Fog Creek Software
Discussion Board




You can't be an accountant and run a software biz

It's a fact.  You can't just be an accountant (or any other occupation) and run a software business.  You absolutely, positively must have knowledge of the product you are making, the process used to make it and how it is made.

True story.  I worked for a small software company out of college.  I took the job because it was the only offer I had.  Big mistake. (Live and learn)  Anyway, The owner was an accountant.  This guy didn't know jack about computers.  He could barely use his own accounting program and would ask me questions about MS-Word and how to download things off the internet... read porn.  At any rate the software was dispicable.  It didn't work.  Customers would call and complain and this guy wouldn't refund their money.  He hired consultants who claimed to be able to write software and had us, the full time employees maintain it.  The business was struggling to say the least.  He couldn't pay me or my co-workers, our checks would bounce.  When I approached him as to refactoring the software so that it would do what the customers wanted he would simply ignore me or become upset and say something like he had payed a consultant 20k to write the program and that "it had better work".  The fact is these consultants he hired were from the boom era, were self-taught and knew how to write code that worked like a boat with a hole in it.  If this guy would have been knowledgeable of the management and construction of software he would never have gotten himself into the mess.  I tried everything from giving him articles on JOS about eating your own dog food to going so far as to demonstrate to him that the software didn't work.  He still didn't want to do anything about it and to this day sells the same garbage.  I could give you the URL, but I won't go that far.  (You would really have a good laugh though.) I quit the position and am glad I did.

Now maybe I'm crazy, but I know for a fact that any business you get into you have to be knowledgeable in that business.  You can't be an executive at a Fortune 500 company and then one day decide you're gonna run a dairy farm.  You have to learn the business first and then you have to have the motivation to run it.  What keeps Joel's business running?  It's his and his employees' knowledge of the business they are in.  That is the software business.  The next company I work for (if the chance presents itself), I'm gonna make damn sure that the person I report to and/or work for/with knows the software business.

My Own Prison
Wednesday, May 28, 2003

I couldn't disagree more.  You have made broad, sweeping generalizations about management ability based on a single personal experience.  Sure, you got a bum ride but don't let that taint your perception of what it means to be a good manager.

Every newly minted MBA has read a case study of what Lou Gerstner did at IBM.  I believe its really nothing short of a miracle.  And he was a bean counter, of sorts.  Certainly no background in hardware or software.

I started working for myself sometime ago since I got fed up with incompetent managers, both with and without software backgrounds.  They were equally as bad, but for different reasons.

At present, I am under contract on a project in which the product manager, director and the VP have absolutely no IT background.  Its the best job I have ever had.  They have taken care to hire great people and then got out of thier way.  And for this, we make an extra effort to ensure things are done properly.

Canuck
Wednesday, May 28, 2003

You can't extrapolate your experiences of a bean counter at the helm of a product most likely done in VB by idiots to anything else but another product don in VB by idiots

Mike
Wednesday, May 28, 2003

A CEO's job is to lead and trust his people. Your boss did neither. It had nothing to do with his lack of knowledge of computers.

In fact, I'll wager that if he *had* known programming, then he would've been micromanaging you guys about how to do your jobs, and this thread would've been "You can't be a software developer and run a software biz"

;-)

Philo

Philo
Wednesday, May 28, 2003

I think the problem isn't the fact that your boss is an accountant.  I think it's that he's an asshole.

anon
Wednesday, May 28, 2003

The original poster makes an excellent point. People producing Gertner and others as counterpoints miss the point that, in such large organisations, those people are managing sub-businesses, not software products or development. At the level where software expertise is needed, IBM has it.

As has been mentioned before on JOS, software is the only expertise-intensive professional activity where important directions can and often are taken by people with no expertise in that area.


Wednesday, May 28, 2003

Where is Domain knowledge fitting in to this... This individual apparently thought he knew accounting and it is implied that this was an accounting package he was hawking.  If a person who only knew software tried to write an accounting package, it would also be a story of " You can't be a developer and run an accounting software biz".  It more sounds like he knew nothing of management and did not understand his market.  This along with a lack of drive to succeed (I don't think a truely driven person would waste there time figuring out how to download items during 'work hours') was his down fall.

A Software Build Guy
Wednesday, May 28, 2003

The no name poster is basically thinking along my lines.

> "it is implied that this was an accounting package he was hawking"

It's not an accounting package.

> "If a person who only knew software tried to write an accounting package, it would also be a story of " You can't be a developer and run an accounting software biz"."

I can see your point, but I respectfully disagree.  You CAN be a developer and run an accounting software biz.  A good software engineer would follow the rules of development and research, study and learn the accounting practice as well as interview accountants etc. before diving into making a software product.  Developers NEED to know the business their software is being used in - in order to write software for it.  As is my original point, the opposite is also true, if you're going to run a software business you need to know the in's and out's of software.

As mentioned earlier, and for those of you who brought up the large companies, and as was mentioned earlier these people really don't manage software products or development.

My Own Prison
Wednesday, May 28, 2003

And yea, you're right,  "A software build guy" - he didn't and still doesn't have the drive to succeed, only to just get by.  Probably because he doesn't know what the heck is going on and needs to learn a thing or two about software development, but is too lazy too and just want to get by on the income that he steals from his customers while using programmers just out of school to do his dirty work.  In fact he claims to "know the way these things work."

My Own Prison
Wednesday, May 28, 2003

Domain knowledge (knowledge of a real world area of need) and technical implementation knowledge are strange bedfellows.

My experience with vertical market software companies (companies that specialize in a SW product for a particular industry, trade or occupation) is that they are almost always technically mediocre; they have terrible technology cultures; and such companies are usually run with a ham handed mentality that software just doesn't count. It's quite rare to see a vertical market company that considers technology a necessary component of their value proposition.

In a sense, things 'have' to be that way.  Technical skills plus no domain knowledge of a real world application area == no revenue. Domain knowledge + barely passable technology implementation == possibility of making some money from people who need a solution in that domain.

Where this becomes 'bad' is the vertical market company that has developed an internal culture that more or less says that technology just doesn't count. These are the companies that (for instance) nursed DOS real mode applications along well into the 1990s because the owners and management wanted to rationalize that Windows was a passing fad.  Or, the many vertical market companies that are run like "boiler rooms" with harrassed programmers being yelled at to work "lean and mean" and "stop wasting time". IE, hatred of technology process leading to a degree of persecution of technology people themselves.
In a sense, a vertical market SW vendor tends to be an "end user" of technology. The owners and the management may believe that technology isn't even something that they need to be concerned about.
I suspect that some of this dynamic is at work with the accountant-boss that the original poster described. He just wants to make money, and he is only interested in tech as a means to an end. In addition, as already noted, the problem is more that he's an a$$hole than that he is a accountant.

Bored Bystander
Wednesday, May 28, 2003

Bored Bystander gets it right as usual:

---"Technical skills plus no domain knowledge of a real world application area == no revenue. Domain knowledge + barely passable technology implementation == possibility of making some money from people who need a solution in that domain. "----

In fact I would say that the most important factor is the ability to sell the resultant crud. And that means finding an expensive "must have" feature, that in fact is quite unnecessary and nobody will ever use. In my field, which is English Language Learning, the "must have" is centralized tracking of every student progress and what they do on the machine. In practice this is never used because the lab supervisor doesn't have the time or inclination, and the results are probably too depressing anyway, but control freakery among trianing managers mean that the two big players in this market Ellis and Dyned, sell their software at $1000 a seat, even though Mindscape sell a program that is just as good (or just as mediocre to be more accurate) at $35 a seat.

Stephen Jones
Thursday, May 29, 2003

I've worked for several vertical market companies. All of them but one bumped along just making a profit out of the customers they didn't upset to the point of leaving.

All of them but one were "sales led". The actual product was a poor imitation of the idea the customers were sold.

"Sales led", I have come to understand means the company philosophy is "Sales promise the customers anything to get them to sign and are heroes when they do. Development are a bunch of tossers who can't live up to the dreams of sales and constantly lose us customers."

The kind of company where the sales force has a beemer each and the most expensive laptop money can buy and development keeps running out of disk space and isn't allowed to buy more.

Now, it may be coincidence, but the exception in both cases was the same company. The guy running it is from the industry they sell to, but he knows that the product has to sell itself. In fact it does - people see it running and come and buy it. It's THAT good. And it's that good because the company has a record of hiring fantastic people - top notch developers and good domain knowledge people and letting them go to town. When drivers for new kit are added, it's not enough that drive a couple of the functions - they have to drive ALL the functions. Properly. And have a cool interface. And be scriptable. And auto-configure, and be as plug and play as humanly possible. And yes. If it's a choice between "ship with bugs" or "ship later", they pick "ship later" everytime. Scary idea huh?

Frankly I'd much rather work for a company like that than the ones where they customers start out grumpy because the thing they get in the box isn't what they thought it would be... but there's only been that one company that's learnt that lesson.

Katie Lucas
Thursday, May 29, 2003

"You can't be an executive at a Fortune 500 company and then one day decide you're gonna run a dairy farm."

A story from my ancestors. My grandparents, among a lot of other things, used to run a hotel. One day they thought, sod this, we'll become sheep farmers - neither of them having ever worked on a farm, or with sheep before.

After ONE year they had established a reputation as producing some of the best sheep in wales, and could sell their sheep at a significantly higher price than their competitors.

It IS perfectly possibly to switch into a completely unknown field and excel at it. Your boss was an idiot and an arsehole. You are generalising incorrectly. I think at least a big a problem is programmer-led companies with no management skills or business acumen.

Mr Jack
Thursday, May 29, 2003

>> "These are the companies that (for instance) nursed DOS real mode applications along well into the 1990s because the owners and management wanted to rationalize that Windows was a passing fad."

Yea, I still provided support for the DOS versions of their products while they try to sell this shoddy windows version.



>> "Domain knowledge + barely passable technology implementation == possibility of making some money from people who need a solution in that domain."

It also = ripping people off.  I can't live with that.


To the person who said something about a sheep farm... If you read the next sentence after the one quoted from my original post you would see that I said it is not impossible.  I'm sure these people had the motivation and learned how (or learned at least a little bit about it) to run a sheep farm or whatever it is they had.

To the person who said there are businesses run by software people that are the same way... name one?  As I mentioned earlier a programmer HAS to know the business he writes software for.  Management is only part of the deal, knowing the business you're in is the rest.

I guess I'd rather forget the whole thing.  Working on 90MHZ Pentiums, having to clean my own office because no one would clean... anything, having to put up with customers threatening to kill me if I didn't get the software working, going to work every day and feeling guilty about what I couldn't accomplish... bleh...

I wish I could start over, but the way the market looks and that fact that my reputation is probably tarnished from working for this guy that's prolly a long shot.

Maybe I could start my own software business.

My Own Prison
Thursday, May 29, 2003

-----"Domain knowledge + barely passable technology implementation == possibility of making some money from people who need a solution in that domain."

It also = ripping people off.  I can't live with that." -----

Is it? After all the customer would buy something else if it was in the market. At least your company is offering him something that does what he's looking for, albeit badly.

And remember that technical excellence is just one part of the equation. One of the leading pieces of English Language software is economically written and fits on a floppy, so I suspect Delphi or C++. I am sure it is technically better than much of the competition written in VB or Javascript, but the stuff written in VB and Javascript allows multii-media questions, pictures as objects and everything else a good language authoring program needs. The best the first program could do was to let you put in a picture as background to liven up the page. Their problem is that features that are trivial to execute in VB require serous effort in programs that directly access the Windows API.

Stephen Jones
Thursday, May 29, 2003

Actually the inverse is more likely to be true, you can't run a business, of any kind, without at least a reasonable knowledge of accountancy. 

You don't have to be a chartered accountant, but you do have to be able to read a balance sheet and P&L and understand what they mean.

Simon Lucy
Thursday, May 29, 2003

(Stephen, thx for the kudo.)

It's really interesting to bring up this "real world knowledge vs. programming skill" topic. Programmers usually jump all over the proposition that a technically retrograde approach that otherwise serves the customer's business needs is deficient.


Stephen also got it right: the customer was free to not buy the crap and to hold out for a better product. The fact that the customer selected a crufty vendor's product indicates that the vendor, no matter how backward technologically, is doing SOMETHING well that the customer values and which can't be found even in a technically superior, or higher quality, package. This kind of sticking power in the face of mediocrity generally comes about by owners who are veterans of the industry being catered to. They know what sells.


Let's also differentiate some things. There is the notion that a product line may not have the latest bells and whistles yet is stable. I described this as an aspect of hackiness. But an alternative explanation may be that the vendor may be judiciously conservative about investing in leading edge technology that they don't feel that they can get right.  Much preferable to the companies that 'sunset' working and adequate products and force their users to upgrade to something fragile and buggy and "new".


It would be ideal if all vendors placed sufficient value upon technical competency and execution to assure that the resulting programming workplace was a nice place to work and generated top quality results. The fact is, the customer's buying habits will win out every time. If the customer doesn't notice internal badness or can work around it, they will still buy and will therefore support another one of the vertical market hack shops of the world.

Bored Bystander
Thursday, May 29, 2003

Thanks for your inputs guys.

The one thing (in my situation) that really irked me was the fact that the software just plain didn't work and this guy didn't want us fixing it.  Oh, you want reports, well they don't work.  Oh, you want to print the results of the MAIN function of the program, well it works but some of the results are missing.  Oh, you want to change the parameters of this or that function, well they're not implemented yet.  Oh you thought all this stuff worked? What's that you say? You want your money back?  Nope, sorry not gonna give you your money back.  Ok, I guess I'll just wait till you guys fix this stuff then.

Isn't that wrong.  These guys had a DOS product that was HUGE, I mean literally dominated the whole IBM PC market (for that niche) (Not Macs).  When Windows came out they didn't bother to start developing for it.  Now the year 2000 comes along and bam! Hey let's hire a cheap consultant to write our Windows Versions.  He may or may not write a good program but that doesn't matter to us.  We just shove software in peoples faces, hype it up and they buy it.  Not to mention that we treat our employees badly by bouncing their paychecks, giving them cheap equipment and working them to death in a trash hole because we can't sell the crap software.  I had to use a 90MHZ Pentium with a 1GB hard disk here.  You think VB likes that?  No it doesn't.

After I quit, (and mind you the only reason I quit was because my paychecks were bouncing) I went back six months later to see if this guy had my money.  Nope he didn't but the programmer I worked with had a brand new computer and 2 "college interns" were there "taking phone calls".  Basically this guy is an asshole and uses people and technology to make money.  In my opinion he should be behind bars.

Also, In my opinion you have to place sufficient technical value on your product or you are literally breaking the law by selling something that doesn't do what it advertises.  Of course you sell it under a contract that says it's not guaranteed to do anything anyway.  But isn't that stupid.  I think this guy will ruin his reputation or are customers that ignorant and stupid?

I mean give me a break, is this really the way it works?  If so I will not work for a vertical market app company again.  Either that or I will look good and hard at them before I do.

My Own Prison
Thursday, May 29, 2003

My Own Prison:

I've worked for some people who have used personal threats and who have engaged in outright criminal means to conduct their businesses.


I can't believe that the guy you worked for is still in business. He is an industrial strength asshole.


If your paychecks bounced and you're in the US, contact your state's labor relations board ASAP. W2 employment is a legally enforceble *promise* with statutory laws on the books that state that employee's paychecks take precedence over vendor payments and just about anything else except tax payments to the government.


Get cracking on this if you're still owed money. I know some hard core geeks who got screwed out of their pay by a criminalistically-tending business and they bought the standard stupid geek rationale that money isn't important enough to fight for. Screw that to the max, don't be naive, make this guy PAY and get your money, dammit!

Good luck. Sounds like you had been through hell. YES, this is the exception. Most of my clients have merely sucked out loud.

Bored Bystander
Thursday, May 29, 2003

My Own Prison:

Start your own biz & show him how it's done.  Use his dissatisfied customers as your target.

victim, jr.
Thursday, May 29, 2003

Steal his customers? Sure! Fight evil with evil. (Ok, the second 'evil' is really "screwed over disgruntled ex employee setting a grievance..." .. heh heh..)

Unless the SOB made 'My Own Prison' sign a non compete.  And guys that have this sort of attitude ("everything is MINE!!!") are usually obsessed with locking their people down.

Bored Bystander
Thursday, May 29, 2003

If he still owes you money, and the statute ot limitations hasn't set in, sue him.

And then leave it all to your lawyer and forget about it.

Stephen Jones
Thursday, May 29, 2003

The fallacy in some of the posts above is that development is nothing more than coding.

Development is about creating solutions to problems in an abstract space, using software which has to be designed and created.

Most problems in vertical markets can be understood by others. The developer then is able to design solutions to them, and implement those solutions in software. That's what it's all about.

Most people are not good at such abstract thought, and of those who are, most lack the understanding of the software environment that's necessary to design to closure.
The ones who do are first class developers.


Thursday, May 29, 2003

Katie's description of vertical market businesses is spot on. They start with sales-oriented people who see a gap in a market for which they have contacts and marketing features.

Those people then hire developers. Because programmers are just a cost to them, they hire the cheapest they can find, and probide abysmal direction and resources.

The software thus produced is 2nd and 3rd class, but performs some vital tasks for that industry. Customers are charged high prices for the products. The directors of the companies become wealthy. Programmers are always treated as dumb cogs.

Eventually industry associations commission someone to develop a competing product that's good and cheaper, or a good developer happens to discover the niche and moves into it.


Thursday, May 29, 2003

>> "The fallacy in some of the posts above is that development is nothing more than coding."

I don't see this in any of the posts.  Maybe you can point one out?


Thursday, May 29, 2003

re: non-compete clauses and the suggestion that My Own Prison should start his own business just like evil boss's, but do it right and beat boss at his own game.

I wonder...if you have a non-compete clause that is in effect for 1 year after leaving the company, can you develop software that does directly compete, but NOT MARKET OR SELL IT until after the 1 year?

loop hole...?

Lauren B.
Friday, May 30, 2003

>> if you have a non-compete clause that is in effect for 1 year after leaving the company, can you develop software that does directly compete, but NOT MARKET OR SELL IT until after the 1 year?

Who would know?

Bored Bystander
Friday, May 30, 2003

Well that fact of the matter is that I didn't sign any non-compete agreement, or for that matter much of anything.  He didn't want me to fill out an appication and made me sign a six months probation period agreement.  Along with the standard w-2 forms and stuff.  Personally I think he would have just dumped me after the six months anyway and was hoping that I would stick it out that long and be able to somehow sell him more software (I'm a programmer not a salesman or even a support tech (not that I have a problem with support)) during that time.  Well I didn't stay to see month 4 and why should I have?  I learned that this guy has a lien on his business from the state and that he had multiple law-suits against him from the contractors that wrote the windows software.  Which explains why that windows software is incomplete and doesn't function properly.  So the state depletes his checking account once a month at a time of their choosing.  Nice way to get your money back eh? :)

Anyway, I could probably develop an application that accomplishes the same function and sell it.  After seeing what the business is like though, I don't know why I would.  Maybe I need more business expertise myself or maybe I just need to dive in and "just do it" so-to-speak.

The thing is, I don't want to end up being driven towards a "profit driven mentallity" so that I end becoming like this guy.  I mean, you need to make a profit, but you also need to do it honestly and pride usually gets in the way because you don't want to see your business die and you're afraid to take up a new occupation.

My Own Prison (Starting to taste freedom)
Friday, May 30, 2003

I'm not clear if you're still owed W-2 pay or not from this guy. It sounded like you were and are.

If you are, call your state offices and inquire as to what department handles labor law violations like this one.

I would urge you to NOT rely on your own guesses of his ability to pay, or other outsider's observations, or any statements from him that if you press it you will drive him into bankruptcy and then you won't get anything. He is a crook and you won't know until you force it out of him using formal means.

The fact is, if a business can't pay its tax bills, then the business's owners and executives become personally liable. The same thing may or may not apply to payroll but a similar principle applies. I'm not sure that payroll goes back on the owners as personal liability but payroll does precede almost ALL other payments, such as to vendors, etc. They have to pay you before they pay the janitorial service or the SW contractors he used.

And you're not being like him or being legalistic in doing this. You're righting a wrong and you aren't a charity obliged to help a scumbag.

Bored Bystander
Friday, May 30, 2003

Yes, I am still owed W-2 money from him.

I have the bounced paychecks and the pay statements, but for the last two week pay period he wouldn't give me a pay statement or a paycheck.  He did go into his file cabinet and pull out the cash to give me half of one paycheck.  So he still owes me 1.5 paychecks.  The thing is he sent me a W-2 form and I filed my taxes, which might have been a mistake as the amount filed included half of the one paycheck that he still hasn't payed me and did not include the one whole paycheck that he didn't even give me. 

So really, I'm a fool for not seeing through any of this and pursuing it sooner.  The fact is I was scared, angry and hurt and my personality is one where I'm shy, held back and I trust(ed) people too much.  Guess I learned a hard lesson.

Anyway, I have went to the department of workforce development and filled out the necessary forms, we will see what happens next.

My Own Prison (Starting to taste freedom)
Friday, May 30, 2003

Prisoner: "The thing is, I don't want to end up being driven towards a "profit driven mentallity" so that I end becoming like this guy."

If you can make a working application, that helps the customers and the market is big enough for you to make a profit, I don't see the problem. By making a solution that helps the customers, you are doing a good thing for the society. Everybody is happy - except "this guy" maybe.

It does take some business sense, and selling effort, though, but its not necessery to be dishonest, like "this guy".

anonymous
Sunday, June 01, 2003

*  Recent Topics

*  Fog Creek Home