Find out as much as you can about the problem: A clear problem is easily understood. A problem you understand is easier to solve. For example, a clear bill is easily paid - one line of fine print, and you'll have questions. Determining the problem domain will help you to take off in the right direction. For example, finding out about financial practices affected by a finance problem is necessary first - anything else will only take you off in a tangent to where you were headed.

Start with the tried and true -- explain the problem to yourself: Communication is, sadly, not mankind's strong suit. We trip over each other for simple errors in understanding. And clients may not often be aware of exactly what they want. The only one who knows exactly what you mean is yourself (even if you may not always say it right).

Tell yourself clearly, what it is you are trying to do. Write up a (one page maximum) description of the problem as you understand it. This will also alert you to any issues you may encounter. Rest assured, if you cannot explain the problem, you will not be able to clarify any loopholes in it, and therefore, you will not be able to solve it completely.

Choose a standard technique: Choose techniques that are well known. This saves time you would otherwise spend on bringing other people on board with your ideas. In cases where others make this decision, assimilate their point of view. Any given problem can be solved in many different ways. Be open to ideas from all members involved, and offer your own. During your career, you will find yourself in different chairs, so be prepared, positive and co-operate.

Fly higher: Once you understand the problem and choose the technique to solve it, you arrive at the point where creativity will separate you from the typist writing code from design. In his book, 'Biplane', Richard Bach says: 'When one is at twenty thousand feet, approaching a major metropolitan area, one has a choice of several runways to land on. As one reduces altitude, the choice gets restricted, until there is only one runway to land on.' Try to find as many alternate solutions - those runways - as you can think of. Remember, your first cut may not seem so first-rate once you evaluate your options. Some of your solutions may be outrageous, some may be practical and some may be downright nonsensical.

Choose wisely: We have chosen the runway. Time to land the thing. A good pilot can land a plane in infinite ways, but not all of them will result in safe landings (production grade systems) and unhurt passengers (clients). This is the turning point. This is where you have to decide how creative you want to be. This is where you choose to live or die.

Run your solutions through constraints on the problem - time, money, skills, equipment, and so on. A viable solution is one that costs the client an average amount of time, money, and skills. An elegant solution is a viable one that employs ALL your current skills to the best possible end.

People lose assignments because they forget that nobody is indispensable in this business. Each of us can be replaced - for better or worse. If you write an obscure solution in the hope that you will be around forever to maintain it, keep your resume ready. If you can achieve an optimum via media between all your constraints, you will be around longer - because you will have gained the client's trust. Trust, while delicate and requiring high maintenance, will keep you in the race for any future development at that client. That is how you win this race against yourself.

Remember to re-group: Several factors influence the success of a landing attempt. Your solution may need to be modified over time and adjusted as the problem becomes clearer to you and to the client. If your solution is as elegant as you can make it, chances are you will be able to spend the fifteen minutes per week required to run through the entire solution to make sure it meets the requirements, while you make those adjustments.

Optimize later: Every implementation has its highs and lows. Optimization will highlight the highs and cancel the lows in your particular implementation.

Once the solution of choice has been implemented, depending upon time constraints, you may get one or two tries at optimization before the solution is generally deployed. Determine for yourself, from your experience and interactions, what the client is looking for - speed, space, glamour, and so on. Then, fine tune your solution to remove any extraneous noise in the relevant department.

Beautify last: Business casual attire dramatically increased productivity at work by making people feel good about how they looked without much effort.

Your solution should be similar. Keep the frills to a minimum. Increase usability over glamour. In the final analysis, the beauty of your solution will be in the simplicity of use and maintenance. Your client may not appreciate the inner beauty of your code, but, trust me on this, you would have earned the respect of the person who follows you to maintain that code. That, is a good thing in this small world of consulting.

Be careful: If you end up maintaining inherited (from someone else) code, you may have a piece of cake or of garbage (or something in between) that the client wants adjusted to her changing business climate. Use your creativity to retain those parts that you are comfortable with, and modify what you know you can improve. Again, watch the development life cycle closely. It is more difficult to maintain code - writing afresh is always easier. But, you will have to maintain a production system, in all likelihood - so be doubly careful. Trust, in this case, will be very, very difficult to retain (and/or regain) if something breaks. In the event you make a mistake, admit to it first. And rectify it immediately.

Intangible though it is, creativity relieves the tedium and boredom associated with mundane IT assignments that are so commonplace in today's workplace. It can help you keep your current assignment and land you newer ones. Develop yours, and watch your job satisfaction index skyrocket!