Kathy Sierra wrote a great post back in 2005 called Featuritis vs the Happy User Peak. It’s well worth a read if you’re involved in any sort of product development - software, dog houses, electronic voting machines. The general gist of the post is that adding features does NOT add value - the secret to creating a great product is adding just the right features to give the most value, making them perfect, and then stopping.

Software companies don’t like this. The last company I worked for in the UK had what might have been a great product, but ruined it by continuously piling on feature after feature. They did this so their sales team could say to a customer: “Yes, our product does that too!” The end result was two years of development, a product that kept crashing through lack of testing, a dissatisfied customer base, and an even more dissatisfied work force. And of course, low sales.

To borrow a line from Kathy’s post: “Don’t give them new features just because your competitors have them!”

I’ve tried to follow this advice with PageFour, choosing to add features only when they add value to the product for most users, and impact little on usability. The Search and Merge Pages in the recent release are examples of this. But I haven’t always been so successful in the features I’ve chosen to add. My feeling is that version 1.50 strayed a little off course, with the inclusion of Smart-Edit.

This is a great feature, and I use it all the time. It was designed to identify over-used phrases - something that has always plagued my own writing - and I’ll be running it on this post as soon as I finish. But it’s complicated. And it doesn’t really fit in with the rest of PageFour.

The problem, is that recently I’ve been thinking of numerous additions that could be made to Smart-Edit, making it much more powerful. Each one would benefit a certain proportion of users, but these users would, of necessity, be people who have no problem running complex bits of functionality, adjusting configuration settings, and playing with the features until they obtain the best results. And this is NOT most PageFour users.

PageFour was designed with simplicity in mind, and Smart-Edit, and the extra features I’ve been thinking about, are not simple. But don’t panic! The current incarnation of Smart-Edit will be staying where it is.

So I’ve been considering a spin off product - a product built around Smart-Edit, and incorporating all the extra functionality that would only weaken PageFour. To offer a general outline: the product would be designed for use on a first draft of your 80-100,000 word manuscript, just as you begin editing and revising. As with Smart-Edit, it would not tell you what to do, only highlight areas that you might want to look at in more detail.

I’ve drawn up a list of features it might contain. It’s a very rough and ready list I put together yesterday evening, but should give a taste of what I believe IS achievable through software.

• List of over-used phrases, as with the current incarnation of Smart-Edit.
• Highlight excessive use of certain phrases at the beginning of sentences.
• Flag potentially awkward tags used in dialog. For example: ‘she snarled’, ‘he bellowed’.
• Over us of ‘…’ of ‘-’ in dialog. A lot of amateur writing tends to suffer from an abundance of dialog interruptions through ellipses and dashes.
• Frequency of adverbs in sentences. How many or what proportion of sentences include adverbs? And are multiple or strings of adverbs used in the same sentence?
• Highlight weak qualifiers - such as very, a bit, fairly, quite, slightly.
• Excessive use of The, A and And to begin sentences, as well as There was or There were.
• Highlight redundant words. For example: a cold chill, the end result.
• Use of weak phrases: The fact that - of the (students of the college instead of college students), She began to - He started - appeared to - seemed to, etc.
• Use of ‘then’ in place of ‘and’ or a new sentence - she did this, then she did the other…
• Flag clichés - a trusted servant, a mighty warrior…
• Flag sentences without verbs - excluding dialog, of course.
• Excessive use of punctuation - exclamation marks, for example.

Before anyone leaves an angry comment along the lines of “But Faulkner did that ALL the time!”, I should point out that features like those above, and like Smart-Edit in the current version of PageFour, only point out POTENTIAL problems. It’s always down to the writer to decide if they actually are problems, and make corrections where needed.

The intention would be to make the product fully customisable, with the user capable of editing lists of ‘weak’ words or phrases and saying whether something constitutes a serious problem or not. Creating a separate product independent of PageFour means that the potential user base would be far larger, while PageFour itself would not be contaminated with new and complicated features.

I’m very interested in hearing feedback on this - do you you think it’s a good idea or not? Does it have potential? Would YOU use it? If not, why not? Have I left anything obvious out?

About 6 weeks ago, I noticed a gradual increase in Google traffic to this site. Without revealing precise figures, I’m getting about 4-6 times more hits from Google today than I was only a few months ago.

It’s always difficult to pin down the precise reason for such an increase, as many different factors come in to play. Back in April, I redesigned the site from the ground up. The old site - in all its green ugliness - was the first website I ever designed, and it suffered from many of the more common website design mistakes: poor wording in page titles, no use of header tags (h1, h2, h3 etc.), static content…

It was not designed with search engines in mind, and the traffic reflected this. The redesign addressed these issues, as did pulling the PageFour blog into the site. To this day, I have no idea what possessed me when I decided that hosting the blog under a separate domain was a good idea.

Most of the ‘new’ Google traffic is very relevant to PageFour, with search values such as “creative writing software” and “software for writing novels” appearing regularly. What has surprised me though, is the high number of people searching for “page four software” rather than “PageFour.” This breaking of the name in two is a VERY recent phenomenon, and I don’t know where or how it began.

Someone, somewhere, must be writing about PageFour in this way, or speaking very slowly when they talk about it, pausing after the ‘page’ to catch their breath before moving on to the ‘four.’ I’m convinced of this, because these searches only began two months ago. To date, I have had no luck in tracking them down.

The title of this post is a signpost to Google - just in case any of these potential PageFour users have difficulty finding me.

By far the most common ‘useless’ search value is a variation of “strikeout shortcut”, with MS Word often appearing alongside. I can’t help but feel there’s a message here for the Microsoft Word development team.

On a lighter note, I’m always amused to come across a search value such as “pagefour license crack.” Those cracks may or may not be out there, but if they are, does anyone really believe they’d be hosted on the PageFour site?

It’s always difficult to work out what you should charge for software. In one sense, as a digital download, the software has no ‘real‘ value. Despite what Microsoft or the record industry would have us believe, digital goods, delivered down a high speed connection, are not the same as a DVD you’d buy in a record shop or a loaf of bread freshly cooked at the local bakery.

The costs are mostly in the development, and once the product is completed, tested, and reasonably stable, that’s pretty much it. Sure, you have website costs and bandwidth costs - all ridiculously cheap these days, as well as support costs - not that high when you have a stable, easy to understand product.

Which brings me to PageFour. Over the past year I’ve played around with the pricing many times - all in an effort to determine which figure produces the greatest return. Apologies for my capitalist tendencies. The conclusion I’ve come to is that for small software products such as PageFour, there is a $30 price barrier.

Moving beyond the barrier, even by as little as $5 seems to have a seriously inhibiting effect on buyers. I’m not sure why this is, as $30 is not a vast amount of money, but the barrier does exist. People seem to hesitate, as if what might have been an impulse buy at $29.95 suddenly becomes cause for serious consideration at $34.95.

With that in mind, the price of PageFour will be returning to a more modest $29.95. The crowds have spoken.

Sales of PageFour have been encouraging over the past few months. It seems that word is slowly spreading as more and more people try it out. In the early days, soon after version 1 was released, people tended to stumble across the product through Google search strings relating to writing, or via software download sites. The numbers were never great, and sales slow to come.

Over the past six months, a large proportion of visitors who reached the site, did so through typing ‘PageFour‘ into Google - not blind searches, but specifically looking for the product. The release of the FREE EDITION and the removal of the 30 day trial limit back in September may also have contributed. As for the download sites, they still play their part, but only in a small way - providing incoming links which help with Google Page Ranking. In terms of numbers of downloads, they account for very few.

So where in the world are the buyers coming from?

As expected when it comes to buying downloadable software, the largest proportion is from the US. The figures that have surprised me over the past year are the comparatively high number of sales to Australia, as well as some of the ‘other‘ countries that pop up.

PageFour is not friendly to languages other than English, either in terms of menus and documentation, or dictionaries and spell-checkers - which is why for a product aimed at creative writers, sales to non-English speaking countries always come as a bit of a surprise.

These countries include Sweden and Norway, Mexico and Brazil, Germany, Spain and India. Over the coming weeks, I’ll be looking into providing friendlier options for those writing in other languages.

Over the past year, there have been a number of enquiries about the possibility of a Mac version of PageFour. My response has always been that the probability of releasing a version for the Mac is very low. I’m a Windows developer, and have been for many years. Despite having a lot of respect for the Mac software community, I have never been tempted to join. My training and working life have been focused primarily on Big Business software - an area that tends to impinge little on Mac users.

Every time someone queries me on the Mac, I point them to Scrivener, a neat piece of software, written - like PageFour - for creative writers. For many months, Scrivener was available as a free beta, as the product was still undergoing development, but version 1.01 of the completed version has just been released.

Definitely worth checking out if you’re a Mac user, as it does share some features with PageFour (the use of Snapshots being the most obvious). You can download it here, or check out the authors blog here.

Ten days ago, PageFour was picked up by a series of British computer magazines, nost notably Computer Active, What PC, and Computing. When I say picked up - it featured as a recommendation on their software download sites and appeared in the newsletter they send out to subscribers. This has resulted in a large increase in downloads, sales, and enquiries. My first thought when the figures shot up was that yet another dodgy download site in Hong Kong or Russia had found me.

PageFour ships as a thirty day trial version, so it’s too early to say how effective this push will turn out to be, but even now, ten days on, the download figures remain high and sales keep trickling in.

I played no part in this whatsoever.

Just over a week ago I attempted a marketing push of my own, where I offered free copies to anyone prepared to blog about or mention the product on their web-site. The plan was to generate some sort of buzz around PageFour, as up to now it had had limited exposure, popping up on blogs and discussion boards only rarely.

They succeeded. I didn’t.

The strategy had merit, as other companies had tried similar drives before. A few months ago, a company called Axosoft released their flagship product for five dollars, with all the money going to the American Red Cross. They shipped thousands of copies over a three day period and appeared on the front pages of del.ic.ious and Reddit.

So the question is, why did I fail where Computer Active succeeded? They had little to gain from promoting a small software product owned by another company, whereas I built PageFour and invested much time and energy in making it as perfect as possible.

What it boils down to is that they have a voice that is heard, and I don’t. Sending a targeted newsletter with a list of new software recommendations to people who’ve specifically asked you to do so, is very different from standing on a soapbox in the middle of an empty square and offering your product to a busy world.

If no one is listening, then it doesn’t matter what you say or how loud you say it.

Programmers don’t like users. In an ideal world, whatever product they are designing would be beautifully written with the latest trendy technology, passed on to a testing department (preferably in a different city), and never sold to a single user. It would work like a charm when configured correctly through hacking various text and script files in Notepad, produce in depth exception reports for every problem encountered, and have a beautifully crafted and fully modular framework of code that can be reused in every future conceivable product - whether a flight simulator for NASA or a stock inventory for Jerry’s Fish and Chip Shop.

The problem with users is that they tend to get this crazy idea that what they see on the screen actually is the product. They seem to think that the hastily thrown together heap of grey dialogs and tacky icons sourced from a free download site is what they’ve paid all that money for.

They’d be right.

Software is designed to be used, which means it’s designed for users to use. Note the emphasis on the word use here. When someone pays you money for a piece of software, they don’t care how the underlying code base has been constructed, and they don’t care about the structure of the database. All that matters is: Can I use it to do the job, or will it cause me grief?

Every programmer pays lip service these days to the importance of user interfaces, but in most cases they don’t really believe it. Like an alcoholic who tells you about his drink problem, they say the words because they are the words people expect to hear, but deep down they still believe that they control the drink, it does not control them, and the GUI really isn’t as important as the engine running underneath.

For too many programmers the GUI is the bit bolted on at the last minute to satisfy users and the guys in sales. It is often designed by someone who thinks typing configuration commands into a text file is far more useful than an interactive GUI with trees and checkboxes; that a visual interface can never allow the flexibility and customisation of a simple configuration file. This is all true - if the sole user of the product is the guy who wrote it!

If you don’t think about the user from day one, you will never design a truly great product; you will never design a product people want to use. Even the giants make this mistake. I stumbled across an article the other day on the failings of Lotus Notes, one of IBMs big groupware products everyone was talking about ten years ago. To quote one of the many disgruntled users:

“Notes’s backend functionality has no bearing on us 100m or so end-users. As far as we are concerned the GUI is the system. And boyo… is the GUI client a heap of ill-conceived, non-intuitive rubbish.”

Lotus Notes now has the dubious honour of its very own hate site where its many failings and embarrassments are displayed and dissected for all the world to see.

There’s a lot more competition out there today than there was ten or twenty years ago. If you build a technological marvel and ship it with a shabby user interface, and the competition builds a functional product that looks fantastic, who do you think will win the sale?

My TV remote has twenty nine buttons I’ve never used; the VCR has thirty; the DVD player twenty six. And before you ask - yes I really did count them all.

There’s a name for this kind of excess: it’s called featuritis, and it pops up in all sorts of design failures. I stumbled across an interesting article on The Featuritis Curve the other day that goes a long way to explaining why we have to endure this.

Good designs are obvious. When aesthetic and functional beauty combine, the result is not just pleasing to the eye, but a pleasure to use. This is why the iPod is so popular. Chances are it’s not the best value for money - there are probably other music players that produce a better sound, have a longer battery life, or have a whole raft of additional features - but the iPod looks like a work of art and is simple and straight forward to use

When have you ever heard someone boast that they couldn’t program their iPod?

So who should we be designing for anyway? The 98% who use a small number of features all the time, or the 2% who use advanced features once a week? Wouldn’t it be a whole lot simpler if my TV remote had a single advanced button that opened a menu for all those quirky, unused features?

Make no mistake here - you’ve never used those mysterious buttons either, and you probably don’t know anyone who has.

Software is often designed in the same way, for the same 2% of users. Feature after feature is added, each just as prominent as the one before, with little thought given to the complexity this adds to the product as a whole.

At the nine to five over the past few weeks, we’ve been putting the finishing touches to a new product aimed at the business market. It’s the sum of eighteen months work for four programmers - plenty of time and resources to produce a first class piece of work. But it’s not first class. Dialog after dialog contains features, options, and checkboxes that no one really understands. The poor unfortunate writing the help file has to suffer every time he asks for an explanation of feature X. In some cases, even the person who designed the feature can’t explain it without re-reading his own code, and even then the explanation can be patchy.

This isn’t funny, it’s critical. Complexity like this is nothing to be proud of. But why does it happen?

You can chant ‘keep it simple‘ until you’re blue in the face, and everyone around you will nod their head in agreement. The problem is, we all have a different perception of what simple means. Programmers are geeks; they belong to the 2% who use all those weird buttons on the TV remote, and for many of them these features are absolutely essential. The sales and marketing types want to be able to say that their product is better than the competition because it can also be used to fry an egg, or do your kid’s homework. But that does not make it better.

Features are not important, it’s the product as a whole that matters. Does it do what it was designed to do, and does it do it fantastically well? Does it look good, run smoothly, and behave intuitively? If it doesn’t, then all the features in the world will not save you.

More features does not equal more sales. More features does not make for happier customers. More features will not make your product better than the competition.

PageFour has reached the stage where featuritis first raises its head. Over the past couple of weeks I’ve been putting together a future development plan, and my focus throughout has been on identifying those mysterious buttons that so few of us ever use. Only time will tell if I’ve succeeded.

When I first created the web-site for PageFour, I’d vaguely heard of the term Search Engine Optimization. It was one of those things I assumed I’d get around to eventually, but it would just have to wait until the more important stuff like rolling out the first, and then the second release of PageFour had been dealt with.

I was wrong.

Had I read those few chapters months ago, I would have realised the importance of the URL, and chosen a domain name with more relevance to the product. www.imbt.co.uk has no meaning to anyone but me. I know it’s short for It must be Tuesday; I know IMBT Software is the company name; but does anyone else? And would it make a difference if they did? It’s the product that matters, not the company.

PageFour is software for writers, and the new URL is www.softwareforwriting.com.

It’s been live now for about a week and has already been picked up by MSN’s very industrious robots. The Google team however, seem to be sleeping on the job, and let’s not even talk about Yahoo. Still, it was something that needed to be done, and it really couldn’t have been put off any longer. All the links that I control on other websites have been changed, and the new download sites PageFour has been submitted to each carry the new URL.

Things are finally moving in the right direction.

It’s make or break time for PageFour. Version 1.2 went live at half past midnight.

One of the prevailing opinions in software development circles is to release early and release often. By doing this, you develop your product in line with your customers. Features being added to the next release should always be features existing users have asked for, or potential users have queried before deciding not to buy.

Releasing against a six monthly or even worse, a yearly development plan is catastrophic. The software market, and the IT industry in general changes too fast for such long term plans. And one year is a very, very, long time.

Small software companies have one huge advantage over the giants, and that is their ability to do things quickly. For Microsoft to release a new version of MS Word is probably a two year undertaking - for a company of one or two programmers, two or three months is not unrealistic.

I’ve been following this maxim with PageFour since day one. Version 1 was released on October 8th this year. The world did not shake, but the feedback and suggestions were inspiring. Two months later and I’m feeling very confident about the new release.

The only question now is: will the world shake? And if not, can I make it shake?