Hi Richard,
Fixed price quotes can be a great way to make good money and to lose hard earned money, you have to take care and make sure you know what you are getting yourself into. Businesses like them, because they know their maximum exposure. I find that there a far too many clients, who think that they can estimate software development time, but they have little (no) true experience in software development. Some thoughts… Make sure that scope is clear with at least one bullet point for every deliverable that you know they are expecting and also clearly state that anything not listed in the deliverables will be cost plus later. This also helps them make sure that they have asked for everything they want / expect. I also typically have a list of non-deliverables, where we have discussed some feature and agreed that it would be included in a later version. Make sure that the contract is not a one way street, if you are expected to take the risk on a late delivery, you should also get the reward of getting it in early. If the client is asking you for a fixed price, then you should have the option to do the work where, when and how you like. This also ties the client into waterfall development, once you give them a version for test / review, there will be a stack of change requests coming in AT THEIR COST! I have found that some projects more than double in size with the change requests, that is why they have to be managed. Take care doing a few for free, or you will set up an expectation. Look out for clauses like: “Must be easy to use”, “Must be documented to our standard”, “Must be fully commented”, “Must follow our coding standard”, “Must be approved by our standards group”, “Must be approved by our DBA”, “Must interface with our un-documented system”, these are all open ended non-objective unclear requirements. My standard way of sizing a project is loosely based on (very loosely) function point counting, I count the number of database tables and simple screens and multiply this by a factor that I have worked out over the years of my realistic productivity, then I add on a margin for complex logic, complex screens, client liaison time, documentation, testing and general stuffing around that all projects have. Number one piece of advice, if you feel you don’t have enough information to make a meaningful estimation, then do your best (worse case) guess and at least double it. After I have done this, I look at the number and ask if it feels right or not, if not I adjust it. When you come to a final number, round it to a number that is not quite so analytical, the last two or three digits should always be zeros. If they say sign now or loose the contract, you say, that is fine, I will sign after I complete the analysis at $X per hour, if they don’t like that, you have to make the call on risk / reward ratio. Your analysis time should be charged or you are setting up an expectation that you will do work for free. Follow every conversation with an email to the client sponsor, CC to who you were talking to “As per my conversation with Fred, you need xyz and do not need abc. This will be provided with an increase in the project scope of X days and Y dollars. Please confirm your agreement by return email.”. Good luck Regards Greg Harris On Fri, Aug 2, 2013 at 7:51 PM, Richard Jones <[email protected]> wrote: > I have been asked by a potential client to work out time and cost estimate > before I have signed a contract to perform the work. They indicated they > didn't want a recruitment company. > > To me this seems a bit strange, as I have never experienced this before, I > have usually signed a contract got in and did the work, however, this is > different. They have indicated to me that they think this type of work will > take 3 months, however, they would like me to confirm/demonstrate time and > cost. > > Has anyone had this type of work?, any helpful comments/suggestions would > be grateful. > > >
