Fog Creek Software
Discussion Board




Making it on your own

I'd like to hear about experiences in starting your own business, software related or otherwise, or links to such experiences (besides the story of fog creek since there's this lil' website that mostly covers it).

Tell us how you got your first customers/clients, or when you actually knew you were doing alright and were going to make it as a consultant/entreprenuer, the uncertainty, the blood and sweat involved (or not), successes, failures, etc.  Not particularly interested in rags-to-riches, but rather in those making a decent living.

John Mohsen
Friday, January 03, 2003

Paul Graham who developed Yahoo store with some others has great articles on starting a software company:

The Other Road Ahead http://www.paulgraham.com/road.html

Beating the Averages
http://www.paulgraham.com/paulgraham/avg.html

Revenge of the Nerds
http://www.paulgraham.com/icad.html

Also this site "Joel on Software" is hard to beat:

See the "Software Strategy: The Big Picture " section: http://www.joelonsoftware.com/navLinks/fog0000000247.html

A great book too is: "Hard Drive: Bill Gates and the Making of the Microsoft Empire"
http://www.amazon.com/exec/obidos/ASIN/0887306292/qid=1041588787/sr=2-1/ref=sr_2_1/103-8965068-7643832

Matthew Lock
Friday, January 03, 2003

I'm currently reading an old edition of "Million Dollar Consulting" that I found on our book-shelf one day:

http://www.amazon.com/exec/obidos/tg/detail/-/007138703X/qid=1041598516/sr=1-1/ref=sr_1_1/103-3829423-9530239?v=glance&s=books

It's REALLY GOOD.  A few key ideas:

1) If you want to make big bucks, stop charging by the hour and start charging by the project.  (People balk when they hear you charge $200/hour, but if they just hear that a perl script costs $1,600 to develop, they might just buy it.  You don't have to tell them that you can write and test it in eight hours ...)

2) You can choose to compete in products, services, or relationships.  If you try to be world-class in all three, you'll kill yourself.  He suggests being good at products, better at services, world-class at relationships.  (Very similar to what Joel has taken:  All his community-building stuff is world-class relationships.)

3) If you want to consult, start with:

A) Calling local non-profits and governments and doing pro-bono work,
B) Bidding on government contract work,
C) Make sure your name appears on the resource list for governments, non-profits, and large businesses. 

4) Articles/Presentations:  Get your name out there.  (Think Joel, AGAIN ...)

5) How much you need to spend to get a professional office, answering service (not machine!), FAX, photocopier, Logo, etc.  It's about $10,000.  Spend less, and you look like a Lone Wolf.  Spend more and create debt ...

6) Which leads to:  How do you finance your consulting shop?

Tons of stuff on strategy and growing your business.  I highly recommend it.  You can probably get it for free from a library in your area.

regards,

Matt H.

Matt H.
Friday, January 03, 2003

Yes, Million Dollar Consulting is a good resource.
It's good advice and at the same time, it's not so simple to have everything up and running like he says...
It's forward moving tough.

Philippe Back
Friday, January 03, 2003

Matt,
            Pretty neat post on consulting. I'm stumped by the $10,000 though. If you're hiring an office for yourself it's too low, unless you're out in the sticks, and if it's just for equipment then it's way too high.

            You can get a fax number free if you pay $10 or so a month for outgoing faxes, and the faxes will be emailed to your computer. Much better than having a physical fax machine or tying up a whole phone line simply to be able to receive faxes on the computer.

              You could pay an agency to take your calls so it sounds as if you have a dedicated secretary. I'm not sure you would not be better off simply giving your mobile number (appeaiing lean might have advantages in today's market). Probably the best thing to do is to get the agency secretary to give out your mobile number when somebocy calls  in.
           

Stephen Jones
Friday, January 03, 2003

>
You can get a fax number free if you pay $10 or so a month for outgoing faxes, and the faxes will be emailed to your computer. Much better than having a physical fax machine or tying up a whole phone line simply to be able to receive faxes on the computer.
<
You can setup a faxserver for the cost of a computer...$200 bucks and charge people for faxes. If anyone is interested I'll do it and you can fax through my system

   
Friday, January 03, 2003

Be very careful about fixed-price contracts, especially if you are starting out.  If you charge $1000 and figure that it's 5 hours at $200/hr, you had better be prepared for the case where it takes 10 hours and you only make $100/hr because you overestimated it.  I've seen quite a few companies end up shooting themselves in the foot that way.

My strategy was a very good estimate for the initial project cost and an hourly fee, and then let the client do change requests and go over the initial estimate.

w.h.
Friday, January 03, 2003

I've almost lost my shirt on the fixed price contracts I have done. You might be able to get away with a $1600 perl script, but those types of contracts are few and far between. Most single person development projects the client is looking to spend between $20K and $50K, and if you charge fixed $20K, you are going to eat the $30K yourself, when the client squeezes you with feature creep. 

consultant
Saturday, January 04, 2003

Feature creep is a problem only if your management sucks and doesn't know how to track requirements.  In a fixed-price project you charge the client for every additional feature.  In fact, that's how some contractors make money despite bidding low in the beginning.  They lowball the competition to win the contract, because they know the project is one that is likely to require many modifications, then as the customer wants more and more they tack it on to the bill, and in the end the changes add up enough that they make a decent profit.

The main problem with fixed-price projects is creating a good estimate that won't put you in the red.  It is especially difficult with software, as managers almost always grossly underestimate the cost and effort of software projects.

T. Norman
Saturday, January 04, 2003

>>Feature creep is a problem only if your management sucks and doesn't know how to track requirements.  In a fixed-price project you charge the client for every additional feature

I agree with this, but getting fixed price to work as a one-man shop, just starting out, is VERY hard.  First of all, the client will not want to pay for a design phase, where you cook up the requirements.  Thus, it is hard to get good requirements whilst getting paid,  in the first place. The requirements will change constantly throughout the development of the software. If someone decides that they want their workflow app to flow in a different direction, after the first rev, they are usually _not_ happy when you say "that will be an additional $8000."  However, if it is priced by the hour, you just continue getting paid your hourly rate.  I'm not saying it is impossible to do fixed price, it is just a lot harder to do than hourly, _especially_ if you are just starting out.  I question a consulting book that recommends fixed price to a beginning consultant, since nearly all other resources caution against fixed price contracts.

consultant
Saturday, January 04, 2003

"In fact, that's how some contractors make money despite bidding low in the beginning.  They lowball the competition to win the contract, because they know the project is one that is likely to require many modifications, then as the customer wants more and more they tack it on to the bill, and in the end the changes add up enough that they make a decent profit."

Should note that this is generally referred to "bait and switch" by the client, and is not conducive to repeat business nor recommendations...  (So if you are counting on word of mouth to spread the news about you, or if the client has the potential to send a lot of business your way, I wouldn't recommend setting up a lowball bid based on the expectation that "add-ons" will make up the difference).

There's always a delicate balance between avoiding the perception of bait and switch while preventing the client from bankrupting the project via feature creep.  The best way to handle this is to have a very clear set of "included" features from the get go and make sure the client understands that changes from this feature set are billable.  Also, it is better to defer many feature changes until a "Phase II" of the project - since they are often nice-to-haves as opposed to critical requirements (provided your needs analysis was appropriately done), and since it will be difficult to stay within your original time frame if you add many new features (as opposed to substitutions).  Of course, that doesn't mean being completely inflexible / insensitive to what the client wants ...

Phibian
Sunday, January 05, 2003

Rather than considering how to charge, you might want to ask yourself whether you are a products person or a services person.

I have sold both hourly consulting and a self created product, and I am much, much happier with the product company I have evolved myself into.

The reason is simple- consulting projects of the size you can do solo or in a small or even medium sized company come largely through personal relationships and referrals. Fine if you have lots, and enjoy selling. Personally, I hate it- and while I don't mind talking to customers via email or the occasional phone call- by and large I have no desire to go out and drum up business.

So, I built a product around my consulting expertise that my customers can use themselves. I sell a lot more, at a much lower price point. At the end of the year I have made about the same in terms of money, but I am a much happier person. I can tinker and create and take long vacations without having to worry about meeting other peoples timelines or placating their neurosis. I don't have to sell- they can visit my website and buy if they like it or move on if they don't.

My point is, decide what you want to do and what kind of person you are not just in light of your basic skill set (programmer, for example), but in view of the fact that going out on your own means you will have to do it all (sell, manage, accounting, customer support, etc.), so how you do it becomes as important as what you do. Consulting is only one option, and not necessarily the best.

Good luck,
MCR

MCR
Tuesday, January 07, 2003

*  Recent Topics

*  Fog Creek Home