* Consider that your audience will not fully understand your idea unless you can effectively present it - take the time to build a visual presentation * Esti

* Consider that your audience will not fully understand your idea unless you can effectively present it - take the time to build a visual presentation * Estimate time/effort to complete the project, the cost of integration into the current IT environment, on going needs (people/ systems / support) * Revenue structure and model - per seat, per month, per licence, one time * off set costs with revenue to show time line of profitability based on a log of sales. be conservative. let other people you can sell more * build a prototype - even a poor one but make it look good. Most business types take visual cues to reflect on code quality. They do not understand Algos. Don't discuss it with people that do not understand the math / concepts. Talk about the solution to a problem instead. * If possible build two documents: one is a powerpoint presentation with the above information. The other is a prose document outlining in deeper information the system/ concept / etc * Give it away - you don't need to own the project. Find the individuals in your firm that can get it done. Get them interested in the project. Let them help you sell it. * Market Research is red ocean. You can only know what has been done and with what demographics. You can not understand blue oceans from market data. Determine if your product / service is new/unique/special/ and if it is red ocean do the work. find the demographic, understand the competition. Figure out the four forces (five) and break down their busienss / profit models. If it is blue ocean, map the shift. Where are you taking the model vs the current solution framework. * Be persistent, as that is the measure of an idea. If you wont fight for it, why would anyone else? The best ideas do not always win. * Break down the presentation with an action list. what to do next. who to do it. time lines. * Make sure you would use it, if you were the customer. Otherwise, why would you build it? (unless you are in finance, then this doesn't really matter does it?) * Patent the idea before you give it to you boss. Just for the hell of it. You can always licence it to the company. :-)