Welcome to the new year.

No resoutions for me, just getting ready for a new baby at the end of Feb.

Now, on to the second of the four parts.  Software that no one uses but me.

Now that might not sound like much.  But in fact, I think this is the largest category of software in the world.  Want to know why?

Because included in this category is every little spreadsheet, database, and wordprocessing document or template with embedded scripting that anyone ever wrote to make their job easier.

And if I told you that number, it would probably make you disbelieve me.  And in fact there are features still built into Excel simply because Wall Street firms have been using them for years, and threaten to move to something else if they have to go and rewrite it (though maybe that might have helped prevent some of the meltdown we had).

But I can give you are real example.  I did an analysis project for a company, it turns over something in excess of $2,000,000,000 (yes, that is 9 zeros there) each year (not bad for a privately owned firm), and 90% of their business is driven by spreadsheets.

We found 40,000 spreadsheets on their network.  Which means each spreadsheet had an average value of $50,000 per YEAR.  Of course, half of those were things like the default Excel templates, and the new document option for the right click menu in explorer, so that really means that they had 20,000 spreadsheets with a value of $100,000 per year.

In truth of course, 90% of those spreadsheets are old, and probably worth something between nothing, and maybe $10,000, but a few were worth over $10,000,000.  The project was to identify them, and plan a way to put some controls and structure around them.

 But anyway, this is what "exemplifies" software that someone builds for themselves.

• Not flexible
• Only one way to perform any action
• Does only one or two things, but does them exactly
• Frequently requires "tweaks" to the code
• Doesn't crash if you make one of your common mistakes, but heaven help you if you do something really weird
• Only one person really understands it

Often these pieces of software evolve over time to become the central core of a business process.  And trouble sets in when the original creator leaves, because no one else understands it, and everyone is afraid to change it.

 Next up, software written for a single business or group (bespoke software)

Wow.  Nearly 2 years since I last posted.  I have been really slack.  Ah well, time to get back into it I guess.

 So, I was busy learning to tap dance.  Long story, something my wife wanted to learn.  Oh, and as an aside, learning new stuff is good for your brain.  In this case, figuring out just how horrible I was at learning a new skill got me to thinking about the four different types of software we commonly write.  Well, more like styles really.

So this is going to be the first of the four.  Software for learning purposes.

That is, the software we write when we are trying to figure stuff out.  Like at university, we had this assignment to write a program that could solve the "Towers of Hanoi" as an example of recursion.

And an hour before the assignment was due, I had this 500 line program full of messy special cases and weirdness and it mostly worked, and suddenly it clicked.  Thirty minutes after that, I had a 33 line program that was perfect.

Or when I was figuring out how to read and write lists from SharePoint using the web services.

You write this messy code, and it has all these hard coded pieces in it, the URL, the name of the list, the values you want to add or delete.

And this is the software we write when we are learning stuff.  It does these very narrow specific things, it doesn't cope with anything that deviates from the expected.  Every time you want to change what it does, you have to change the code.

This is all good for figuring stuff out, but it is completely useless for giving to someone else.

Next up, software for ourselves.  The software we write to use for ourself and no one else.

Now comes the research portion of your software project.

If you have not yet registered with the site (or sites), you can do that now.

Then, go through the "top sellers", or "hot buys" list for each site.

 Visit the product page for each item and note down (Excel or are text file are both fine) the following:

• Name of product
• General category (e.g. Helpdesk, advertising, security)
• Any questions and features that are often mentioned.

Once you have done at least 20, go back over them and pick a general category you want to create a product for.

Go back through those products and note down all you can find out about the features they have, and those that are asked for.

You will use this to plan out your list of requirements before you create your product.

Note that while this shortcut gets you lined up on a product that people want, you are unlikely to end up on the top sellers list yourself unless you do a really good job.  Many of the products you find there will have been around for several years.  This is a quick way to get a product done and selling so you can use the revenue for other things.

Much like any other for of selling, to sell software, you need to start with a target market.  A group of people who are likely to be interested in what you have to offer them.  Fortunately, the internet makes this quite easy to discover, though it can take some time.

For the purposes of being able to sell without needing a web site or anything else that might cost, you need to go looking for a specific type of web site.

It needs the following features:

1. To be popular.  Fortunately, Google already figures this out for us.
2. To be focussed on a target group or market
3. To host and process payment for software products for you
4. To not charge any set up fees.
5. To have a "most popular" or "top sellers" list

Any easy way to get a ready made target market is to produce "add-ons", these are often referred to as "modules", "extensions", or "add-ins".  These are small pieces of software designed to extend or alter the behaviour of existing applications.  Your target market is then defined as the owners of that application, or a sub-group of them.

Some examples of applications you could target are:

• Photoshop
• DotNetNuke
• PHPNuke / PostNuke
• Microsoft Word / Excel / Access / etc
• FireFox
• Joomla
• WordPress
• Internet Explorer

I recommend you do some searches for programming languages you know, and  see what you can come up with.  I'm going to go with "DotNetNuke" for this as I'm familiar with it from previous work I've done.

Searching Google for dotnetnuke purchase commission brings up a number of sites.

Two that are potentially useful are:

• SnowCovered.com
• DotNetNuke Marketplace

You are likely to arrive somewhere deep inside the site, so make sure you check out the home page.  Both these sites are suitable for this because they meet all the criteria that I specified earlier.

If you want to follow along with this process, go and do your research now, I'll be posting the next step within a few days.

I must apologise for the long time between posts.

I'll be posting the first part on creating and selling a software product for $0 shortly, but in the mean time, I just wanted to let you know that I've released a guide to customer service that you can get from this site Customer Service Guide. It's just $4 until the 1st of Jan, and $10 after that.

It should be useful to anyone running, or looking to run a business, especially one which is mainly online.

I hope everyone has had a great holiday, and I'm looking forward to a fun and profitable New Year with you all.

By the way, if anyone has suggestions or questions for other guides they would like, drop a comment here and let me know about it.

Yes.  You can do it.  And since I like the idea of showing how, and someone else though it would be interesting, I'll go through it.

This will take me a few posts to go through, from locating an apropriate website, identifying a product, producing the product, and putting it up for sale.

Points to note:

• This is not hugely cost effective, you'll be giving away ~25-30% of your revenue to someone else in exchange for not doing payment processing
• I don't recommend this for long term use, but it works well for getting some cashflow up front.

You WILL need:

• A PayPal account (if you want to get the money quickly)
• The ability to program, or someone who will do the work in exchange for a profit share.

Things you need to do:

• Locate a website that acts as a "marketplace" and processes the payments for you
• Research the products that sell well
• Create a product specification
• Create the product (or get it created)
• Package it
• Set it up on the website
• Write sales copy for it

I typed that into Google, and I got nothing back.

Which is odd.  Google is usually pretty good at defining things.

Anyway, here is a good working definition of a microISV (borrowed from Wikipedia):

A Micro-ISV (abbr. mISV or μISV), a term coined by Eric Sink, is an independent software vendor with just one software developer. In such an environment the company owner develops software, manages sales and does public relations.

The definition is commonly accepted to be slightly wider than that.  Often encompassing up to 10 people, but most often one or two. An "Independent Software Vendor" by the way, is a company that sells software, but doesn't quite reach the scale of Microsoft or Oracle.

What this means is that one person can basically do everything required to run that business.  And that comes down to about 10% research, 5% design, 10% development, 10% support, and 65% marketing and promotion.

In fact, you can even farm out the development and support parts (though I recommend keeping that) if you are not a programmer.  The real keys are:

1. Finding out what people want (no technical skills required)
2. Designing something give it to them (some logical/planning skills required)
3. Getting it somehow
4. Marketing it (marketing skills)

Notice that programming skills don't actually appear there.  The only place those are needed are to create software (and that's just one of the ways of getting it), and it would help for designing it, but is not critical.

So, now you have the technical term for how I spend my evenings.

A lot of people think that businesses run on a bunch of complex and expensive applications.  SAP, JD Edwards, MS CRM and the like.

The truth is that the vast majority of businesses are SMALL businesses.  And a small business usually runs on a combination of an accounting package, pen and paper, MS Word, MS Excel and maybe MS Access.

However, a time comes when a growing business finds that the systems they have been using are now causing them more trouble than they solve.  A smart or lucky business finds that out before the whole thing breaks down.

Then they need to progress to a better system. 

• Pen and paper moves to Excel or Access. 
• Excel moves to Access. 
• Access moves to SQL Server, and a web or desktop front end,
• or maybe an off the shelf application.

Applications move from being single user to being able to function with 5 or 10 people using them at once.

A very large proportion of what I do as a bespoke developer involves taking an existing Excel or Access based system, and moving it to the next step.

I'm considering taking an Access application I wrote over 10 years ago, and turning it into a .NET Windows application.

As part of the exercise, I can record the process on video and create a training course out of it.

If anyone would be interested in learning how to convert Access applications to .NET "nicely", drop a comment in and I'll start the process.

You need to give some consideration to how you license the software you sell.

There are some important points to consider.

• All license issuing has a cost overhead.  You need a commerical licensing tool, or to write your own.  Plus support costs for issuing new or updated licenses.
• All licenses can be hacked, eventually, even if it requires someone to create binary patches of your code.
• A license will stop most people from trying to cheat anyway
• Licensing may reduce your sales

Because of all that, I recommend structuring your licensing as follows:

• Under $50 - Don't bother, the extra support costs will eat large chunks of your time.  But make sure you build in some ways for people to find out about other (more expensive) products you sell.
• Under $200 - Make it simple and Generous.  E.G.  A person can put it where ever they like.  All websites on a server can use the component.
• Above - Look at "per site", or "per user" licensing.  Give generous discounts as the number of licenses climb.
• Where possible, provide a service, not software.  This means you host the software on your own server, people connect to it and use it.  Monthly membership or lifetime membership fees apply.  But remember that lifetime fees imply a contractual obligation to provide the service, and that you will have ongoing hardware and bandwidth costs to pay.

In response to a question I had about licensing, and e-Commerce providers, take a read through:  http://successfulsoftware.net/2009/10/12/a-survey-of-ecommerce-providers-for-software-vendors/

There are many ways of coming up with ideas for your own software product, even if you don't already have an idea in the back of your mind.

Let's talk about one of the options.

If you'd like to jump into things quickly and you have capital available to spend to acquire software outright, you might consider purchasing existing source code.

You can purchase software that is already being sold publicly or you can find a programmer who has put code together but hasn't taken the time or energy to make it publicly available.

Buying Existing Software

Programmers are a notoriously introverted group.  In an episode of ScreenSavers on TechTV, a software programmer had put together an amazing piece of software that allowed artists to "paint" movies using film that had been shot with digital cameras.  It allowed for a surreal movie to be built in a matter of months using techniques that would have taken traditional artists years to complete.  This was some awesome software!

The host asked the developer the exact question I had in my mind: "Do you have any plans to market this software?" The response: "Well, I'm not sure yet." What was he thinking? This stuff was great! I wanted to download and try it out myself.

Even though it was an amazing piece of software, the developer hadn't even begun looking at ways to capitalize on it beyond what it was being used for the film they were currently producing.  You're going to find many similar stories for thousands of products that exist - but aren't on the market.

Yet.

THERE ARE THOUSANDS OF SOFTWARE PRODUCTS AVAILABLE, READY FOR YOU TO PACKAGE AND MARKET RIGHT NOW!

If you think I'm kidding.  Let me tell you some of what I have sitting on my hard drive waiting to be completed at the moment, and I sell software for a living.

• A ClickBank search engine, with Google style search functionality.  I'm just trying to figure a good licensing model, but I've left it for over a year.
• Improved version of my ConversionBooster split testing software.  Just need to install it and validate the results a few places.
• New Affiliate Program software.  Needs some more functionallity, but I'm currently testing the tracking / payments portion of it.
• Product Sales website.  Integrates membership, security, downloads, ordering via PayPal and 2Checkout.  Can include the split testing and affiliate software easily.  Need to package and promote it.
• License Key Generator.  Uses 256 bit Public/Private key encryption to secure license information.  You might be able to crack it to read the content, but you can't modify the license and have it work.