Freelance Programmer Survival Manual

Don't be afraid of losing jobs because you charge too much. When I first started on oDesk I charged $10 an hour, because I didn't want to put people off. Then I realized that kids in my neighbourhood make more walking dogs. Since then I've put my rates up considerably, and even though I still get nervous when talking money, I've never had a client say it's too much.

Charging more lets people know you're serious. Low paying customers typically ask for way more "can you just" tasks, like changing the font colour a dozen times or moving something "just a little to to right". You don't call your lawyer to change the typeface on your legal letters.

You are right about the "he can have partners who will make your life a living hell" part. My judgement might be incorrect, and I can't predict what will happen next as well (e.g. the project got taken over by someone).

Precisely. The biggest mistake newbie freelancers do is go for fixed prices projects and then allow feature creep.

You would still have a scope, right? And I assume you would give the client a quote or approximate duration of the project, right?

So one week iterations with a showcase at the end?

So are you collecting money up front?

I wish the customers at this part of the world (Malaysia) are sophisticated enough to accept the concept of "time & material". Most of them will refuse to pay if they are not satisfied with the work (in regardless of the completeness); and a partially completed work is of no value to a typical customer (especially non-IT customers).

The only marketing I did is word of mouth: friends & family, those are usually the good paymaster due to our "relationship". Yup, returning customers are the best.

I didn't try any elance stuff, because I felt it's a race to the bottom price with programmers from India.

I move on to develop a restaurant review website (sort of lifestyle business): http://food.malaysiamostwanted.com/

Maybe because you spread the risk? Many small will allow you to do a lot of small projects ; when a client leaves it doesn't matter, while if you have a bigger client (who will have bigger wishes), you are usually suddenly stuck for month in a project and if that client leaves you are done?

That was my reasoning at least when I was a freelancer; that's why I didn't continue doing that and started hiring people quite soon; the risk of losing a big client is too stressful. YOU depend on it; you feel it immediately. Funded startups; I would never do that anymore as freelancer. I had a few; cases in point here. Working on a project for 3-4 months only to see the funding had run out and the company was broke => no money. Funded startups are horrible to work with/for if you cannot peek into their reserves and financial situation. They will play nice weather until the very last moment (they have to of course) when you receive an official letter that they are gone and you can call some lawyer to get in line for your money (which you will never get).

If you don't want to 'work from home', I would say corporates (at least in north EU) are your very best bet at stability. I know guys who have held stable freelance jobs for decades at huge corporations. It's like having a job though, so I don't really see the point.

I just like small projects because I get bored quickly. Three months is about ideal. Projects for larger clients also have a tendency to go nowhere, that is they never make it into production. It is easier to get good rates from larger clients, though.

I guess because if you have one large enough client they become your defacto employer but without any of the benefits of salaried employment.

It can also make you complacent so you won't bother marketing yourself so much. That means that if they do decide to ditch you then you are in trouble.

Perhaps the answer is to provide smaller niche services (performance optimisation , pen-testing, UX design) to larger clients and charging a higher hourly rate so you don't end up becoming their general purpose code monkey.

Having more smaller clients means that my status as a freelancer increases due to there being more people who can vouch for me. There also is the people factor. I like to work with startups and help get their MVPs out of the door. Which means that I'm not only coding, but helping people follow their dreams. You don't get that big personal sense of accompishment when some big-Co. is succesful with something that you helped build. When the smaller client hits it big with your code it is a very good feeling. Nothing like helping the little guy get his dreams.

Another point is that smaller clients are more straightforward with me. They just want to get their project done. Big-Co. usually has a list of stupid requirements and proposals that make me waste too much time. In my experience, smaller clients have always paid in time. Big-Co, not so much.

It's easy to get started with smaller clients, with more manageable scope (for newbie) and to built up the portfolios.

Many local clients (especially from non-IT companies) are not sophisticated enough to use services such as elance, thus they are fairly happy to found a programmer which they could work with locally (my customer still expect face-to-face communication).

Eventually I did landed on a semi-government job, where they pay is probably 3x more lucrative than smaller clients, but longer duration + more politics + more stress to handle the human side of things.

The worst is "uneducated" clients, meaning they don't know how custom software development works, and asking why you charge 10x more than off-the-shelve accounting software and still deliver a product "full of bugs" (you should have tested vigorously and solve all the bugs, right?).

It is just hard to get freelance work from large employers, IMHO is just not on their comfort area.

I came from years as a contractor for large Fortune 100's who pretty much beg me to come back as a contractor but won't/can't send me projects as a remote freelancer. It is to radical for their business culture.