* The Business of Software

A former community discussing the business of software, from the smallest shareware operation to Microsoft. A part of Joel on Software.

We're closed, folks!

Links:

» Business of Software FAQ
» The Business of Software Conference (held every fall, usually in Boston)
» Forum guidelines (Please read before posting!)

Moderators:

Andy Brice
Successful Software

Doug Nebeker ("Doug")

Jonathan Matthews
Creator of DeepTrawl, CloudTrawl, and LeapDoc

Nicholas Hebb
BreezeTree Software

Bob Walsh
host, Startup Success Podcast author of The Web Startup Success Guide and Micro-ISV: From Vision To Reality

Patrick McKenzie
Bingo Card Creator

RULE: Scratch to beta in a week

Some people have been indicating that they take a long time to develop products, anything from 6 months to many years. I feel a bit disconnected, this doesn't seem to match my ISV view of the world.

My criteria for new products is that I should be able to develop from nothing to an almost complete working version in 7 days. That leaves another week for testing, polishing, snagging, ecommerce, website and initial marketing push.

Is anyone with me on this?
Adam
Sunday, March 05, 2006
 
 
Adam,

It all depends on the application, but unless you live in some parallel universe, I doubt you can build a worthwhile product in a week.

If you can, please tell us how!
Bob Walsh Send private email
Sunday, March 05, 2006
 
 
I'm releasing a Hello World "BETA" in about 10 minutes.
Floyd Price Send private email
Sunday, March 05, 2006
 
 
"My criteria for new products is that I should be able to develop from nothing to an almost complete working version in 7 days."

I've seen systems that take years to develop with multiple developers.  It all depends on what the fucntuionality of the app is.  Blind assumptions such as this get you no where.

A complete working app in 7 days would be a very small and limited program IMHO.  Reality is that you can only do so much in a limited time period.

Sunday, March 05, 2006
 
 
There is some truth to the idea that for a micro ISV you should choose something small enough that the first beta is quickly obtainable. However, it would be difficult to make a rule. Even the implication about starting out with a Beta is not a general rule.
Anon and anon
Sunday, March 05, 2006
 
 
Not with you on this. Unless you want to give the product away for free, you will be selling into a market with very low barriers to entry. That will be a very crowded market place and I figure you'll get out of it pretty much what you put in.  Very little.
wip
Sunday, March 05, 2006
 
 
Seems to be a lot of nonsense posted here lately, basically to summarize - "Quit your day job and make 3X your salary in a year with little to no effort or investment.  In just 1 week and with $1000 you too can be an microISV millionaire!"

Sounds like promoting microISV is the new infomercial to sell books and business plans.  Make is sound all upside with little cost and everyone will buy in.

It is a business like any other.  You need to invest the proper time, money, and effort to make a product that people need and want.  It is not going to happen over night and if you churn out crap in a week you will end up with crap to sell.  If anything running a microISV is going to be harder and cost more time and money than anything you have ever done before.  You are 100% on yor own, with no team or cash pool to draw on.  Success or failure is up to what you put into it and what the market deems as useful or not atthe end of the day.  Sure you may get lucky and hit a product that sells like hotcakes and costs nothing to do but that is not the norm.  We all can't sell chemical software out of our parents garage and not even try for sales and still get rich; that only happens in the movies and in blog posts.

Sunday, March 05, 2006
 
 
"With my shooting software, I conceived the product, coded it, and was selling it within a week."

"Getting the software out there so quickly showed me that starting simple is best because you can add the features that are most requested later while starting to make money immediately."

-- Brian Plexico of http://microisv.com as interviewed in Micro-ISV: From Vision To Reality

[Just an interesting case in point]
Nick Hebb Send private email
Sunday, March 05, 2006
 
 
I agree with noname. It appears the ISV-bubble is growing thanks to Bob's and other authors' books.

Not that ISV is a fad, but too many people are jumping ship to get rich quick.

The bubble will burst and the ones with the sound bizplan will survive, but there will be lots of casualties!
dil.b.ert
Sunday, March 05, 2006
 
 
"With my shooting software, I conceived the product, coded it, and was selling it within a week."

In all honesty I question how any one person can properly design, develop, and test a product in one week.  It certainly cannot be properly tested or documented in that time frame.  This is not an attack on anyone but jsut a comment from reality.  If your rushing a product through that quick you have to be makign shortcuts and the final product has to suffer to some degree from that fact.

Is there a user guide or help file?
Has it been tested on multiple OS versions?  Against multiple date/currency and other international settings?
Has anyone other than yourself tested it?

Sunday, March 05, 2006
 
 
> Not that ISV is a fad, but too many people
> are jumping ship to get rich quick.

This is much better for them than joining open source projects, working for big coporations for free *, and ruining software development as a profitable job for all the rest of us.

* - big coporations such as IBM are making a lot of money from open source
Jarred The Arab
Sunday, March 05, 2006
 
 
Excellent, my post is seen as an indicator of a MicroISV bubble!
Adam
Sunday, March 05, 2006
 
 
I had one of those perfect "Ah Ha" moments and coded an alpha of my application in 10 days worth of spare time.  It was really well received, but has no documentation, no website and no sales channels.  I expect to do a small amount of refactoring of the code itself, but mostly to make the code usuable in other vertical markets.

I am not however deluding myself into thinking I have a business ... I expect to finish beta testing and start marketing around the beginning of September.

I guess my main point is ... the application is actually a reasonably small part of a successful business.
Steve Moyer Send private email
Sunday, March 05, 2006
 
 
What's the definition of "beta," at least the commonly accepted definition?  Software that is for all intents and purposes ready to sell but needs a last-time workout by real users to uncover any final gotchas.

Only a very small, limited-functionality product can possibly be at that stage in a week.

Adam, are you talking about the above, or are you talking about a first prototype that shows static screen mockups to give potential buyers an idea of what you will have when you do finally go beta six months down the road?

The product my company writes has over 1M lines of Visual Basic code, 225+ SQL Server tables, probably 250 forms,  more than ten separate .EXE's in its suite, a desktop and handheld component, and links to other related apps in our industry.  If I wanted to write a competitor to that product, even with my domain knowledge, it would take me at least three months, full-time, just to write a spec.
Karl Perry Send private email
Sunday, March 05, 2006
 
 
Steve Moyer - well said.  An alpha, mockup, prototype, design, yes this may be possible in a weeks effort, but anything more than this would be suspect at best IMHO.

Sunday, March 05, 2006
 
 
I don't know whether there's a bubble or just the perception of one. It seems to me that there have been simple little apps listed on CNet and Tucows for years.

I think that somewhere in the wide universe of code, there exist some marketable applications that could be coded in a week. But they are the outliers.
Nick Hebb Send private email
Sunday, March 05, 2006
 
 
A ONE WEEK rule!  You must be on drugs if it takes you a full week to develop a product.  My rule is that when I wake up in the morning with an idea, I expect to have it coded and tested by noon, then if I haven't made 3X my salary by the end of the first day I scrap the whole thing.  It's too easy to make money as an ISV to waste time on anything else.

Sunday, March 05, 2006
 
 
"With my shooting software, I conceived the product, coded it, and was selling it within a week."

This was in reference to his second product.  It's not 100% clear from the interview, but it sounded to me like there were a lot of similarities between this and his first product -- that he took his firt product and customized it to skeet shooting enthusiasts. 

Also, I don't know too much about the skeet shooting software market, but I'm assuming that there's little to no competition there.  Certainly if you found a narrow enough niche, you could develop a product quickly to fit that need.  But that's going to be a lot different than if you're developing a product that has a lot of competition.
Ade Olonoh Send private email
Sunday, March 05, 2006
 
 
"My criteria for new products is that I should be able to develop from nothing to an almost complete working version in 7 days. That leaves another week for testing, polishing, snagging, ecommerce, website and initial marketing push."

I think some of the posts put on this site are merely there to get a debate going.  It's difficult to take such a post seriously.

But one thing this post does prompt is that we probably are taking far too long to get products out and how to trim down the time-to-market is a very worthwhile discussion.  And I guess what I would offer up that is being as well-schooled in the development process as possible.  Knowing and applying software engineering best practices, knowing and using design patterns expertly, using code generation, application frameworks, having libraries of well-tested code that are well-designed and can be re-used.  These are all things that can trim down that time to market.
Mike Stephenson Send private email
Sunday, March 05, 2006
 
 
Well, Aaron Swartz claims to have only been working for one week on http://infogami.com/.

Of course he started last summer and ended up throwing it all away and starting over, but that's not entirely relevant.

And what it does is approximately the same as those 15 minute ruby-on-rails demos do.
Joel Spolsky Send private email
Sunday, March 05, 2006
 
 
"I'm releasing a Hello World "BETA" in about 10 minutes."

If it's "Web 2.0" I'll give you $25 million for it.  :-)
!
Sunday, March 05, 2006
 
 
I some respects, he's right.

In my domain, the abstractions allow me to rapid prototype a system in a few days.  A few days spent designing, a few days spent coding, and a few days spent making it solid.

I end up with a proof of concept in less than a "coders" week.  Is it a finished?  Not by a long shot.  Does it demonstrate the fundaments, which I can then build upon? Indeed it does.

Unless you know every little detail of the final system/product, don't get stuck in analysis paralysis.  Get out there and start coding.  Having something for others to play with in a week is a great way to get early feedback and to help make the best/cheapest damn product/service possible.
Jayme Dunlop Send private email
Sunday, March 05, 2006
 
 
1 week is completely reasonable to release a web app if you have about 60-70 hours available of uninterrupted programming time. 

I really don't kow what takes people so long, I see things like 6 months, 2 years, etc, around here.  I just couldn't imagine.

However, when other people are involved, communication with partners and potential clients, then waiting for beta testers (such as in a small app I'm involved in right now) it can take a few months - but not all spent programming.
Ben Mc Send private email
Sunday, March 05, 2006
 
 
One week is a possibility. It depends. But as others have said, it wouldn't be polished enough.

I once wrote a script in ASP, uploaded on my website, submitted to a script directory and went away for a few days. Surprisingly the script survived longer than my website. It was a 3 or 4 hours effort. It was free and I didn't bother to include any documentation at all.
Umair Send private email
Sunday, March 05, 2006
 
 
+5 Adam

The replies are interesting because I think they illustrate that most of the posters are thinking in terms of a general purpose product with wide appeal.

The people who are talking about niches have it exactly right. There are many simple products that can be built in a day. The key is specialized domain knowledge combined with software development skills.

The funny thing is that in the time of DOS, a one week app would never have raised eyebrows, but many of the tasks those little DOS apps used to do years ago still need doing. And not all of them are included in XP, etc. Maybe the key is to look around and find some specialized small applications that people still keep DOS boxes around to run (I used to be one of those until recently) and create Windows versions.
lw Send private email
Sunday, March 05, 2006
 
 
And how much can you sell a "one week product" for?
Sounds like freeware to me, not something your going to sell for more than $5 or sell thousands of copies of.

Sunday, March 05, 2006
 
 
As the OP, just to clarify, this was a serious post and a philosophy I follow. One week for me is 60 hours of work. Much of the application code is reused from other applications.
Adam
Monday, March 06, 2006
 
 
"One week for me is 60 hours of work. Much of the application code is reused from other applications. "

So in reality a week and a half and not from scratch as you origianlly stated.

The problem with many of these recent posts is that people define things totally differently.  To me from scratch means you started with 0 ideas, code, etc.  A week means approx 40 hours.

Same goes for the cost to start a microISV.  Some don't count all their current personal equipment, software, or time while other do.

THe only real way to measure these thigns is to count everything and be clear on the definitions.  If you are resuing code then it is not from scratch, it is based on existing libraries, screens, etc.  If your working more than a typical employee week then it is not a typical week.  Need to compare apples to apples to make any real comments or suggestions.

Monday, March 06, 2006
 
 
I thought the post was outrageous too.  Then I remembered that version 1.0 of CeaMuS (under a different name) was done in about a week.  The product was useful in a limited way and some people (including myself) used it.

There were huge caveats though:

1. It was completely unsalable.  It didn't significantly differentiate itself from the available competition, and suffered from some limitations that competitors didn't.

2. I started with a lot of domain knowledge, since I'd been writing small content management systems for 10 years.

3. My development toolkit was as solid as it had ever been at that point, with all the kinks worked out and everything clicking just wonderfully.  It had taken me two years to put this toolkit together in this language, and a few more when you consider that I started it in another language entirely.
Clay Dowling Send private email
Monday, March 06, 2006
 
 
Because my interview with Bob was quoted in this thread I decided to provide more insight to my experiences about creating and selling an app in 7 days.  You can read my post at http://www.microisv.com
Brian Plexico Send private email
Monday, March 06, 2006
 
 
It is entirely possible to go from idea to beta in a week.  Particularly if you have done a program before that is similar to the one you are creating.  Add in the use of frameworks, domain knowledge, existing code bases that the company (or programmer) already has, code generation, macros and really high level languages and getting something working in a week is not so farfetched.  If you get a couple of weeks to think about it while you are driving 3 hours per day to your day job, then once you actually start coding it can easily take less than a week (week = 40-60 hours).

Further what if you miss your week deadline, it's not like you didn't get a whole bunch done and now you just need to finish everything up the next day.  Also, it's not like the coding stops once you get to beta.  You are going to show it to people, they are going to use it and make suggestions.  That's the point of doing something this quickly.
Joshua Volz Send private email
Monday, March 06, 2006
 
 
"If you get a couple of weeks to think about it while you are driving 3 hours per day to your day job, then once you actually start coding it can easily take less than a week (week = 40-60 hours)."

And this is all time applied to the project.  As I said before there seems to be some selective inclusion on what counts as time/money spent on a project.

I could probably build a car in a week too from scratch, if scratch meant I had all the plans and parts ready for assembly and by a week I actually meant 16 hour days for 7 days and I studied the plans for 3 weeks prior and made a plan of attack.

Monday, March 06, 2006
 
 
It seems to be this entire discussion is really a debate as to what constitutes an application.  In essence, this is an apples and oranges discussion.  To some, it is simply a software program that does something, regardless of how miniscule that something is.  To others, it is a polished work that probably does all the things you expect of a polished app (undo/redo, drag and drop, export to various file formats, etc....).  So yea, something can be cranked out in a week.  I have tons of "helper" apps that I've written for my own use that are useful to me but also very unpolished.  They generally took less than a day to develop.

While I was one of those that said a week is too short, I actually thought about my own experience, and yes I have done such an app and even sold it. In fact, it was developed in less than a week.  Was it anything fancy?  No.  Did it sell well - not really, but in terms of dollars per hour for my time invested, I got a very good wage.

Most of my apps have taken weeks and months to get out the door.  It boiled down to adding enough functionality for it to do what it was supposed to do and to have the polish that justified the price.  These apps - none of them - were one week apps.
Mike Stephenson Send private email
Monday, March 06, 2006
 
 
"It's possible to develop and sell a small tool in less than a week." Check.

"All applications no matter how complex must be developed and completed in less than 7 days, including all documentation, testing, internationalization, etc." Yeah, sure, anything you say.

My problem with the OP isn't the fact that some things an be done in a week - it's the assertion that everything worth doing can be done in 7 days or less. Even the "oh, alright, you can have a second week for polishing and marketing" doesn't make it a more sensible rule.

Monday, March 06, 2006
 
 
Huh. So people pay $17.99 for that app?

Cool.

I guess in my own mind, I have a hard time believing people will pay money for an application that could be developed entirely within an excel spreadsheet.

I've been working on my own application for the last five months (spending 15 to 20 hours per week on it). But my application involves simulation, machine learning, statistical analysis, etc.

It's a strategy-testing tool for poker players. Originally, I had no intention of selling it. It's a tool for my own AI research (sharpening the saw, you know?) but some of my friends have expressed an interest in it, so I'll probably try to sell a few copies when I get it into saleable condition.
BenjiSmith Send private email
Monday, March 06, 2006
 
 
Ok, so Brian found some degree of success with a product that was written in a matter of days.

What kind of revenue can you possibly generate from a product that is that simple to make? And can you repeat the success?

If something is that easy to produce then it's going to have little value on the market. Doesn't mean you won't sell a few copies, but I don't think it's going to make for a stable business.
Mark Hoffman Send private email
Monday, March 06, 2006
 
 
Benji,
Yes, people pay $17.99 for the app and it is pretty cool that people buy the app.  Not everyone has Excel or wants to track everything in Excel.  For some people, setting up Excel to track their scores is simply too complex. Clicking a square to change it from red to green is the simplicity they are looking for so they are willing to spend $17.99.  Also, my post only talked about version 1.  I have since spent much more than 7 days adding increased functionality that was necessary to make the software appealing to a broader audience.

Mark,
My opinion is that you can make as much money with an app this simple as you could any other app.  Its about the marketing and getting the word out to the people who are interested.  An app that solves complex problems will be usually be complex.  An app that solves simple problems can be complex, but not always.  Look at all the non-software examples there are in the world.  People are always coming up with simple products that make millions.  Software doesn't have to be different just because its software.  The best current example is the Million Dollar Home Page, a simple idea that made the guy a million dollars.  While his exact idea probably isn't repeatable, some new and slick idea will pop up soon.
Brian Plexico Send private email
Monday, March 06, 2006
 
 
Brian - What sort of ballpark sales figures does the product have since launch if you don't mind me asking?

100s, 1,000s, 10,000s?

Tuesday, March 07, 2006
 
 
100's.  which is more of a marketing issue.  I used eBay as my primary sales venue and once the eBay fees increased time and time again, coupled with the time spent mailing CD's it wasn't worth the time.  My non-eBay marketing was dealt a blow when Google disallowed a lot of my AdWords ads for being "gun related".
Brian Plexico Send private email
Tuesday, March 07, 2006
 
 
That's a brilliant idea. For the first release of the http://www.ubercode.com compiler I could release something that would compile IF statements and integer assignments only. Eg:

  Program ::= [ Statement ]*

  Statement ::= Declare | Assign | IfSmt

  Declare ::= "var" ##id "as" "integer"

  Assign :: ##id "<-" ##int

  IfSmt ::= "if" ##id "then"
            [ Statement ]*
            "end" "if"

Then I could extend it by adding a statement each week!
William Rayer Send private email
Tuesday, March 07, 2006
 
 
Thanks for the info.

So approx $5000-$6000 in sales for a week+ of effort plus sales, production, and marketing costs and time.  Approx $1500-2000 profit?

So probably a bit better than break even at this point, which I guess is not bad for a small applciation such as this.

Tuesday, March 07, 2006
 
 
If you've never programmed before, it takes 8 days to turn out a product:

Day 1:  Sams Teach Yourself <whatever language you use> in 24 Hours

Days 2-8:  turn out your application
Couldn't resist Send private email
Tuesday, March 07, 2006
 
 
actually, you could.

You would be amazed at how much people will spend for a custom MS Access database that anyone with a brain could churn out in less than a week.

Have a friend who does this as a nice little side earner on weekends. Think doctors' surgeries and the like
Tapiwa Send private email
Wednesday, March 08, 2006
 
 

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics
 
Powered by FogBugz