one title earning $50,000 a year every time, but to every small business I've worked in or known about.

The Pit And The Internet Article Writer

The arc of knowledge dissemination on the Internet is traced by a surprisingly ponderous pendulum, one that has yet to reach the end of its second full swing. Period Web 2.0 has been defined by quality search and the spread of social networks, along with enough streaming audio and video to lighten the darkest fibre. But the geometrical increase in available information can be largely characterized as a thousand places to check the schedule of a train that was late twenty years ago and will be late again tomorrow. It's fashionable to see the Internet as a technological marvel that redoubles in merit with the regularity of Moore's Law, but the assembly line generation of search fodder challenges both the gatekeepers (Google et al.) and the humans seeking enlightenment.

I suspect that there will be a special place in The Pit for internet article writers who work from templates. They are responsible for all the sites populated with variations on: "Ten rules to write a great book," and intentional misspellings "10 rules to right a great book," ad infinitum. Sometimes I get stuck treating the casualties who mistake them for a resource and try to launch a publishing business based on bullets - bullet points that were written in a 21st century collaboration between a clueless writer and a sophisticated content generation program.

The three R's of bad article writers are Rehash, Rewrite and Reduce. All three types of plagiarism lead to the introduction of errors and exaggerations, but worst of all is reduction, which creates a false impression of authority without leaving the telltale signs of charlatanism. It used to be that an imposing edifice was key to instilling respect and even religious awe into the hearts of men. But modern man has found something even more imposing than monumental demonstrations of organized labor, capital expenditure and genius. Today's converts are won at the point of anything resembling a PowerPoint slide.

When the author of a new web resource wins past the front line of paid article writers, the hired guns of content spam, he's likely to run smack dab into a volunteer corps, hundreds of millions strong. Volunteer content generation sites, often disguised as social networking sites, run the quality gamut from WikiPedia to a million flavors of "Answer questions for bonus points." Taken as a whole, the content generation capabilities of social networks represent the industrial age of the Internet, an explosion of mass production at the expense of the individual artisan. Think of social networking content as the equivalent of asking a stranger on the streets how to find a particular address. Not so bad. Now think about asking a stranger on the streets for business advice. Now consider the results if you can only hear the advice of the stranger who yells the loudest.

Experience teaches long-time discussion list participants that the definitive comment is the one most likely to be exactly wrong. Psychologists or philosophers may know the reason that uncertainty manifests itself online as statements of canon law or unimpeachable truth, I can only observe that it does. It's no longer a surprise to find something I've written referenced as the authority to carry a point of argument that directly contradicts the text. No doubt somebody will reference this very blog post to argue that top ten lists produced by internet article writers are the very height of the information age. After all, I did say something about the upswing of a pendulum.

Perhaps contrary to popular belief (I've never seen a poll on the subject), I also believe that social networking tends to limit rather than encourage new authors. For all of the advantages of a public space, most authors crave a room of one's own, a place away from the madding crowd to work without interruption when the spirit moves or the bills need paying. Social networking is nothing if not a giant, ongoing interruption, and in many cases, the energy that could be channeled into writing a book instead gets scattered across hundreds of discussions (or thousands of tweets). Blogging has the same negative affect on authors, I could be the poster child:-)

Freemium or Fearmium and Scare Me? Um...

I invested six hours this week listening to Chris Anderson’s free unabridged reading of “Free – The Future of a Radical Price.” It’s the first complete audio book I’ve listened to in my life, and I greatly enjoyed it. For those of you who think all my blog posts are twice as long as they need to be, you might want to pay for the three hour abridged version, in accordance Anderson’s stated strategy for audio book sales. I thought about summarizing the book in this post, but I don’t want to compete with that abridged version, and besides, you couldn't afford it.

Free is nothing new to me, my publishing career has been based on free from the inception. I started publishing my work online back in 1995, and my first Freemium model was based on giving away the full version of a book as HTML and offering to sell the printed version for less than the cost of an inkjet cartridge. Free is what established my reputation as an expert and brought multiple offers from trade publishers for my work when I went that route in the 90's. I’ve been very active in encouraging unknown authors to publish online to establish themselves, you have nothing to lose but your anonymity.

Having spent six hours in listening to Free, I wanted to do something new I could point to as a direct return on the investment. So I launched a new venture to compliment my website and maybe add one or two readers to the couple hundred thousand a month who already show up for free content. The experimental first step was to publish a collection of blog posts from the last few months as a PDF, both on my website and on SrcibD. I know ScribdD must be a good place to publish PDFs because they were the first site to host a pirated copy of my laptop eBook, though I’m not inspired to link to them for that reason. I’m no apologist for piracy (and I believe Chris falls more to that side of the fence than the other) but I’m also not going to cut off my free distribution options to spite my reader count for something I’m already giving away.

While publishing a simple collection of free blog posts as an experiment is unlikely to accomplish much of anything, I plan to expand into topical collections. Since there are well over 400 blog posts in my archive, enough for a half dozen full length books, I think I should be able to put together a dozen or so collections on subjects like: eBook publishing, market research, Amazon and the retailers, content website design, publishing business issues, self publishing challenges, etc. Those collections I could publish as an eBook library or bookshelf both on my website and on sites like ScribD, and hopefully they will help build readership for my blog. The only problem with this as a business model is I don’t make any money through writing new blog posts, in fact, it’s a huge creativity drain with no measurable return outside of reputation.

Anderson would point out that I can use this enhanced reputation to earn larger lecture fees, sell T-shirts, or offer consulting services. There’s something to be said for this, in that I’ve already been offered two prestigious speaking gigs this year, I get regular requests to buy Foner Books hats, and despite my protestations that I don’t sell my time, there’s a new inquiry about my consulting rates around once a month. Unfortunately, I’m scared of public speaking, haberdashery, and selling my time, so the reputational currency is tough to cash in. And that brings me to the mirror story of Free and Freemium, which is Scare, and Scaremium, which I prefer to spell and pronounce: Scare Me? Um…

I don’t think I have to explain Scare, life is full of scare and fear, but I want to avoid fearmium because it sounds too much like freemium. Low grade scaremeum is popular with unethical freelancers and other service providers who participate in free forums, such as discussion lists for authors and publishers. I would define discussion list scaremeum as “Marketing services under the guise of giving free advice that it would be suicidal not to purchase said services.” When somebody posts a legal question about contracts or copyright, and a lawyer responds with post that explains all of the things that can go wrong without offering any advice on how to make them right, that’s scaremeum. When somebody asks what font to use for a book and a designer responds by describing horrendous scenarios of zero sales due to a poor book design without offering a font recommendation, that’s scaremeum. From accountants on business lists who explain why your shouldn’t file your own incorporation or taxes, to business consultants who have plenty to say about what you’re doing wrong and nothing to say about how to correct it, that’s scaremeum at work.

The next level of scaremeum is reminiscent of the “protection” sold by organized crime. Protection for your credit sold by the very credit agencies which are responsible for letting crooks access and destroy your credit in the first place is scaremeum at its best. They can even charge to lock your credit record so nobody can access it and charge again to if you want to unlock your record to apply for a mortgage. Creating both the problem and the solution is an ideal scaremeum business model. Phony virus protection software sold through pop-up ads that pretend your computer has just undergone a free scan (and failed) is another example, where the alleged cure carries a Trojan horse bearing the disease.

The biggest users of scaremeum are the many businesses that make up the medical industry, which prefers the friendlier “Health Care Industry” moniker. Friendly scaremeum ranges from online symptom checkers hosted by famous clinics that seem to end every diagnostic path with “See your physician immediately!” to “informational” websites run by pharmaceutical companies that would make Spock question his sanity if he spent enough time following their chicken-and-egg logic. Practically all of the medical information I’ve read online outside of WikiPedia and some support groups could be classed as scaremeum. Even medical spammers practice scaremeum, as demonstrated by the occasional e-mail that slips by my filter to insult my anatomy. All insurance, including health insurance, is sold through scaremeum tactics. Special lose-lose scenarios like the $1,000 plus tax penalty for not joining an HMO in Massachusetts turn health insurance into a pure protection racket.

Scaremeum is simply a more successful business model than freemeum. The largest employer in the United States (after the government) is the medical Industry, with the government funding over 50% of their gross. Under a freemium model, drug companies would give away the pills you have to take every four hours and sell the 12 hour or 24 hour time released versions. Or they would sell reformulations with fewer side effects or with fruity flavors. Instead, drug companies raise the prices of their older, supposedly less beneficial products, to get their users switched to the new drugs before generic competition becomes available on patent expiration. Where are the hospitals and surgeons offering the first joint replacement or eye surgery free in order to sell the second? Why not restore patients to health for free and make a profit selling them activewear and ski vacations? How about free heart bypass surgery as a loss leader to keep patients alive for the jackpot of long term care?

Free in its various incarnations is a neat business model, but it remains a niche model, even in the information economy where it thrives. Fear remains the killer application for both marketing and motivating. Ironically, and perhaps intentionally, one of the strongest arguments Anderson makes for Free is that piracy will force your electron friendly business into a free model whether you like it or not. That means that even the strongest advocate of freemium is marketing the concept with scaremeum. And why not, since it works?

Warning: Reading Morris Rosenthal’s publishing blog can lead to morning sickness in the morning, sleeping sickness at night, and certain sexual side effects (though we aren’t saying which). Readers of a placebo blog report experiencing bliss.

Think Like A Reader To Find Book Buyers Online

You can improve your chances of building a successful online business around a content based website if you learn to think like your customers. Established publishers have plenty of experience building brands around personalities and imagery, but they make a mistake by seeing the Internet as a mere extension of their pre-existing marketing techniques. Branding is traditionally synonymous with spending money on advertising, even "viral" advertising campaigns cost big dollars, especially if you amortize the failures against the small number of successes. Every new publisher's main goal for an Internet site should be attracting new readers through search. You can serve your fan base through author blogs, videos and other one-to-many broadcasts, but it's the new readers who build a new publishing business.

Quality writing and useful information will always be the main key to engaging, and perchance selling to, the readers who arrive at your website. But getting new readers to arrive at your website in the first place is the obstacle that most book publishers never overcome. You may think the best way to engage new readers is through turning a pretty phrase or drawing a clever allusion, but unless you are plagiarizing directly out of the brains of your potential readers, it's unlikely that they will be searching on your pretty phrase or clever allusion. Try to think of your website for a moment as an independent bookstore. Do you think you'll get more useful walk-in traffic if you have a sign that announces your store as a Turkish Bath with red satin curtains over the windows, or if your sign includes the word "books" and a number of books are displayed in the windows?

Useful walk-in traffic on the Internet is known as targeted traffic, visitors who are likely to be interested in what you are selling rather than the merely curious or red curtain voyeurs. The way to reach out to that targeted traffic with a content based website is to find out how they go about looking for whatever it is you have to sell them. If you publish books about architecture, you need to figure out how your target audience goes about searching for design ideas or the history of timber framing, whatever your titles may be. You don't need to draw up an FBI profile of your readers, it wouldn't help in any case. What's key is the words and phrases they use in searching for your subject matter online.

Since thinking like a reader turns the publishing concept on its head, I'll risk repeating the point to drive it home. Attracting readers through search isn't about turning a pretty phrase on your web page, it's about understanding what phrases your potential customers are likely to type into a search engine. Once you have that information, you can properly title your web pages, and write the content so that the search engine will understand that your content is a good match for their users searches. Otherwise, your quality content may never be visible in the search engine results, drowned beneath the weight of inferior pages and even keyword stuffers. It's like the old joke about the religious man who cries to heaven that he never wins the lottery, and finally the Lord speaks to him from a cloud and thunders, "Buy a ticket." You have to meet your readers halfway.

So how do honest publishers go about finding out what their readers are searching for online? Well, if you have a lot of money and time you can fool around with focus groups, paying to watch people do searches or hiring academics to put their slaves, I mean students, on the job. But for most of us, a knowledge of the field (which you better have as the publisher) and some research with online tools is the best approach. Today I'm going to stick with a couple of Google tools, Google Trends and the Adwords Keyword Tool. Google Trends is ideal for investigating the use of popular search phrases over time and is a topic I've already blogged about in the context of vampires and SciFi. The Adwords tool is less familiar to most publishers, but it delivers much more detail on less popular phrases that are likely used in searching for your books.

The main thing you need to be aware of when using the Adwords Keyword Tool is the pull down menu to the right for "match type". There are four different match types, the first three of which are different flavors of the same thing. The default match type is "broad" which means Google includes all searches that included the phrase words. The next is "phrase", which only includes searches that included the phrase words in exact order, and the last is "exact" which is limited to searches that consisted of the phrase and nothing but the phrase. The "broad" and "phrase" results will often be the same if the phrase is practically a compound word, like "Self Publishing", for which the result for the month of June is rounded to 165,000 in both cases. But the "exact' result for "Self Publishing" is about a fifth of that total at 33,000. A phrase containing words that are often used in the same sentence without being connected, like "Laptop Fan", yields 110,000 for "broad" results, 49,500 for "phrase" and just 9,900 for exact. Note that in all the cases above I'm using the U.S. numbers, the global numbers are a higher if the phrases are used outside of North America.

The most important information the keyword tool offers is not the results for the phrases you think will be popular, but the results for the related phrases that Google offers up under the heading "Additional Keywords To Consider." This is where Google politely points out to advertisers who are looking for customers that they may be targeting the wrong phrases. For example, additional keywords for self publishing include such suggestions as "book printer", "book marketing", "writing a book", phrases that contain neither of the original keywords or variations. What Google is telling us is that companies who advertise for the phrase get results from people who search on these other phrases as well.

In terms of how well the number of searches reported by Google matches up with the results I see on my website, I'll have to sit down some day and put a serious effort into working it out. The quick check suggests that for exact phases that bring readers to my website, I get somewhere between 25% and 100% of the volume reported by Google when my site comes up #1 in the Google results. The lower percentages tend to correspond with the shorter search phrases, which makes sense since it's more likely that the title and snippet displayed by Google is enough for the searcher to know that it's not what they were thinking when they typed their two word query. I'm also using my Google rankings from August to compare with search results from June (the latest available) so some of the pages that are now ranked #1 may well have been lower a month or so ago.

All this business with phrases is something you need to know about, but not anything you should obsess over to the point of spending money on SEO. In the end, a good content website will get the majority of its traffic from Long Tail phrases, as described in the video below.

Women (And Men) Of Trade Publishing on LinkedIn

I didn't set out to blog about gender in publishing. My goal last week was to spend a of couple hours researching the career mobility of acquisitions editors on LinkedIn to go along with a video I had already shot. The couple hours turned into a full week of reading through hundreds and hundreds of public profiles of editors, associate publishers and assorted vice presidents, and puzzling over the career paths of trade publishing professionals. Initially I tried relying on Google search techniques, adding and excluding keywords in hopes of getting some statistically significant results without having to read every profile, but that turned out to be a weak methodology. By the end of the week, I had an idea what it would feel like to be CEO of a large trade publisher and to read a stack of resumes for a senior position instead of employing a headhunter.

The initial goal of this study was to see if the publicly available information on LinkedIn would be useful in determining how long acquisitions editors remain in their jobs. This is critical information for unagented authors who are looking for a publisher and are going about it the right way - investing their efforts (and postage) in contacting editors with a stated interest in acquiring works in the author's field of endeavor. Authors do their homework by buying the latest edition of Writers Market or Jeff Herman's Guide, and searching for the names and contact preferences of appropriate editors. As often as these books are updated (go by the publication date, not the year on the cover), a large number of editors will have moved on by the time the author purchases the book. So I started by looking at the profiles of current acquisitions editors and senior acquisitions editors to see how long they had been at their jobs. For acquisitions editors, the median time (half shorter, half longer) in the current position was 1 year, 9 months, and for senior acquisitions editors, the time was 2 years and 4 months. The moral of the story is that authors should check LinkedIn or use Google search to try to determine if the editor for whom they are crafting a proposal is still in the same job before investing too much time.

Now a quick look at the data I gathered earlier in the week on the relative career mobility of women and men, before I realized that the methodology was too weak. The reason I'm giving these results is because the obvious gender inversion with seniority is what led me to invest several more days trying to increase the accuracy. Note that these percentages came from paging through Google results generated with the " parameter" and the job title. Gender was determined by first name, which means I had to exclude many unisex and foreign names. Also, just because the job title is included on the public profile doesn't mean the individual held that job, i.e., an intern with work experience as "assistant to the associate publisher" would have been counted with associate publisher, etc. Equally important, many of the same job titles are used in magazine publishing, radio, television, and advertising as well as trade book publishing.

Associate Editor
Men 35% Women 65%

Acquisitions Editor
Men 41% Women 59%

Senior Acquisitions Editor
Men 49% Women 51%

Senior Editor
Men 57% Women 43%

Associate Publisher
Men 61% Women 39%

Group Publisher
Men 79% Women 21%

Now before anybody gets excited about these early returns, remember that the methodology was flawed. But it motivated me to take a closer look at a specific group, big trade employees (think Random House, HarperCollins, McGraw-Hill, Pearson, etc) with the job title "publisher" or "associate publisher" in their public profile. Rather than just counting the boys and girls in this group, I also counted the number of years from college graduation (100% claimed to have graduated from college) to when they first claimed a job with "publisher" in the title. The final group was limited to 61 senior trade publishing employees, as most listings had to be disqualified for reasons like incomplete job histories or omitting college graduation dates. Worse still, if somebody spent 15 years or a whole career at a company where they eventually became publisher without breaking down their progression of jobs, I couldn't use it. In retrospect, I wish I had retained a record of whether those omitting college graduation dates were more likely to be men or women.

After spending a few days reading all of the relevant public profiles on LinkedIn, I came to the conclusion that while men edged out women in senior positions, women reached senior positions a little faster than men. Given the small final sample size, both conclusions are probably in the error band, with men accounting for 56% of the publisher jobs (44% for women) but women first claiming the publisher job title after 15.96 years on average vs 17.32 years for men. As a side note, the 12 employees with graduate degrees, all Masters, reached the publisher job position in 13.33 years on average, so graduate degrees were a minor help to men, but not much use to women.

Some people manage to encapsulate their career history on LinkedIn, like one ex-acquisitions editor now working as a grade school teacher who I'll paraphrase: "In the morning I would read a children’s book and in the afternoon I'd negotiate with authors – it was a wonderful job until I got downsized." On the other hand, a former VP and Associate Publisher at a top trade who became a “reading specialist” at a grammar school after 30 plus years in the business is, I assume, happily retired. In some cases, you can see where the march of time and the Internet has changed the publishing landscape, as in the profile of an editor who moved on after 18 years in one job as an acquisitions editor for an encyclopedia.

On many profiles of these educated and highly experienced managers, self employment is clearly a euphemism for laid-off rather than a career choice. I wish I could have sat next to them on a plane ten years ago (thanks to a free upgrade to business class) and convinced them to start their own websites and publishing companies rather than working for corporate America. Of course if I had a time machine and could do it, would I really want to create all that competition? I was also struck by all the people who seemed to have a firm plan in mind for their direction in life before they even left college. How come all of the successful people I know in self publishing came to it after years of this and that?

Many editors enter the trade at higher levels, including one who made the jump from store clerk to senor acquisitions editor at a major trade – being underemployed to start helps. One of the longest job descriptions on LinkedIn was by an acquisitions editor (he included all the authors he was currently working with) who has been in the job just 2 months. In one case, a publisher I used in the study didn’t give her graduation date but was a member of the LinkedIn group for her class of 1984 college reunion – gottcha! Another publisher’s first job in publishing was “desktop publisher”, a related but no equivalent position.

In general, I didn't see much humility on display in profiles, some of the claims were pretty fantastical and would no doubt be hotly debated by fellow employees if LinkedIn had a reader comments option. I noticed that people stay at TV jobs. I didn’t make any calls to Random House or HarperCollins to confirm the claims on LinkedIn profiles, I took everybody at their word. Perhaps the most interesting thing about this study is that you can do it yourself. The information is public, all it takes is time. Analyzing the public profiles for different publishing jobs on LinkedIn might make a decent August assignment for your summer intern who is running out of things to do, now that everybody is off to the shore or mountains for a few weeks vacation, and it's too hot for the underlings left behind to drink coffee.

And finally, here's the video that cost me a week of reading virtual resumes: