Data Conversion Projects

Hi, sKumaran,

I am not sure whether (1) you want to get a project or (2) you want help with a project.

Assuming (2), nothing is without cost. If you are offering your services free-of-charge there is still a cost, only you or your employer will be absorbing it.

The approach I have used over quite a few data conversion projects is as follows.
(a) ask ?What is the target system?? You need to get an understanding of what the new system is and what it does and the kind of data required for it to function correctly (or near correctly, as you may not have all the required data). Find out what data is mandatory and the system cannot function without.
(b) Find out what the existing (old) system is (what business functions it supports) and what data it contains.
(c) Perform function matching
This may be the biggest problem. The old system may offer some functions or sub-functions no longer supported in the new system. The business has to decide what to do with the data associated with the obsolete functions. If it is customer-related, are you getting rid of these customers?
(d) Perform a data-matching exercise. This needs both technical and business involvement. One needs to be careful here in that data with the same name in both old and new systems may not be the same data. E.g., ?sex? may be ?M? and ?F? in one system and ?1? and ?2? in the other. Or ?address? may be 3 lines of text in one system and ?house number?, ?street name?, ?street type?, ?city?, etc., in the other. Or ?location? may be a bin or shelf number in one system and a city/country in another.
(e) Perform data cleansing
This may involve running programs to ?clean up? wrong data, require manual input to fill in missing data, or require data substitution or reformatting (e.g., ?address?). The business has to decide what to do with records that cannot be "fixed", i.e., you cannot correct the data and you cannot get in touch with the customer (or whoever) to correct the data.

Hope this gives you enough to start with.