Hi Dave,
I had your problem when I first started my consulting business, 20 years ago, and it only took one or two clients to realize that I was doing something wrong.
In a phrase: charing by the project. ALWAYS sign a contract and ALWAYS charge by the hour, and ALWAYS bill no less than every two weeks. If they stop paying you stop coding.

Most potential clients do not prepare an RFP, or if they do it inadequately describes the project they want done. You cannot accurately predict how long it will take you to do a project without a very detailed description. You do not want to write such a description, which envolves many hours of interviews and data analysis, for free. Seems like a catch-22, but there is a way out of it.

I created a questionaire (sorry, mine is not available) which I gave to the potential clients to fill out on their time and money. The questionaire asked simple things like the name of the project and which people would be the primary contact persons, and complex things like the descriptions of existing software, data structures, input fields, data to store, reports to output, etc... I would review their response on my time, and sometimes call to ask clarifying questions. After a few exchanges they usually had what they wanted in their own writing. In effect, I had them create an RFP, to which I gave my proposal, in the form of my standard contract (sorry, not available) to which I attached an appendum describing the the job in detail using their RFP (and quoting from it).
The main body of the contract described my rights of prior knowledge and skills, plus any knowledge or skills I may acquire during the course of coding their project which did not arise from propriatary information, and their rights of propriatary information. I also included conditions allowing me to work on at least two other projects at the same time. (I cound never keep more than three going at once: one just starting, one in the middle of development and taking 80% of my time, and one wrapping up.) Explicit in the contract was my per hour rate billed bi-weekly, with pauses for non-payment, and verbage which said, in effect, that I will pour until the job is complete or you say stop. The contract also had a cancelation clause for such purposes. I gave a NONBINDING ESTIMATE as to how many hours I thought the job would take, which was usualy twice what I thought it would really take, just like Barbara stated. Neither party can predict in advance what problems will arise as the project unfolds, and since the client is the one who wants the work done s/he should pay for that work. As you may already know, computerizing a business is a process that reveals the inner workings of that business in ways many owners never understood. And, of course, the computer must always be programmed for the worst case mode.

You will lose bids with this method, but you will never lose money and, like Barbara, you will find yourself being called in to fix messes left by fixed-price programmers. In fact, I usually equate fixed-price programmers with inexperienced programmers. (Sorry, no intention to insult).
JLK



>Hello everyone,
>
>I've had my own software development business for about 3 years, and I've come to the conclusion that I'm ripping myself off. Since I work off-site, I give my clients a flat rate. For example, if I estimate that a job will take me 24 hours to complete, I will charge the client $75.00 x 24 hours. The problem is that these jobs rarely ever conform to the original time estimate. Communication is often very bad on the clients end, since they are either slow in answering my questions or send me incomplete or inaccurate descriptions of what they really want. A 24 hour job could end up not being completed for a month because of these communication issues. I've wasted a lot of time redoing applications, because I did not have the correct information to start with.
>
>Since I try to foster a good relationship with my clients, I would hate to tell them that I have to charge them more money because they did not communicate their specifications accurately enough or in a timely manner. Given all of this, I'm thinking that my rate is not high enough or I'm not estimating enough time from the beginning. Considering human nature, It will probably always be an ongoing battle to get the information I need from clients. I guess this is called consulting .
>
>Ok, here is my question: A potential client recently asked me for a quote and time estimate for building a database application which tracks the location of machine parts in a plant. The client also wants the database to keep a schedule of when parts will be sent off-site for repairs, and track their location. A sophisticated calendar interface is required by the client. In addition, the system will need to track which machines produced which products. No doubt, I will need gather much information about their operation before I build this.
>
>Finally, here is the question: What should I quote this client as far as a price, and how long should I estimate that it will take? I was going to quote approximately $3000.00 for 40 hours worth of work, which includes consulting, development, testing, implementation, and training. However, my partner thinks that I continually keep under quoting the price and under estimating the time for projects such as this. My dilemna is that I don't know if I'm charging too little or too much (I've lost a few prospects). Granted, if I knew exactly what a client wanted before I began programming, I could probably build this within a couple of days. But, that rarely happens. In the past I've lost quite a few prospects, and I'm not really sure if this has to due with sounding too cheap or too expensive. Is it my rate that is off, or do I not estimate enough hours?
>
>Any advice would be greatly appreciated....:)
>
>Thanks,
>
>Dave

Nebraska Dept of Revenue